Howie Singer, Bill Rosenblatt - Key Changes - The Ten Times Technology Transformed The Music Industry-Oxford University Press (2023)
Howie Singer, Bill Rosenblatt - Key Changes - The Ten Times Technology Transformed The Music Industry-Oxford University Press (2023)
Howie Singer, Bill Rosenblatt - Key Changes - The Ten Times Technology Transformed The Music Industry-Oxford University Press (2023)
Key Changes
The Ten Times Technology Transformed the
Music Industry
HOW I E SI N G E R A N D B I L L R O SE N B L AT T
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
DOI: 10.1093/oso/9780197656891.001.0001
To Jessica, who kept telling me “You have one more book in you” until
I believed her.
–Bill Rosenblatt
Contents
Acknowledgments ix
We have each had our own journey through music and technology that
brought us to the point of writing this book, but the rest of the journey—to
the product that you hold in your hands or see on your screen—has required
a lot of research, help, and companionship along the way.
Many current and former music- industry executives were generous
with their time with us, including Elektra Records founder Jac Holzman;
Ralph Simon (cofounder of Zomba Records); former Warner Music Group
executives Paul Vidich, Jacob Key, George Lydecker, and Jordan Rost; former
Sony Music lawyer and former Napster General Counsel Gene Rhough;
former Columbia Records executive Paul Rappaport; former CBS Records
executive and NBC Friday Night Videos producer David Benjamin; former
Sony Music executive Rich Appel; former Sony executive Marc Finer;
Tucker McCrady of The Orchard (ex-WMG); Ted Cohen of TAG Strategic
(ex-EMI); MQA executive Mike Jbara (ex-WMG); former UMG execu-
tive and CEO of presssplay Mike Bebel; Matthias Röder of the Herbert von
Karajan Foundation; RIAA data guru Josh Friedlander; and Jonathan Taplin
(Director Emeritus, Annenberg Innovation Lab at USC; former Bob Dylan
and The Band manager; Oscar-nominated film producer). We also talked to
several music technologists and studio people, including former RIAA Chief
Technology Officer David Hughes, Bob Stuart of Meridian and MQA, Isabel
Garvey of Abbey Road Studios, Niclas Molinder of Session, and Bill Klinger
of the Association for Recorded Sound Collection.
We were privileged to be able to talk to several Grammy-winning recording
artists and producers, including Jerry Casale of Devo, Nick Rhodes of Duran
Duran, Imogen Heap, Arabian Prince of N.W.A., Peter Asher, Rob Fraboni,
and Albhy Galuten. From the world of radio, we spoke with Jim McKeon
(KWST Los Angeles, various Detroit stations, and various record labels),
John Platt (WXRT Chicago, WNEW New York, and WFUV New York), and
Sean Ross (Ross on Radio and Edison Research). Dr. Todd Boyd of USC was
x Acknowledgments
kind enough to share his expertise on the issues surrounding race and pop-
ular culture.
Finding historical information about (and usable photos of) con-
sumer audio electronics from the mid-twentieth to early twenty-first cen-
tury turns out to be surprisingly difficult; this subject tends to fall into the
cracks between museums and current information. Luckily there are sev-
eral online sources curated by passionate people who have filled in the
gaps, including WorldRadioHistory.com (a goldmine of searchable PDFs
of old audio magazines, consumer electronics retailer catalogs, and much
more), RadioShackCatalogs.com, JukeBoxHistory.info, WikiBoombox.
com, RadioMuseum.org, Reel- Reel.com, 8trackheaven.com, and
Tonbandmuseum.info. Flickr is also a rich source of photos of gadgets going
back decades. And let’s not forget the Internet Archive Wayback Machine,
which is one of the true miracles of our age.
We’d also like to thank various people and institutions for all kinds of
help, including but not limited to, and in no particular order: Karl Sluis for
the graphics; Lisa Shaftel of Shaftel s2do for help with photo clearances;
Dave Hermann for old Schwann catalogs of commercially available
recordings; Jedi Master Librarian Gary Price for help in finding research
sources on old consumer electronics; Max Poser of My Radio Berlin
for boombox history expertise; Brian Wallace, Curator of the Sarnoff
Collection; Regan Smith of the US Copyright Office for pointers to cop-
yright legislative history; Mike Lupica and Dante Sudilovsky of WPRB;
Louise Barder of Glossophilia.org; Viberate for its data analytics platform;
the NYU Bobst Library; the New York Society Library; and last but defi-
nitely not least, the amazing people and resources of the New York Public
Library for the Performing Arts.
Finally, much gratitude to our agent, Pamela Malpas of the Jennifer Lyons
Literary Agency, as well as to Norman Hirschy, Sean Decker, Egle Zigaite,
and everyone else at Oxford University Press.
If you had told me at the start of my career that I’d be discussing the future of
the music business with Neil Young or Robert Plant, I would have thought
you were crazy. I would never have reached that point if not for a PhD in
Operations Research and a successful career at Bell Labs, which depended
Acknowledgments xi
upon the support of Ben Operowsky, Alan Tucker, Jack Muckstadt, Janet Nici,
Dave Menist, John Sheehan, and many others. My transition to the music in-
dustry would have been impossible without Larry Miller, Tony Grewe, Sandy
Fraser, Tsvi Gal, Paul Vidich, and the entire a2b music team. My fifteen years
at Warner Music Group gave me a ringside seat to how the music business
handles disruption and provided the background needed to create this book.
There are far too many people across the industry who helped to educate
me to list here, but special thanks to Alex Zubillaga, George White, Michael
Nash, Paul Sinclair, Laird Popkin, Mike Elias, Bill Ardito, Mike Jbara, Ron
Wilcox, Craig Kallman, Edgar Bronfman, Jr, and Jac Holzman for their gen-
erosity and guidance. And thank goodness Bill Rosenblatt offered to join me
on this journey. Without him, this project would have taken far longer and
the result would have been far poorer.
Neither my career nor this book would have happened without the
sacrifices, support, and love of my family. I know my parents, Bernie and
Rhoda, would have shown off this volume with great pride and I expect that
my sister Helene will do the same. My sons, Matt and Doug, and their re-
spective spouses, Melissa and Camilla, were all cheerleaders for this project.
My granddaughters Riley and Eloise pitched in by demonstrating that even
the youngest music fans can now use voice assistants to play their favorite
tracks. The final and most heartfelt thank you to my wife Sandy: my love,
my best friend, and my number one fan. Here’s to many more adventures
together.
The path that I’ve taken into music technology might seem inevitable in
hindsight, given that both of my parents were musicians, I played a few
instruments myself, and I was a math nerd who liked to tinker with elec-
tronics. But the die was truly cast the summer after my senior year of high
school in Philadelphia, when I learned that the radio station I loved to listen
to, WPRB in Princeton, New Jersey, was the student-run station at the college
I was going to attend; thus began an association that led to 12 years in radio
and that continues to this day. It took a while to make my way from there into
the music industry proper, but I’m here, and I’m home. People I’d like to es-
pecially thank for my journey into the music industry over the years include,
certainly without limitation, Dan Harple, Chris Harrison, George Howard,
xii Acknowledgments
Adam Kidron, Larry Miller, Panos Panay, Bruce Rich, Rajan Samtani—and
Howie Singer.
Howie and I have known each other for roughly 20 years. After I wrote a
book on digital rights management in the early 2000s, his introduction led
me to a heady three-year stint at a digital-music startup, which in turn led
to a number of consulting projects with record labels, digital music services,
and music tech companies. Howie and I have worked on various projects
together, at Warner Music Group and elsewhere, most recently teaching
different sections of the Data Analysis in the Music Industry class at NYU.
When Howie approached me with his idea for a book during the lockdown
in the summer of 2020, I saw immediately that it held much more appeal
than the rather pedantic ideas for books on copyright and technology that
I was toying with at the time. My partnership with Howie during this en-
deavor has been the second-best I’ve ever had—the first-best being the one
I have with my wife Jessica Lustig, who has been a life partner in every sense
of the phrase as well as an unstinting cheerleader for me and for this book.
1
Introduction
Whole Lotta Shakin’ Goin’ On—Jerry Lee Lewis
On April 3, 2018, Spotify began trading its shares publicly for the first time.
Since its October 2008 launch, Spotify has added more than 200 million
paid subscribers and a half a billion total users, including those who use its
ad-supported tier. The largest technology companies in the world—Apple,
Amazon, and Google—all offer streaming music as well. And around the
world, other services—such as Deezer in Europe and Tencent in China—
provide access to an enormous catalog of songs to millions more music fans
at the touch of a finger to a smartphone.
This explosion in streaming has brought growth to a music business that
had basically been treading water since the massive declines of the Napster
era. Selling music downloads generated billions in revenues but never made
up for consistently declining Compact Disc (CD) sales. Apple dominated the
era of digital downloads with the iTunes store, and Steve Jobs often voiced his
opposition to the streaming business model.i
He famously said, “People have told us over and over and over again, they
don’t want to rent their music.” But the realities of the mass-consumer adop-
tion of streaming finally drove Apple to abandon its founder’s views—after
his passing in 2011—and ride the next wave. At this writing, Apple Music is
now estimated to have about 80 million music fans paying for access to its
streaming service.
People “renting” music by the millions required the melding of a range of
technologies. It represents a seismic shift in the industry, not just in terms
of the innovations that create a viable alternative to what was the dominant
music format of the day, but in the aftershocks that altered fans’ expectations,
the birth (and sometimes death) of the distribution channels to reach those
fans, the methods that artists use to create their music, the legal framework
around copyright, and the business models that ultimately fuel creators and
rights holders.
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0001
2 Key Changes
The music industry has existed on the fault lines of evolving technologies
for more than a century. In the late nineteenth century, two competing
formats for recorded music—the cylinder and the disc—altered the expe-
rience of listening to music and established the foundation for a new busi-
ness. Phonographs were followed by radio, the long-playing record and the
45-rpm single. Then came 8-track tapes, cassette tapes, and MTV. Analog
audio became zeros and ones, and the CD took over the music business and
drove it to new heights of financial success. But those same digits seeded
the conditions for the industry’s eventual decline. With the adoption of the
MP3 standard and advancements in computer technology, music could be
compressed, and the resulting audio files could be sold—or shared—over
ever-faster networks. And now downloading those files seems antiquated
as streaming, whether on Spotify, Amazon, or Apple, drives listening to un-
precedented levels. Fans have become more active participants in creating
and distributing music through videos on YouTube or TikTok.
Format changes in recorded music are the best lens through which to view
the evolution of the music industry; that’s because changes in formats have
pervasive effects on the rest of the industry. Every time the format changes,
the foundations of the music business shake. Nikki Sixx, cofounder and pri-
mary songwriter of the band Mötley Crüe, said, “Formats are going to change
because this is what the people want. It’s not what the labels want.”ii
For each of the formats in which recorded music has been distributed, this
book will review how technological advancements came together to enable
new experiences for fans. The first version of the new product was often not
quite good enough to drive mass-market adoption. In today’s terminology,
that first release represented a “Minimum Viable Product,” or MVP—that
is, a version of the product that enticed early adopters with a somewhat
compromised and limited feature set that was not quite good enough to at-
tract a mass audience. But those early users provided the feedback that helped
to point the way to the enhancements that broadened the product’s appeal.
Once the technology advanced to make the format attractive to a large au-
dience, the resulting mass-market adoption had wide-ranging impacts on
business models and ways of garnering revenues. As we’ll show in this book,
the changes reach far beyond consumer experiences. Artists change their
creative processes and musical styles to take advantage of the new formats.
New distribution strategies emerge, as do new ways for artists to get paid.
Rights administration, rights enforcement, and even copyright law itself may
shift to reflect the realities of the new formats.
Introduction 3
programmers selected. By 1932, just a few short years after the crash, US
record sales dropped over 90 percent to 6 million units. Music fans chose
convenience and price over quality in what would turn out to be a recurring
theme as music technology and formats evolved through time.
The broad reach of radio (Chapter 3) into a majority of US homes helped
to create the first multimedia stars. Bing Crosby leveraged his enormous pop-
ularity on the radio in the 1930s to become not only one of the largest sellers
of recorded music but also a huge box-office draw in Hollywood movies. The
improvement of microphone technology itself helped to make his crooning,
intimate style a success. AT&T’s telephone lines enabled the creation of the
national broadcast network, which, in turn, enabled advertising to thrive as
a mechanism to support the music business. And the radio quickly made its
way into the car as technology advances in electronics shrank the required
form factor sufficiently.
The recorded music industry standardized on 78 revolutions per minute
(rpm) in the 1920s. That limited the 12-inch diameter record—first made of
shellac, then vinyl—to less than five minutes of music per side. If you wanted
to hear the original Broadway cast recording of “Annie Get Your Gun,” you
had to listen to all 12 sides of the 6 records that were bound together as a
“record album.”
This was the case until after World War II: Peter Goldmark of Columbia
Records introduced the new 12-inch “long playing” (LP) vinyl record spin-
ning at 33 1/3 rpm in 1948. With LPs, listeners only needed to get up once to
flip their record to enjoy more than 40 minutes of music. That was enough
time to hear classical concertos in their entireties, or all the songs from a
Broadway musical. Even though only one disc was needed, the album termi-
nology stuck. The individual vinyl disc became known as the “album,” and
it drove decades of financial and artistic achievements for the music busi-
ness. In fact, even in today’s world of music streaming, the term album lives
on. Billboard charts that report sales and streams together measure “stream
equivalent albums” (SEA), and artists still make album releases into media
events. Not to mention that vinyl albums have now become the fastest-selling
physical format.
In 1949, RCA Victor introduced the 7-inch, 45-rpm format to compete
with the LP, but fans were not particularly interested in purchasing groups of
45s to represent a collection of related works. The LP and the 45 (Chapter 4)
quickly became the dominant formats for music collections and single songs,
respectively. As baby boomers became teenagers, their love of rock & roll
Introduction 5
drove their parents crazy. And their ability to buy 45s with their allowance
or babysitting money was the fuel that propelled that segment of the music
business, along with the pocket-size transistor radios that appeared in the
1950s, which those teenagers could afford as well. The LP also provided
bands and songwriters with a new creative canvas for the album as a unified
artistic concept.
Vinyl was virtually the only format for recorded music products for sev-
eral decades, although prerecorded reel-to-reel tapes had a small presence
in the market, particularly for classical music aficionados, starting in the
1950s when “high fidelity” (hi-fi) listening equipment first appeared. Figure
1.1 shows how different formats—physical and then digital—began to pro-
liferate after that (the RIAA began tracking annual revenues per format con-
sistently in 1973).
More user-friendly tape formats began to appear in the 1960s. The cas-
sette tape (Chapter 5) was introduced to the US market by Philips in 1964.
The sound quality of those first tape cassettes was poor, but they were far
easier to use than reel-to-reel. And the freedom they provided to listen to
30
25
TOTAL
REVENUE
20
CDs
15
Vinyl
10
Tapes
Streaming
5
Downloads
0
1973 1978 1983 1988 1993 1998 2003 2008 2013 2018
the music you wanted to hear on the go and to make copies for your friends
were major improvements. Shortly after that, 8-track tape players appeared,
starting in certain Ford automobile models. The format was derived from
tape cartridges used in radio stations. The 8-track was designed specifically
for use in cars, where the driver could simply “plug and play.”
At first, 8-tracks were better suited for use in cars than cassettes because
of their superior sound quality, durability, and ease of use while driving. But
in the early 1970s, consumer electronics and tape makers introduced home
cassette recorders and blank tapes with sound quality that began to approach
vinyl. By 1979, Sony had introduced the iconic Walkman, the first cassette
player that was not only as portable as those transistor radios but also offered
excellent sound through headphones, making cassettes desirable on both
the road and the sidewalk. These developments brought the dominance of
8-tracks as a “music on demand and on the go” format to a rapid end.
The cassette was also the first format that enabled easy recording of music
at home—another factor that contributed to the format’s success over 8-
tracks. Copying albums onto cassettes at home and sharing them with friends
became common, and the technology had evolved by the late 1970s to en-
able fans to make cassette copies of LPs that sounded almost as good as the
originals with moderately priced equipment. The industry pushed back hard
on what they viewed as damaging behavior. In 1980, a British band called
Bow Wow Wow released a single on a cassette tape with a blank “B-side” to
make it easier for fans to record music. Shortly thereafter, allegedly because
of the concerns over home taping, their record label EMI dropped the band.
Around the same time, the British Phonographic Industry (BPI), the trade
group for the UK recording industry in launched a major ad campaign with
the slogan “Home Taping is Killing Music.”
But that blank, inexpensive, recordable medium didn’t just let you make a
copy of Led Zeppelin IV for that sophomore across the hall in your dorm in
the 1970s. It also enabled DJs such as Grandmaster Flash and other pioneers
of hip-hop to make and distribute recordings of their live club performances.
It gave them the tool they needed to create a new genre of music and a busi-
ness model to support it, despite the major record labels’ initial lack of in-
terest in rapping instead of singing.
While hip-hop was burgeoning in the streets, the delivery of television sig-
nals via cable under those streets was setting the stage for the next major
game-changer for the music industry. Sixteen million US households were
receiving cable television and choosing among 28 channels in 1980. The
Introduction 7
Cable Act of 1984 loosened regulatory constraints and unleashed a rapid pe-
riod of growth. Soon a new wave of US teenagers were saying “I Want My
MTV.” And by the end of the decade, MTV was one of almost 80 channels
being broadcast over cable into more than 50 million US households.
Though not a physical format like the vinyl record or the cassette tape,
there is no doubt that the music video and the 1981 launch of MTV
(Chapter 6) drove significant changes in the music business. The creative pal-
ette expanded so much that when you see a song title such as “Beat It” or
“Material Girl” you think of the images from the videos as quickly as you re-
call the melody. For top musical artists of the 1980s such as Michael Jackson
and Madonna, it is impossible to separate their iconic dance moves and
visual imagery from the songs themselves.
The heyday of the cassette was relatively short lived compared to the
decades of vinyl popularity. That’s because Philips and Sony Electronics
introduced the CD format in 1982 (Chapter 7). The first CD players cost up
to $1,000 in the US but still sold several hundred thousand units in the first
couple of years in the market. Within a decade, volume drove player costs
down, and performance improvements made portable players a reality. CDs
zoomed past both vinyl records and cassettes in total volume and became the
choice of the mass market of music fans by the end of the 1980s. Unlike those
earlier formats, the reproduction of the sound rendered from the digital bits
placed on the CD did not degrade over repeated plays. And the perceived
quality advantage of the CD motivated millions of baby boomers to repur-
chase their music collections, leading to the greatest financial success the
music industry had ever experienced, as Figure 1.1 shows.
The seeds of the music industry’s massive decline that was to come at
the end of the century were planted with the ones and zeros embedded on
those CDs. Compact Discs usually held 700 megabytes of data. The typ-
ical personal computer of the mid-1980s contained 20 megabytes of hard
disk memory. The limitations of technology in terms of storage and pro-
cessing power meant that you listened to CDs on CD players and only on
CD players. But the fledgling personal computer industry was hard at work
leveraging Moore’s Law to drive massive improvements in computer perfor-
mance, and CD drives for personal computers also became available in the
mid-1980s.
As CDs were propelling the industry to unprecedented revenues,
audio experts at Bell Labs and the Fraunhofer Institute of Germany were
innovating with “Perceptual Audio Coding” algorithms. They could soon
8 Key Changes
discard 90 percent of the data on the CD and create a far smaller audio file
with an adequate audio experience that could be distributed over networks.
Those techniques were standardized by the Motion Picture Experts Group
(MPEG), and the resulting file was known as an “MP3” (Chapter 8). America
Online (AOL) was the leader in providing Internet access to millions of
homes via dialup modems, and it took more than 10 minutes to download
one of those compressed song files.
In June of 1999, Shawn Fanning released a beta version of Napster at
Northeastern University that ultimately triggered epic shockwaves across
the music industry. Those small MP3 files were ideally suited for the “peer-
to-peer” (user-to-user) exchange over the Internet, particularly over uni-
versity networks that were much faster than the typical residential dialup
connections at that time. But as broadband connectivity replaced dialup
modems in homes, the unfettered copying of the most popular songs became
an everyday occurrence across the globe.
The rapid declines in the music industry caused by Napster and its
successors created a sense of desperation in the industry as millions of people
simply stopped paying for music on CDs. Internet file-sharing went beyond
home cassette taping in three major ways: while blank tapes cost money to
buy and to send, digital files cost almost nothing; while tape copying had to
be done in real time (or close to real time), copying digital files via broadband
was instantaneous; and while cassette copies of LPs or CDs had degraded
sound quality (even if only slightly), copies of digital files sounded like the
originals.
The industry needed to find a way to attract millions of fans to pay for
music obtained online. Universal Music Group and Sony Music launched
an online music store called pressplay in late 2001 to compete with Napster,
while other major labels partnered with streaming audio technology pioneer
RealNetworks to launch MusicNet. The limited catalog and compromised
user experience of these services were not comparable to the simplicity and
price of the pirate services. In 2003, Steve Jobs and Apple created the first
viable competitor to free; the easy-to-use combination of the iPod and the
iTunes Music Store was soon selling billions of songs and albums—though
not enough to offset those cratering CD revenues.
By the beginning of the 2010s, advances in computing and networking
technologies enabled hundreds of millions of fans to pay $10 per month for
on-demand streaming from a huge song catalog on their mobile phones,
tablets, or PCs. These numbers continue to grow at a rapid rate around the
Introduction 9
world (Chapter 9), now that paying for access to music as a service rather
than purchasing music as a product has become the preeminent music ex-
perience. That growth led, starting around 2015, to the first increases in
overall recorded music revenues in almost 20 years. And that growth is ex-
pected to continue for a long time: In the 2022 edition of the Goldman
Sachs “Music In The Air” report, they predict that the worldwide revenues
for recorded music will exceed $50 billion, more than doubling the 2021
industry level, with over 85 percent of that total attributed to streaming
revenues.iii
Music videos evolved as the Internet provided the two-way communica-
tions channel that cable TV lacked. You can still watch the very first video
ever uploaded to YouTube in April 2005. It is titled “Me at the zoo” [sic]
and shows one of the three founders of the company, Jawed Karim, at the
San Diego Zoo. YouTube created a home for “User Generated Content”
(UGC) that enabled anyone to share videos with what would soon become
a massive, worldwide audience (Chapter 10). Creators generated revenues
by leveraging the advertising capabilities that are also part of YouTube’s
platform. If you search for the most viewed videos of all time on YouTube,
you find a list that is dominated by music videos featuring major artists,
with “Despacito” by Luis Fonsi and Daddy Yankee at no. 2, with more
than 8 billion views at this time of writing.1 User creativity has become
even more crucial to newer, short-form video platforms such as TikTok,
where individuals complement popular songs with their own singing or
dance moves.
The popularity of smart speakers, particularly the Amazon Echo and
Google Home, has made it easy for a whole new group of users to discover
and to enjoy music streaming services controlled via voice command.
Though we are still at the early stages of this technology, it is already clear
that it will have profound effects on how people discover and engage with
music and how the distribution platforms and artists optimize what they do
to leverage this new mode of interacting with music. Artificial Intelligence
(AI) techniques make voice response ever more accurate; they are being
deployed to improve music services’ playlists and personalization and to
find the most promising new artists. Creative AI tools that can generate
music appropriate for specific needs, such as synching to a video or new
songs in the style of famous artists, are already emerging as well. “Generative
AI” is clearly going to have major repercussions throughout the industry in
the coming years (Chapter 11).
10 Key Changes
What is the next set of technologies that will alter the structure of the
music industry? As we write this in 2022, we are reasonably certain that
the answer is “Web3.” As we explain in our Afterword, Web3 is a catch-all
term of distributed ledger or blockchain and related technologies; it should
usher in significant changes to the ways in which music is distributed to fans
and rights holders are paid. For example, nonfungible tokens (NFTs) have
captured the imaginations of fans and artists alike as attempts to meld the
advantages of digital and Internet technology with the scarcity and owner-
ship attributes of physical media. Although it’s too early to tell exactly how—
and to what extent—the industry will change in response to Web3, we offer
some observations.
As we said before, format changes are a powerful lens for viewing the de-
velopment of the music industry because they cause many follow-on effects.
We have identified six distinct categories of change in the industry, all of
which are related to evolutions in formats. We call these the 6C Framework.
They are shown in Figure 1.2.
We’ll be referring to these icons in each chapter, but generally the
categories of change are as follows:
Cutting-edge
Technology How innovations came together to create the new format
Copyright How laws are revised to address issues of the new format
Cash How artists and labels make money from the new format
Cutting-Edge Technology
The innovations that converged to enable new ways to enjoy and to dis-
tribute music. Several separate technological innovations typically come
together at a given time to enable the new format. For example, interac-
tive streaming relied on streaming audio technology but didn’t become
mainstream until other innovations emerged: mobile devices with pow-
erful processors and ample memory, mobile apps, and high-speed mobile
Internet access.
Creators
The changes artists and songwriters have made to their creative processes
and their creative outputs in response to the new format. New technologies
lead to new ways for artists to create music, ranging from the electric guitar to
the digital sample library. But new formats also lead to changes in styles and
genres of music. For example, the LP was originally created in the late 1940s
to hold classical concertos and Broadway musicals on single albums. But by
the mid-1960s, LP players had become affordable for young pop music fans,
and this created an opportunity for pop artists to consider the LP as more
than a collection of singles; they turned the format into a canvas for albums
that held together as cohesive artistic wholes.
Channels
The places—real and virtual—that labels and artists use to provide fans ac-
cess to their music. Our definition of channels has three components: distri-
bution, or the ways in which music is packaged and sent to the places where
fans access it; merchandising, or the ways in which the channel presents the
music to fans for discovery and selection; and consumption environment,
or the ways in which consumers can play the music distributed through the
channel. For example, labels manufacture vinyl records and distribute them
to retailers; retailers choose which items to stock and how to display and pro-
mote them; consumers buy records and take them home to play on turntables
and read the liner notes.
12 Key Changes
Consumers
Cash
Business models, royalty rates, and other industry practices that are in place
to gather money and to distribute it to artists and songwriters. Often these
models and practices are quite complex and take time to stabilize. For ex-
ample, radio stations pay royalties to songwriters and music publishers
through collecting societies such as ASCAP and BMI, which estimate airplay
of individual compositions through combinations of monitoring, sampling,
and self-reporting by stations; these practices evolved over the course of
decades and were improved by technological innovations such as automated
music recognition and large-scale databases.
Copyright
for recorded music was that copyright law does not include a performance
right in sound recordings; this meant that radio stations could play records
on the air without having to pay royalties to labels. As an example of the
former, the Digital Millennium Copyright Act (DMCA) of 1998 limited the
legal responsibilities that online technology platforms bore for the actions of
their users (such as uploading music files without permission). It provided
the legal framework for YouTube to launch in 2005 and ultimately to become
the world’s largest music streaming site. Various other digital content serv-
ices have flourished since 1998 by exploiting loopholes in the DMCA. This
has prompted many in the music industry to complain that the law should be
revised because it hasn’t achieved the balance of interests between tech and
content businesses that Congress originally intended in passing it.
Legal and regulatory frameworks beyond copyright, such as antitrust law
and broadcast regulations, have also contributed to structural changes in the
music industry. We touch on these as well when appropriate.
We use the 6C Framework to analyze each of these formats (phonographs,
LPs, streaming, etc.) in the ensuing chapters. For each one, we start with the
background story of the technologies that were on the cutting edge at the
time and the ways in which they coalesced to enable meaningful innovation
for the mass market of music fans. We follow up with discussions of how
the other parts of the business—the rest of the 6Cs—morphed in reaction
to this new reality. Sometimes these shifts occurred quickly, or at least they
appeared so with the hindsight of history. Other times the music industry
was reluctant to change course and dragged its feet for years before adapting.
It’s the same old song again and again as the established order of the music
industry gets “all shook up.” Advancements in technologies converge to en-
able new ways to distribute and to enjoy music. Often the first market entries
are simply “proofs of concept,” and feedback from the early adopters helps
to guide innovators to improve their products until they meet the needs of
a mass market. Those successful new formats not only give fans new music
experiences but also cause upheavals in the rest of the industry landscape. The
business models that remunerate creators for their work need to be redefined
to address the characteristics of the new format. New distribution outlets
emerge that are better suited to the new format, while old channels change or
disappear. Fans change how they interact with artists and their music based
on the new product configurations. Often there are demographic shifts that
coincide with these changes as younger people move first to join the next
wave, which can result in new genres of music coming to the fore. Artists
14 Key Changes
modify their creative processes to maximize the value they can now extract
and to connect with their fans more effectively. And although it often takes
years, the law eventually catches up with what the format has wrought.
Judge Sidney Thomas, in his ruling in an important file-sharing sharing
case before the Ninth US Circuit Court of Appeals (the Grokster case; see
Chapter 8) wrote:
Once that equilibrium is in place, all the parts of the industry have
adjusted to the “new normal.” The resulting stability enables the players to
thrive, but only for a limited time. Scientists, engineers, and entrepreneurs
are always working to come up with yet another innovation that can unsettle
the foundations of the music business. Learning how this cycle has recurred
through the history of the music business helps us to be better prepared for
the next go-around. As William Faulkner wrote, “The past is never dead. It’s
not even past.”
In the concluding chapter, we reexamine each of the 6Cs to consider the
effect it had across the entire recorded music timeline. We’ll also show how
each of the 6Cs affected another industry in analogous ways.
This book will appeal to anyone who is curious about an industry that touches
the lives of billions every day, from casual music fans to industry insiders as
well as those who want to understand the aftereffects of technology disrup-
tion in other businesses.
This book is designed for music industry professionals who seek to deepen
their understanding of the business, are considering the future of the in-
dustry, or may not be aware of its history or how it came to be the way it
is today. It’s also designed for journalists and industry analysts to help
them place current developments or consider the implications of strategic
Introduction 15
There are almost as many books about the music industry as there are cover
versions of “Yesterday.” Some concentrate on a single era, such as the birth of
rock & roll or hip-hop. Others dive deeply into how the industry reacted to a
specific disruptive technology; for example, we’ve read at least a half-dozen
books about Napster’s origins and how the industry dragged its feet while the
CD business collapsed. Other books take a longer historical view, but only
from a single perspective, such as how radio broadcasting or copyright law
has evolved over the decades. No one has examined this long line of format
changes with a consistent analytical framework that encompasses the tech-
nology, the artists, the fans, the channels, the cash, and the law.
This book originated as an idea that Howie had for a class in the Music
Business program at NYU, where he teaches the Data Analysis in the Music
Industry course for graduate students. He had a front row seat to the enor-
mous shifts in the music business after Napster through the dominance of
streaming by working for a start-up as well as for one of the largest music
companies in the world. But as he started to map out a course on technology
and its impact on music, he realized that the changes he had lived through
were not unique to the Internet era. Instead this was a theme that had re-
peated itself again and again across the history of recorded music.
Howie mentioned the idea to Bill, a longtime colleague, who teaches the
undergrad version of the Data Analysis in the Music Industry course and
has worked in book publishing as well as in music and radio. Bill had come
to a similar conclusion about the structural history of the music industry in
16 Key Changes
Around the turn of the century, people used novel technology to experience
music in a completely new way. They said the name of a particular song, then
inserted their earpieces, and were amazed when they heard the track they
had requested. Typically, they had to pay for this privilege, but some opted to
hear an audio ad either before or after the song instead and then enjoy their
selection for free.
That experience resembles what we have become accustomed to with
the proliferation of smart digital assistants controlling streaming music
applications. But the scenario just described was in place at the end of the
nineteenth century. It was enabled in the 1890s by the ground-breaking de-
velopment of the phonograph. If you lived in a major city, you could visit
an “Automatic Phonograph Parlour” equipped with “nickel- in-the-
slot”
machines. You would use a speaking tube to tell an attendant which of the
hundreds of available songs you wanted to hear. Once the recording was
mounted on the talking machine, you inserted a pair of rubber tubes in your
ears to hear it. Some songs were free if you listened to an ad, but most required
the listener to insert a token first. The novelty of hearing a recorded human
voice at these locations was a big draw initially, though it failed to become a
sustainable business.i This is the first of the many examples of commonalities
that turn up across the history of recorded music.
Thomas Edison, often referred to as America’s greatest inventor, saw the
phonograph’s most promising and immediate application as a dictation ma-
chine for businesses. The early devices were difficult to use and the quality
of the recordings often less than satisfactory. That limited their viability in
offices and made them more suitable as proto-jukeboxes for “parlours” with
trained staff. Edison, and other inventors, including Alexander Graham Bell
and Emile Berliner, competed to refine their new machines until they were
simple to operate and inexpensive enough to become mass-market products
purchased by millions. Each of their “talking machines” had different names,
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0002
18 Key Changes
Though Edison was ultimately successful in his search for a filament for the
electric light, his repeated failures would have deterred most people. Instead,
the “Wizard of Menlo Park” famously said, “I have not failed. I’ve just found
10,000 ways that won’t work.” This determined and systematic approach
proved unnecessary when it came to the phonograph. When searching for
a way to record telephone messages, Edison found that speech caused a dia-
phragm to vibrate which, when connected to a stylus, made indentations in
the wax-coated paper underneath. When that indented paper was run under
the stylus again, one could hear a faint reproduction of the original sound.
Relying on those findings, Edison sketched out a machine designed to
record and to play back the human voice on November 29, 1877. He turned
that drawing over to one of his workmen, John Kruesi, who returned a
day-and-a-half later with the first version of the machine. Instead of paper,
the device now used tinfoil as the recording medium. Edison set the stylus
against the tinfoil wrapped around the cylinder and yelled, “Mary had a
little lamb” into the mouthpiece containing a diaphragm while turning
the crank. When the stylus later ran over the indented foil, Edison’s voice
could be heard reciting the poem. Unlike the incandescent light, this pro-
totype device worked on the very first attempt. Edison said later that he
was never so taken aback in his life. He foresaw a wide range of uses for
the phonograph, including recording speeches and music as well as ena-
bling talking clocks and toys. However, he viewed the feature of making
recordings as essential to fulfill its primary mission as a business dictation
machine.
The next month, Edison filed for a US patent for his invention and took
the train from New Jersey into New York City to demonstrate his invention
to the staff at Scientific American. As the December 27, 1877 issue reported,
“Mr. Thomas A. Edison recently came into this office, placed a little machine
on our desk, turned a crank, and the machine inquired as to our health, asked
how we liked the phonograph, informed us that it was very well, and bid us a
cordial goodnight.” Edison’s ability to capture and replay speech was seen as
Phonograph 19
Figure 2.1 Thomas Edison with tin-foil phonograph, 1878. Edison National
Historic Site, National Park Service. Public Domain, Edison National Historic
Site, National Park Service. Reproduced with permission.
remarkable, and his new machine garnered attention in both the New York
and national press.ii
Seeking to capitalize on the publicity, Edison sold the rights to manufac-
ture and sell his phonograph to Gardiner Hubbard for $10,000. Hubbard,
20 Key Changes
would never amount to more than a novelty. The player piano, on the other
hand, was seen as a revolutionary musical instrument. To be successful, the
phonograph would need to compete with the already-existing player piano
as a home entertainment product.
Various versions of the player piano had been around for decades when
Aeolian introduced the Pianola in the United States in 1898. The operator
used foot pedals to generate suction that powered mechanisms that caused
the hammers to strike notes encoded as perforations on the piano roll.
This specific product was so successful that “Pianola” soon became the ge-
neric name for player pianos. Between 1919 and 1925, more player pianos
were manufactured than standard pianos. However, the $250 retail price—
equivalent to more than $8,000 today—put the player piano out of reach for
most households. The same was true even when models with modestly lower
prices were introduced. The ability to manufacture low-cost phonographs ul-
timately gave that device the edge in the home market.
In 1893, Edison introduced the Type M phonograph, which was specifi-
cally designed to serve the consumer market. The battery and electric motor
that served the needs of businesses were replaced with a clockwork mech-
anism that reduced three things: the manufacturing cost, the weight of the
machine, and the need for maintenance. However, the $150–$200 price
(equal to around $5000 today) meant that the phonograph, like the player
piano, still remained too expensive for the average household. The machine
had only two controls: a switch to activate the drive that turned the cylinder
and a screw to adjust the speed of rotation to deliver the correct pitch.
To lower the price further, Edison moved away from his commitment to
office dictation as the primary use case and made the recording mechanism
an option rather than including it on every model. The simpler playback-
only version could be sold for $100. Columbia, the manufacturer of the
Graphopone, went back and forth with Edison’s company, each trying to
outdo the other with ever-simpler designs and manufacturing improvements.
Home phonographs marched down the price curve to $10 for the Edison
“Gem” model in 1899, paving the way for millions of households to afford the
machines. However, mass-market acceptance required not only less expen-
sive players but also affordable recordings. The ability to make high-quality
copies of recordings at scale was essential.
At first, making multiple cylinders of the same performance was about as
low tech a process as one could imagine. The singer or musician performed
into the recording horns of a bank of several phonographs, each one cutting
Phonograph 23
the sound into its own cylinder. To make more copies, the artist simply re-
peated their performance. This approach was obviously neither scalable nor
consistent in terms of quality. A mechanical improvement soon helped to
provide somewhat more efficiency: a phonograph for making copies was
created, with two cylinders that operated in a coordinated manner. One cyl-
inder held the original to be copied and the other held a blank. As the grooves
were “read” on the original recording, a mechanism copied them to the blank
cylinder that was running in tandem. The recording material on the original
wore down every time it was played; it could be used to make 100 copies at
most. The duplicates created by this method were relatively poor quality that
only declined further as the original became more worn.
Edison and others experimented with a plating and molding process to
reproduce cylinders that delivered superior sound quality. The original cyl-
inder was plated with metal and then turned into a durable mold through a
series of additional plating steps. That mold was then filled with heated mate-
rial to create the duplicate recording, and when the material cooled and was
removed from the mold, it contained a copy of the indentations in the orig-
inal cylinder. A reliable molding process required a material that was suit-
ably malleable when melted yet got very hard after cooling. It took several
years to perfect that material, and by 1901, Edison was able to introduce Gold
Molded cylinders, and molding became the primary means of duplication.
The cylindrical shape of the mold created difficulties in extracting intact
copies. The alternative form factor of the disc proved to be far more ame-
nable to stamping out duplicate records.4 There was yet a third talking ma-
chine that, unlike the phonograph and the Graphophone, used a disc as
the recording medium. Its inventor, Emile Berliner, gave it another similar
sounding name: the Gramophone.5
Berliner emigrated from Germany to the United States in 1870. His first
major invention was a new microphone that could be used as a telephone
transmitter. It became embroiled in a lengthy patent fight with Edison,
who ultimately prevailed in court in 1901. Berliner’s first patent for his
Gramophone in 1887 used a cylinder but he quickly shifted his efforts to re-
cording music on a flat disc.
Berliner made the entertainment market his primary focus from the start.
He was remarkably prescient in understanding the key success factors for
players and records that would in turn form the foundations of the modern
record business. After presenting a demonstration of his Gramophone
at the Franklin Institute in Philadelphia in 1888, he proceeded to offer his
24 Key Changes
expansive view of the future. He foresaw that recorded discs would become
standardized so that music fans would be able to listen to them on a player. He
predicted that there would be a manufacturing process capable of producing
millions of copies from a single original disc. He also stated that “promi-
nent singers, speakers or performers may derive an income from royalties”
for sold recordings and that those performance would be “registered to pro-
tect against unauthorized publication.”vi Berliner’s ambitions to serve the
consumer market were reflected in his design choices. Gramophones were
playback-only machines unencumbered with the complexity and cost of re-
cording functionality. The discs were made of materials such as rubber or
metal, making them far more durable than wax cylinders for multiple plays.
Rather than carving the grooves in the up-and-down (known as “hill and
dale”) recording pattern used on cylinders, the stylus of Berliner’s recording
mechanism moved laterally, creating a spiral groove representing frequency
and amplitude that started at the exterior edge and moved toward the middle
of the disc. Because the turntable rotated at a constant speed, the stylus made
its way around the disc more quickly as it approached the center, unlike the
cylinder where the speed of the stylus remained constant throughout.6 That
increasing speed caused sound quality issues as the needle neared the center,
and louder sounds could cause the needle to jump out of the groove. A paper
label was placed at the center of the record so that the recording would end
before the needle ever got too close to the middle. That contrivance pro-
vided a perfect place to include information about the recording, including
the song and artist name along with the company issuing the disc. That last
piece of data ultimately gave rise to the record companies being referred to
as “labels.” Having that information placed on the center of every disc proved
far more user-friendly than the label on the end of the cylinder package.7 In
addition, discs provided louder sounds than cylinders, and their shape made
them cheaper and easier to manufacture. Consumers found them easier to
handle, and discs in sleeves took up far less storage space than a comparable
number of cylinders in their boxes.vii
Given that discs ultimately prevailed over cylinders as the recorded music
format of choice, there is a misperception that, in addition to these other
advantages, they provided superior sound as well. This was simply not the
case. Berliner’s first hand-cranked Gramophones sounded even worse than
the cylinder-based machines. The early Gramophones (which were offered
in Europe only) were considered, like Edison’s first versions, a mere curi-
osity rather than the foundation for a meaningful business. The duplication
Phonograph 25
process was similar to the plating and molding approach Edison had devel-
oped, although it used acid to etch the grooves into a zinc disc. Unfortunately,
that could erase some of the audio information, which, together with the hiss
of the steel stylus on the record, delivered an audio experience that was de-
cidedly inferior to cylinders. Eldridge Johnson, who was to become Berliner’s
partner in the United States, said that the Gramophone sounded like “a par-
tially educated parrot with a sore throat and a cold in the head.”viii
Berliner first approached Johnson, an American engineer and machinist,
in 1895 to craft a spring-driven motor for the Gramophone to compete more
effectively with the already motorized cylinder machines. Johnson met the
design challenge of providing enough torque when the needle was at the outer
edge of the record while maintaining a constant rotational speed. Johnson
undertook a series of follow-on projects to enhance the sound quality of the
machines. He designed an improved soundbox that held the diaphragm. He
perfected the duplication process by following Edison’s lead and using wax
for the original recording instead of zinc and then electroplating multiple
layers to make the mold. (He actually melted down some of Edison’s cylinders
to obtain the wax he needed for his experiments.) A shellac-based material
that included pulverized minerals such as slate and cotton fibers for strength
soon became the standard for the duplicate discs.8 These refinements yielded
discs that could compare more favorably with cylinders in quality; these
were commonly referred to as “shellac” records. But they had an unappealing
grayish color, so carbon black was added to the mix to provide the color that
became emblematic of records. By the turn of the twentieth century, both
Edison and Johnson were ready to mass produce recordings. That capability,
combined with the lower-cost players, led many middle-class families to in-
vest in this new form of home entertainment.ix
In 1901, Johnson and Berliner formalized their working relationship by
combining their interests to form the Victor Talking Machine Company
in Camden, New Jersey. That same year, they introduced 10-inch records,
followed two years later by ones measuring 12 inches. The play time of these
new formats exceeded three and four minutes, respectively, giving discs
a major competitive advantage over two-minute cylinders. This scenario
existed for several years as it took Edison time to find a new compound
that could close the capacity gap while still meeting the demands of mass
production.
The Victor company found another way to double the amount of music
pressed onto the records they were selling to the public. In 1904, it improved
26 Key Changes
and eventually forced it into bankruptcy, at which point Edison acquired the
rights to Lambert’s innovations.
In 1912, Edison introduced his Diamond Disc player and records. Unlike
his immediate success with the tinfoil prototype, he had to test over 2,300
materials exhaustively until he found one he could use for a stylus that could
pick up the recorded music patterns on the disc accurately. Up to that point,
Edison had adamantly stuck with cylinders alone, relying on what he saw as
their technical superiority, yet even he recognized that the disc was winning
the “format war.” Although Edison appeared to be surrendering to the disc,
he could not bring himself to concede completely. His Diamond Discs did
not use the same lateral recording method of Gramophone discs. This deci-
sion prevented any buyers of his new players from listening to any already-
released records. Furthermore, Edison continued to produce Blue Amberol
cylinders until his company shut down in 1929. However, from 1915 on, in a
clear indication of the relative importance of the formats in the market, the
recordings on those cylinders were simply duplicates of the Diamond Disc
versions.
The leading innovators within the phonograph industry competed ag-
gressively against each other to improve their products and processes. Yet
these efforts were confined to the field of “acoustic phonographs”—the cre-
ation and playback of recorded sounds without the benefit of electricity.
Although some talking machines had battery-powered motors, the drive to
reduce costs, size, and weight led all the manufacturers to deploy clockwork
mechanisms instead.
The improvements in function and style that we’ve described so far were
incremental; the most significant advancement to talking machines arrived
during the 1920s, when the “electrical phonograph” came onto the scene.
Once again, the innovations emerged from researchers working to improve
telephony. Volta Labs eventually became the research arm for the AT&T
Corporation. Initially housed within that company’s manufacturing arm,
Western Electric, it was reconfigured as a separate division in 1925 and be-
came known by its far more recognizable name: Bell Telephone Laboratories
(although Alexander Graham Bell’s involvement had waned). The top busi-
ness priority for AT&T was to improve long-distance telephone service,
and its Labs’ research agenda was directed toward achieving that goal. Their
scientists believed that electronic amplification was key to that improvement.
E. C. Wente invented a new condenser microphone at Western Electric in
1917. The rights to the three-element triode, a type of vacuum tube, had been
Phonograph 29
purchased by AT&T in 1917; it was used to develop a tube that could reliably
amplify a microphone signal. To run meaningful experiments, the engineers
needed to replicate their test conditions to evaluate performance. And that
meant using the exact same audio input each time.
Rather than trying to improve the acoustic phonographs already in the
market to meet their requirements, they decided to use the innovations at
their disposal to record sound electronically. The sound entering the micro-
phone was turned into a series of electrical currents. The vacuum tube ampli-
fier increased these currents to make them strong enough to create grooves
in the recording medium. The cutter operated within a magnetic field and
made grooves that represented the varying currents. The microphone’s elec-
trical diaphragm moved an infinitesimal amount compared to the previous
acoustic diaphragm, which allowed it to capture a much broader range of
frequencies with far less distortion. The engineers’ efforts to improve the
inputs for their long-distance tests had the unintended consequence of
improving the phonograph.xii
By 1924, the Western Electric team was ready to produce demonstra-
tion records for Edison, Columbia, and Victor, the Big Three phonograph
companies of the time. At first, none of them were interested in helping to
commercialize this new technology despite the promise of much improved
sound quality. The incumbents did not want to abandon their substan-
tial investments in the older generation of machines and the products and
profits they were already generating. However, radio’s growth, which had
already begun to erode the revenues of the record business, was gathering
momentum. The weakness in their business soon became evident, and the
Columbia and Victor companies decided to bring the Western Electric
engineers into their respective recording studios to perfect this new ap-
proach as a response to the radio threat. After months of experimentation,
the Western Electric solution became a commercial reality in May 1925,
when Bessie Smith stood in front of a microphone rather than an acoustic
horn to record her newest release, “Cake Walking Babies (From Home).”xiii
Though the first electrical recordings were issued in the spring of
1925, Victor and Columbia agreed to make no public announcements
about these improved discs until the end of the year, to allow them to sell
off their inventories of older acoustic recordings. In the fall of that year,
Victor announced its new “Orthophonic Victrola,”10 which was specifi-
cally designed to play the higher fidelity electrical recordings. A magnetic
pickup in the arm moved across the record and turned the amplitudes and
30 Key Changes
frequencies embedded in the grooves back into electrical current. The lower
frequencies that had been so difficult to capture acoustically were now re-
corded in the grooves of these new records. To play those bass tones back
accurately, a nine-foot acoustic horn was required; the engineers twisted the
horn into a spiral shell-like shape to reduce the total space required. That
allowed the manufacturers to use the same expensive cabinets sitting in their
inventory.
The New York Times front page on October 7, 1925 reported on the first
public test of the new Victrola with the headline: “New Music Machine
Thrills All Hearers.” John Philip Sousa, the composer and band leader, had
disliked the quality of his recordings on cylinders, despite their robust sales,
using the derogatory phrase “canned music.”11 He had a much more enthu-
siastic reaction to this new device, saying, “Gentleman [sic], that is a band.
That is the first time I have ever heard music with any soul produced by a me-
chanical talking machine.”xiv
Brunswick, one of the smaller manufacturers, was the first company to
go completely electric. Brunswick’s new player had a motorized turntable,
an amplifier, and a loudspeaker. By the end of the decade, most new record
players were fully electric. Edison doggedly tried to improve his acous-
tical machines; he was the last of the larger companies to make the change.
But even he conceded the superiority of the electrical machines when he
introduced his “Edisonic” models.
Early acoustic phonographs had rotation speeds that varied from 60 to
130 revolutions per minute (rpm). Cylinders typically turned at 150 rpm or
more. By 1910, the record players had settled into the range of 78 to 80 rpm.
Beginning in 1897, regulators or “governors” were added to the machines to
ensure that as the machine was cranked, a steady speed was maintained. The
shift to electrical motors provided the opportunity for a more precise solu-
tion, and 78.26 rpm became the standard because it worked well for most of
the existing records: that number of turns was the result of using a standard
3600-rpm motor to turn a 46-tooth gear.12 The new discs proved popular
and, before long, all records came to be referred to as “78s.”xv
Volume control in acoustic phonographs was decidedly low tech. The
Victrola’s horn pointed toward louvered doors in the cabinet that were
opened or closed to increase or decrease the volume. The owners of ta-
bletop machines with an exposed horn resorted to an even less precise tech-
nique: they kept a rolled-up piece of soft cloth nearby and shoved that into
the horn when they wanted to muffle the sound—giving birth to the phrase
Phonograph 31
“put a sock in it.”xvi The loudspeaker in electrical phonographs meant that the
product could have a small knob that adjusted the volume, providing a huge
improvement in the user experience.
Further innovations were yet to come for the phonograph record, and we’ll
describe the LP and the 45 in Chapter 4. However, despite the many advances
in electronics and digital technologies that were yet to come, one aspect of
early phonograph technology has persisted right through the current resur-
gence of vinyl records: the industry still manufactures records by creating
a “negative” mold from an original recording using metal plating and then
employing that mold to stamp out duplicate discs.xvii
Over roughly 50 years, inventors and their companies improved the quality
and manufacturability of the phonograph and recordings from primitive
prototypes to products with mass-market appeal. The products themselves
were necessary but far from sufficient to build a meaningful new business.
The entire infrastructure for the industry had to be created, including how
to decide what music to record, how record companies and artists should be
compensated, and what laws were needed to protect their financial interests.
Taking a cue from the most successful businesses of the day, run by moguls
such as Rockefeller and Carnegie, the phonograph companies opted to inte-
grate their operations vertically. They extended their business upstream to
identify the talent to record as well as downstream to manage the distribu-
tion and sales of their products to consumers.
The phonograph companies understood all too well the limitations of the
technologies they had invented. These limitations constrained the types of
music and sounds that could be recorded most accurately. The companies
and their executives were in the best position to select both the repertoire
and the artists to record it, and then to manage the recording sessions care-
fully to provide the best results possible.
The time limit for recording on any cylinder or disc was the most obvious
restriction imposed by technology. As materials improved, the duration of
recordings increased from two minutes to four, and one-sided discs were
replaced by those that carried music on both sides. Though the increases were
significant on a percentage basis, the absolute length remained fundamen-
tally incompatible with symphonies that often lasted an hour. Many popular
32 Key Changes
songs from the stage or vaudeville were much shorter, but even that mate-
rial often needed to be edited for length. Recognizing this, creators began
to watch the clock when crafting new works. For example, Igor Stravinsky
structured his “Serenade in A” to fit onto two two-sided discs, with each of
the four movements lasting less than three minutes.xviii The recording me-
dium had, in effect, circumscribed the music into a three-minute single, a
market reality that constrained songs for decades.13
For longer artistic works, the solution was to issue a collection that
consisted of many short selections on separate discs. Verdi’s Ernani, the
first opera recorded in its entirety in 1903, required forty discs. To hold all
these discs safely, “record albums” were sold, consisting of empty sleeves
and a hard cover mimicking the construct of a photo album. Eventually,
the record companies assembled and sold multiple 78-rpm discs in their
own albums as single products. The record albums offered in this way in-
cluded symphonies, Broadway shows, or simply collections of songs from
single artists. The “record album” has remained part of the industry’s lex-
icon ever since, even when technology enabled a single long-playing record,
a compact disc, or a group of digital files to embody a curated collection of
songs.xix
In addition to the time limits inherent to the recording media, there were
no recording studios and no editing capabilities whatsoever. Early acoustic
phonographs often performed unreliably, and a skilled operator, a “re-
cordist,” was required to make quality recordings. If the stylus was positioned
improperly or a musician played a wrong note or the vocalist sang an in-
correct lyric, the recording was thrown away and the process needed to be
started all over again.
Only those sounds within a few feet of the acoustic horn could be captured.
Quiet passages required performers to position themselves right at the open
end of the horn. If that soft section was followed by a loud high C, singers
had to move back quickly to prevent distortion on the recording. If multiple
performers were involved, they would all have to crowd around the horn.
Placement was essential, as the closer musicians would drown out players
standing further away. For ensembles, the vocalists would have to move out of
the way so the instrumentalists could get to the horn for their part of the per-
formance. In some more “choreographed” sessions, “pushers” maneuvered
performers seated on rolling boxes around the room to ensure that the right
people were in the right place at the right time. Broadway and vaudeville
performers such as Al Jolson, who were well-practiced in projecting their
Phonograph 33
voices to the back of the house, could do the same when singing into the re-
cording horn.xx
Given these hurdles, it is not surprising that many recordings opted for
simplicity and featured a single vocalist and accompanist. Creativity was
required to make the recording process work. If the accompanying instru-
ment was an upright piano, the back was removed and the entire piano was
boosted up higher so that the sound could be at the same level as the horn.
Even so, pianos still sounded somewhat coarse and tinny.
Other instruments were a poor acoustical “fit” for the recording process as
well. Very loud sounds could cause the cutting stylus to jump out of its groove,
so drums were often muffled or replaced by wooden blocks or cowbells. Soft
sounds such as those from stringed instruments were also problematic to
record. The lower-frequency double bass and cello were replaced by tubas
and trombones to make better recordings. John Stroh, an electrical engineer,
added a soundbox and horns to a violin to improve recordability, rather than
substituting another instrument. The resulting instrument, known as the
“Stroh violin,” was much louder than the standard instrument, and its main
horn could be directed right into the recording horn of the phonograph. (See
Figure 2.3) A smaller, secondary horn was also added to make it easier for
musicians to hear their own performance. Other sounds or instruments were
by their nature more conducive to the early recording process; hence more
records were made that featured the banjo, the xylophone, and even “artistic
whistling.”xxi
Dealing with the intricacies of acoustic recording had impacts beyond
any individual track. Louis Armstrong was in the studio with his band, the
Hot Five, to record a song called “Heebie Jeebies.” Armstrong was singing
when he dropped his lyrics sheets. Rather than stopping when the recording
was going well, he started “scatting”—singing with improvised wordless
syllables—and he was surprised that the take was not discarded. Armstrong
was not the first to scat, but “Heebie Jeebies” turned out to be a hit and
Armstrong was credited with popularizing what became an indispensable
technique of jazz.xxii
Though not as culturally significant or popular as jazz, klezmer music was
impacted as well. Originated by Eastern European Jews, klezmer absorbed
many of their themes and religious melodies. If you know the genre well or
are just familiar with the klezmer-influenced songs of Fiddler on the Roof,
you are aware of the prominent role of the clarinet in the music. However,
before 1900, the genre’s primary instrument was the tsimbel, a dulcimer
34 Key Changes
played with hammers. When attempts were made to record klezmer music
for the burgeoning immigrant population in the early twentieth century in
the United States, the tsimbel’s sound proved difficult to capture. To compen-
sate, the lead melodic lines were given to the clarinet, cementing its plaintive
notes as the trademark for the genre.xxiii
Given the constraints, the brass band turned out to be a good fit for early
recordings, albeit with fewer pieces than one might find in a live concert to
make the ensemble easier to locate in front of the horn. John Philip Sousa
had reservations about the fledgling technology.14 He thought that the
convenience of listening to music in the home would deter people from
coming to his concerts and from learning to play instruments themselves.
His concerns notwithstanding, Sousa agreed to have Columbia record “The
Stars and Stripes Forever,” which became a resounding success. The company
signed him to a recording contract, and the dozens of subsequent discs of his
performances helped cement his position as “The March King.”
At that time, Sousa was the exception, as most other recordings featured
people of little prominence. Many of the major performing artists of the
Phonograph 35
day steered clear of the phonograph because of the perception of the new
device as little more than a novelty offering poor quality music. The Victor
Company changed all that by recording music and artists with the cachet of
high culture and then spending lavishly to advertise the results.
The American phonograph companies licensed their technology and pro-
vided expertise to enable Europeans to set up businesses in their domestic
markets. Those European markets were viewed as the home for “good” music
respected for its artistic ambitions. In 1897, the Gramophone Company
was established in London. Fred Gaisberg, an American with experience at
several of the US phonograph companies, was sent over by Victor Talking
Machines to set up its recording studio in Britain. His role soon expanded
from recording engineer to one of the very first and most successful A&R
executives ever. His efforts to identify the “Artists & Repertoire” to record
took him to cultural centers across Europe. His powers of persuasion were
less than successful with higher profile artists until he broke through with the
Imperial Opera in St. Petersburg. A Russian Gramophone dealer, wanting
to cater to upscale customers, suggested that these new records be distin-
guished from the typical music that was already for sale. Instead of a black
label on the discs, these would feature a red one; and they were priced higher
to impress the public. This marketing tactic proved so successful that the
practice spread, and soon many more well-known performers were featured
on premium-priced “Red Seal” records, or their own distinctive label colors,
to help their music stand out from the Sousa marches, instrumental solos,
and duets that had constituted the bulk of the catalog.
Gaisberg’s next stop was Italy. His visit to the La Scala opera house in
Milan led him to sign the opera singer who became the foundational artist
of the recorded music business. Enrico Caruso, approaching 30 years old,
had a growing reputation for conveying emotion in his performances, but
he was far from world-renowned. He had not yet performed in the United
States. Unlike other “highbrow” artists, Caruso was not put off by the less-
than-stellar quality reputation of the phonograph. Sopranos were the most
popular operatic performers, but as with certain string instruments, their
vocal qualities were difficult to capture accurately. The frequency range and
tone of Caruso’s tenor, on the other hand, were particularly well-suited to the
early recording process. Furthermore, the singer was able to stand in front
of the acoustic horn and provide take after take of consistent performances,
even under the less-than-ideal conditions of recording in a Milan hotel
room.xxiv
36 Key Changes
across the industry. Edison understood that recordings could be more prof-
itable in the long run than selling phonographs themselves; nevertheless he
preferred to emphasize the technical quality of the recordings and his appa-
ratus and to downplay the role of the performer. Edison’s company offered
cylinders by volume and not by the subject of the recording, and he resisted
highlighting the performer’s name or likeness on the packaging for cylinders.
Instead, his boxes prominently featured a photograph of Edison himself.
Downplaying the role of the performer on the product was consistent with
Edison’s overall less-than-artist-friendly posture. Though the name of the se-
lection was printed on the edge of his cylinders, the artist’s name was omitted.
He complained that artists would jump to another company if offered even
a little more money, saying that for artists, “it is money, and money only
that counts.” Rather than get into a bidding war for popular musicians, he
chose to seek out and record less well-known and therefore less expensive
local talent, including choirs in Newark and West Orange, New Jersey. Once
again, Edison’s limited business acumen consigned his company to an infe-
rior market position.xxix
We would be remiss if we did not consider the issue of race in discussing
creators and the phonograph. The history of the music business is riddled
with examples where Black artists led the way in creating and performing
new kinds of music but were pushed into the background in favor of white
artists. We will examine the repeated use of this practice in Chapter 12. There
was a perception that the majority white audience would simply not accept
music from people of color. This deprived black artists of the fame and for-
tune that they deserved.
“Alexander’s Ragtime Band” provides a particularly instructive example.
Ragtime, an African American idiom, emerged in the 1890s and was fea-
tured prominently at the 1893 Chicago World’s Fair. Ragtime was typically
composed as syncopated piano music, but ragtime recordings often featured
banjos because of the difficulties of rendering the piano’s sounds accurately.
Ragtime’s popularity had faded somewhat when the white, Jewish Irving
Berlin wrote his song. Music fans loved the recording by the American duo
of Arthur Collins and Byron G. Harlan; it was the top-selling recording in the
United States for 10 weeks in 1911.xxx
It became a signature song for Al Jolson, the most famous entertainer
of that time, who often performed in blackface. His vocal style when re-
cording, and the idioms he used such as “the bestest band what am,” were in-
tended to convey the impression that the performer was African American.
38 Key Changes
The song’s phenomenal success led the press to label Berlin “The King of
Ragtime.” However, that title had already been bestowed upon Scott Joplin,
whose body of work was far more deserving of the appellation. Joplin, an
African American composer and pianist, wrote over one hundred “rags”
as well as a ballet and two operas. Yet Joplin’s songs were rarely recorded.
An even greater affront, at least to Joplin, was that he believed Berlin had
plagiarized the melody of “Alexander’s Ragtime Band” from him. Upon
hearing Berlin’s composition, Joplin said, “That’s my tune.” Berlin denied
the accusation, though there is no debate that he reaped far greater rewards
from ragtime than Joplin, who died penniless before turning fifty years
old.16 This, of course, was merely the first in a very long series of white artists
achieving greater fame and remuneration with Black music styles than the
originators did.
The shift from acoustic to electric phonographs in the 1920s brought fun-
damental change to the recording process and, as a consequence, to the most
popular artists and music. Instead of selecting the content to suit the vagaries
of the acoustic horn, the microphone captured subtleties of tone and dy-
namics and a broader range of frequencies that had only been available in
live performances. Performances like Caruso’s expansive tenor or Al Jolson’s
Broadway belting right into the horn were no longer necessary to create a
quality recording. Big bands and larger orchestras became viable options, as
it was no longer necessary to crowd a few individuals close to the horn. The
result was a major expansion of the repertoire of music that was feasible to
record. And the same microphone technology that enabled the crooning of
artists such as Bing Crosby to dominate the radio waves captured his more
subtle style for recordings.
For decades, phonographs and the recorded media were sold side-by-
side in the same stores. Spiller Records of Cardiff, Wales was the first such
outlet, offering phonographs along with cylinders and discs beginning in
1894. (It is still operating today, although its location has changed.)xxxi It
wasn’t until the 1930s, decades after the industry had standardized on the
78-rpm disc, that stores relying on record sales alone began to appear. In
the United States, the first such store was George’s Song Shop in Johnstown,
Pennsylvania.xxxii
Phonograph 39
By then consumers understood the value of records, but that was not
the case early on. Convincing people that this unfamiliar and perplexing
device was worth having in their home was a challenge; it required hands-
on demonstrations and education. To accomplish that, the phonograph
companies exerted their influence “downstream” to shape how their products
were explained, sold, and priced for consumers.
The companies preferred retailers that would maintain the manufacturer’s
suggested prices without discounting. They wanted stores where employees
had the technical expertise to adjust and to demonstrate the product for po-
tential purchasers and to repair the machines when required. Although the
companies each sold some products at wholesale, they established outlets
that were exclusive to their brand; this helped them avoid competition and
ensure that employees had the expertise required.
In larger cities, department stores could devote significant floorspace to
phonographs and records while offering multiple brands with discount prices
to encourage traffic. Even upscale stores like Bloomingdale’s sold records and
players. Furniture dealers in smaller cities and towns were preferred options
for distribution, as the wooden cabinets housing the talking machine proved
popular. Hardware stores placed phonographs next to sewing machines and
dry goods. For more rural markets, the Sears Roebuck and Montgomery
Ward catalogs provided distribution. The 1900 Sears Catalog featured “The
Wonderful Home Gramophone” for $5. By 1906, there were 25,000 stores
selling records and phonographs.
The Victor company took the most active role in guiding disparate retailers
to sell more professionally. Beginning in 1906, they published the Voice of the
Victor, a monthly trade magazine for dealers. By 1916, more than six thou-
sand dealers across the country received it. The magazine provided technical
information on the products, guidance on effective sales techniques, advice
on in-store displays, and advertising collateral. Victor even established a
“Window Display Service” as a separate line of business to provide specific
recommendations and goods such as pedestals to create more compelling
and standardized in-store displays.xxxiii
Perhaps the greatest support that Victor provided to its retailers was the
substantial investment it made in advertising. From 1901 to 1929, the com-
pany spent more than $50 million, or 8 percent of its revenues, on ads. Not
only did Victor use artists such as Caruso to sell its machines, it flipped the
script and used the machines to sell the experience of listening to realistic
reproductions. And no image captured that idea better than the fox terrier
40 Key Changes
cocking his head to listen to “His Master’s Voice” emanating from the horn
of a gramophone. The original painting by Francis Barraud showed his dog,
Nipper, listening to a cylinder phonograph. The recording feature of those
machines meant that Nipper could actually hear his master’s voice being
played back. However, Edison’s London affiliate passed on the opportu-
nity to purchase the rights to the image, and Barraud repainted it to con-
tain a gramophone. Despite the fact that the slogan made little sense for
Victor’s machines that could not create recordings, the concept captured the
public’s imagination and became emblematic of “supreme musical quality.”
Periodical Publishers’ Association rated that image as the world’s most fa-
mous trademark in the 1920s, and it remains one of the most recognizable
trademarks of all time.xxxiv
Though Edison eschewed the use of artists (or pets) in print advertising,
even he saw the need to promote the listening experience and the technical
superiority of his “Diamond Discs.”xxxv “Tone tests” were recitals where art-
ists would perform live at the same time their recordings were played. At the
end of the show, the house lights were turned off, and the performer would
leave the stage to demonstrate that the audience could not discern exactly
when the recording was the only source of music. Audiences flocked to these
recitals between 1915 and 1925.xxxvi Though the results impressed thousands
with the quality of the reproductions, the artists were selected, in part, for
their ability to sound like their own recordings thus biasing the tests in favor
of the machines.17
Advertising fueled interest in buying phonographs, but even lowering the
prices for some models to $5 represented a financial hurdle for many, as it
is equivalent to $160 today. And, of course, models geared toward upscale
consumers were even pricier. Extending consumer credit to purchase more
expensive goods “on time” helped to overcome this obstacle. The vast ma-
jority of phonograph purchases were made via installment plans, and, by
1920, those purchases constituted 5 percent of consumer debt in the United
States.18
Those credit arrangements established ongoing relationships between
retailers and purchasers. The need to acquire additional music to enjoy once
someone owned a record player added even more value to that relation-
ship. Victor promulgated that message in its Voice of the Victor magazine, in
an article titled “Are You Selling Music or Mechanisms?” in which it urged
its retailers to encourage customers to build a music library. Many stores
Phonograph 41
scheduled regular demonstrations to play new music they had received. They
even recognized the importance of the music catalog by providing advice re-
peated by music marketers through the years: “The oldest music is new to the
man [sic] who has never heard it.”xxxvii
Larger stores often are willing to accept slimmer margins on a particular
good in the hope of selling more items overall to customers. Retail stores,
then and now, thrive on foot traffic, and price promotions have always been
a surefire method to get people to come through their doors. Even in these
early days, merchants considered reducing prices for music products as a
“loss leader” as a tactic worth pursuing. As we will see, lowering the prices
for music goods to attract a crowd who then purchase other items, from
clothing to smartphones, is a recurring theme throughout the industry’s his-
tory. The phonograph companies provided encouragement and advice to the
retailers while protecting their parochial interests via strict contractual terms
that forbade discounting and required maintaining professional standards.
Breaching the terms would result in a write-up about the violation in the
trade press, or even a suspended or canceled agreement. This economic arm-
twisting did generate legal complications that will be discussed later in this
chapter.
To build greater demand for their products, the phonograph
companies encouraged their dealers to pursue one more market segment
as a “civic” duty: schools. Educational reformers such as John Dewey
were proponents of including music in schools, and what better way
to experience it than to hear it played on a phonograph. Victor formed
an Education Department to advocate for this and sent millions of
brochures and letters to schools and superintendents. It created “musical
memory contests” to encourage familiarity with certain compositions.
Participating in this early version of “Name That Tune” required
schools to have phonographs. The machines being purchased by schools
represented incremental revenues in the short term, with the prospect
that student familiarity with the device would lead them to lobby their
parents to buy one for their home. Or taking an even longer-term view,
students who enjoyed listening to music on phonographs would eventu-
ally become purchasers themselves once they were done with school.19
These programs were hugely successful with more than ten thousand
schools using Victor phonographs playing music into the ears of millions
of impressionable schoolchildren.
42 Key Changes
The typical record cost roughly seven cents to manufacture and sold for
25–50 cents. Even with payments to artists and retailers, most recordings
recovered their costs once sales reached 5,000 units. Victor’s decision to
distinguish records with higher quality music based on its Russian suc-
cess increased that retail gross margin further. Red Seal and other records
targeted toward upscale music fans sold for $2. Seventy million of those discs
were sold between 1903 and 1925, though they never accounted for more
than one-fifth of the total units sold. These premium-priced records featuring
respected artists not only yielded incremental profits, they also increased the
phonograph’s cultural relevance. That relevance, along with more afford-
able prices and the ability to reproduce the human voice as well as a variety
of instruments, provided the phonograph with a competitive edge over the
player piano. By 1923, the phonograph dominated the home entertainment
segment, outselling the player piano by a factor of five to one.xxxix
Revenue and profit growth in the industry impacted record companies’
financial relationships with artists. Performers demanded larger and larger
upfront payments for their contributions. Though some recordings earned
back those payments many times over, other projects remained in the red.
Rather than accepting the entire financial risk associated with a single large
flat-rate payment, the labels reconfigured the business arrangements with
the talent. Artists would receive royalty payments based on sales, but those
payments would not commence until an advance payment was recouped,
and given downstream royalties, these advances offered were more modest.
The success of Caruso’s early recordings prompted interest in signing
him from the other record companies. That gave him leverage to nego-
tiate an overall set of terms, including royalties, that proved quite favorable
compared to others. The Victor Company and Caruso concluded their first
contractual agreement in 1904. Caruso would receive a royalty payment of
50 cents per 12-inch record once the advance he was to be paid had been
recovered. These royalties would continue to be paid to his estate in the event
of the tenor’s passing. He was guaranteed a minimum of $10,000 in royalties
in each year that he made three or more recordings. In addition, Victor ac-
quired the exclusive right to record Caruso’s performances in exchange for
an annual payment of $2,000 each year of the term. If the company required
Caruso’s presence in New York for recording or any other business, he would
be reimbursed a further $2,000 for his travel. Finally, Caruso had approval
rights for recordings made under the agreement. If he was dissatisfied with
the quality of the result, the product would not be released. This contract
44 Key Changes
proved so lucrative for both parties that when it was revised in 1919, Caruso
was guaranteed an annual minimum payment of $100,000 and a royalty of
10% of the price of the record rather than a fixed rate per disc.xl
These arrangements proved to be incredibly rewarding for the most
successful artists. Over the course of 20 years, Caruso received more than
$5 million from his record company, or $100 million in today’s currency. In
1914 alone, he earned $220,000 or $6 million today. In many ways, this struc-
ture established the pattern for agreements between labels and recording
artists that remains in use today. One element in such agreements is an ad-
vance that is recouped via a royalty payment set as a percentage of the price.
Another is the artist demanding to renegotiate a higher royalty rate when
sales climb significantly. Despite substantial payments to performers, these
deals ensured that the labels retained the lion’s share of the financial rewards
as recompense for “fronting” the advance. As Enrico Caruso Jr. put it, his
father earned a “small fortune” while his record company received a large
one.xli The disparity between the outcomes for major labels and artists re-
mains a major source of contention in today’s music industry.
Government rationing to support the efforts in World War I reduced pro-
duction of both record players and discs. Pent-up demand and the postwar
economic recovery carried the music industry to new heights. Record sales
eclipsed sheet-music revenues for the first time ever in the early 1920s. Victor
sold 54 million records worldwide in 1921, surpassing their sales in any year
to that point in time. Industry sales of records hit $110 million that same year,
with overall revenues exceeding the equivalent of $5 billion in 2021 dollars.
The growth was, in part, fueled by shifts in patent protection. The pho-
nograph patents that had blocked companies from entering the burgeoning
business expired, leading to a tenfold expansion in the number of companies
manufacturing phonograph- related products. Victor lost a court case
involving its patents on flat disc technology; this allowed new, independent
labels to offer a broader music selection. The resulting records fueled de-
mand for players. Ironically, the legal loss ended up benefiting Victor as the
market leader in phonographs.
Record sales began to drop after 1921 with the launch of commercial radio.
The shift to electrical phonographs and recordings, using technologies from
radio, halted the declines by the middle of the decade. Worldwide record sales
began climbing again, reaching 200 million units by 1929, with half of that total
in the United States. Yet that recovery was short-lived as the Great Depression
brought the record industry to its knees. By 1932, US records sales dropped
Phonograph 45
with lower manufacturing costs targeted those with less disposable income.
For middle-class families, there were machines built into basic cabinets that
could fit in with the rest of their furnishings. Cabinets crafted from fine
woods in the Queen Anne or Chippendale style were offered to match the
luxury appointments in the homes of consumers with the deepest pockets.
Music and performers with demonstrated popularity in live theaters and
concerts were the first artists to be sought out for recordings, as exemplified
by John Philip Sousa’s Band with “The Stars and Stripes Forever.” The vaude-
ville circuit provided stars such as Billy Murray, who recorded for almost
every early record label with songs from George M. Cohan’s Broadway shows
such as “Yankee Doodle Boy” and “Give My Regards to Broadway.” Songs
with the broadest appeal were most desirable, but those that catered to sizable
audiences with specific musical preferences also proved their worth quickly.
A focus on operatic music by artists like Caruso appealed to fans who al-
ready had exposure to performances of this repertoire in larger cities or had
aspirations for more sophisticated “high art.” Europe was not only the home
for operas, it was also the birthplace of millions of immigrants arriving on the
shores of the United States in the early twentieth century. More than 8 mil-
lion people came to America between 1900 and 1910, representing more
than 10 percent of the population. By 1910, more than three-quarters of the
populations of New York, Boston, Chicago, and Detroit were either first-or
second-generation immigrants.
Though many immigrants often arrived in abject poverty, they made
rapid strides toward achieving some measure of economic security. That up-
ward mobility enabled them to spend money on luxuries like music. Many
yearned to hear songs that were sung in their native language or portrayed
the immigrant experience and their desire for assimilation. The larger record
labels were more focused on making records with mass appeal, which left
opportunities for smaller independents like Okeh Records to step in and
serve these more niche groups of music fans. Okeh made records in German,
Yiddish, and Polish; they featured performers who used Irish brogues
or Italian accents. Many of these songs were comedic and often relied on
stereotypes or racist generalizations.
Billy Murray was the top-selling recording artist of the first decade of
the century, and his recordings demonstrated this point.xlv He not only re-
corded George M. Cohan’s Broadway compositions, he also sang “Twas Only
an Irishman’s Dream” and “My Cousin Caruso,” appealing to ethnic music
consumers. He sang the latter in a thick Italian accent that may have been
48 Key Changes
Orleans to Chicago; once there, they attracted large audiences to their live
performances, though the music industry initially declined to record them.
Instead, jazz was repackaged to increase its perceived commercial appeal
to white audiences, pushing Black creators into the background yet again. In
1917, the Victor company brought the Original Dixieland Jazz Band (ODJB)
into its New York studios to make the very first jazz recording of a song called
“Livery Stable Blues.” ODJB was formed in New Orleans with a completely
white roster of musicians. To accommodate the constraints of time and
volume, recorded jazz eliminated the lengthy improvisations and muffled
the heavy percussion. Those changes along with more structured and ornate
orchestrations “toned down” the more raucous aspects of the new genre in
the hopes of increasing its acceptance by white audiences.
Paul Whiteman and his Orchestra epitomized that approach with their
“Symphonic Jazz.” Whiteman was the most popular band leader of his day,
and his syncopated arrangements sanded down the rougher elements of jazz
to the point where some detractors did not even consider it worthy of that
label. However, his “jazzy” renditions of songs like “Whispering,” played by
his highly talented but all-white ensembles, were top-sellers. As with rag-
time, this was yet another example of a historical pattern of the industry
“white-washing” Black artistry for white audiences while reaping the finan-
cial rewards.
By the 1920s, more jazz and blues recordings did feature African American
artists such as The Original Creole Orchestra and Bessie Smith. Their
releases, initially known as “race records,”20 were created expressly to appeal
to African American consumers and advertised exclusively in newspapers
owned by Blacks. African Americans’ response to the music was positive,
and white-owned record companies jumped on the opportunity and created
new units to focus specifically on this segment of consumers. For example,
General Phonograph’s Okeh label launched an entire series of “Original Race
Records.” The line included 50 recordings featuring Louis Armstrong and his
band, which cemented his position as a jazz virtuoso and a seminal figure
in American popular music. Over the ten years from 1922 to 1932, over
20,000 of these “race records” were produced, with sales reaching five million
units annually. This success paved the way for Black entrepreneurs to launch
record companies of their own. Harry Pace founded Black Swan,21 the first
label marketing to African Americans owned and operated by Blacks.xlvii
The company’s slogan was “The only genuine colored record. Others are only
50 Key Changes
passing for colored.” The label had some measure of success and ended up
being bought by Columbia in 1926.
Swing music evolved as a subgenre of jazz and became a fan favorite in the
1930s. Swing was more rhythmic, making it easier to dance to. The financially
strapped Depression-era consumers who were unable to afford phonographs
and records flocked to nightclubs and bars to enjoy the swing music they
were hearing on the radio. In addition to hiring live bands, many of those
venues purchased jukeboxes to provide musical entertainment, which drove
a significant share of record purchases. The big bands of swing were led by
musicians like Earl “Fatha” Hines, Duke Ellington, and Benny Goodman.
Swing’s popularity would span the Second World War until the introduction
of the vinyl record in 1948 made it possible to record the newer bebop style of
jazz, as we’ll see in Chapter 4.
Urban music consumers, both Black and white, might have preferred jazz,
but those in rural areas were bigger fans of “hillbilly music.” Ralph Peer, the
recording director of Okeh Records, heard the term in the South, and he used
it to refer to a genre that included bluegrass, gospel, and country and western
songs.22 Okeh issued the first country record that included lyrics and vocal
performances in 1923. It featured “Fiddlin” John Carson, who followed up
that release with a series of others that sold well. The big labels, seeing how
many consumers were interested in these country tunes, jumped on the op-
portunity that Okeh had identified. Victor followed suit with its own country
artist, Vernon Dalhart, whose rendition of “The Prisoner’s Song/Wreck of
the Old ’97” became the first million seller in the genre. By the 1930s, country
fans were buying millions of copies of records by artists like The Carter
Family and Jimmie Rodgers, and “hillbilly” music constituted a quarter of
all sales.
Consumers were not only interested in the music itself: they were
fascinated by the people who made that music, and the most popular artists
became worldwide celebrities. Once again, Caruso paved the way. He was
the first recording star who had to deal with the downside of fame that offset
the enormous financial rewards. A century before Facebook and Twitter,
the details of Caruso’s life—both mundane and salacious—became tabloid
fodder. News stories about Caruso featured him shaving his mustache, the
theft of his wife’s jewelry, and blackmail threats by an Italian organized crime
group. The most sensational Caruso story was undoubtedly his arrest for al-
legedly molesting a woman in the monkey house at the Central Park Zoo.xlviii
His trial was featured on the front pages of the New York newspapers for
Phonograph 51
The US copyright law in effect when the phonograph was invented was
enacted in 1831. That statute explicitly covered musical compositions; it pro-
vided composers and music publishers exclusive rights to the reproduction
of sheet music. However, that law did not contemplate copying a music per-
formance onto a cylinder or disc, nor did it cover encoding notes played on a
piano onto a perforated paper roll. However, soon after the turn of the twen-
tieth century, it was quite clear that “piracy”—the unauthorized reproduc-
tion of records—was an issue for the industry.xlix
The fact that the law did not expressly address these new music formats
did not deter rightsholders from going to court to address that piracy and
to seek compensation. Stern & Company, a Tin Pan Alley music publisher,
saw recordings as a way to “plug” songs to potential buyers of sheet music
and founded the Universal Phonograph Company to manufacture cylinders
for Edison’s machines. It soon realized that the sale of those cylinders was
itself a viable business and not simply a vehicle for promotion. George
“Rosey” Rosenberg recorded and released cylinders with his versions of
songs written by Stephen Porter, the composer and vaudeville performer.
Stern had purchased those compositions, but the publisher sued because
it viewed Rosey’s recordings as infringing on its copyrights and damaging
to its financial interests. In 1901, the District of Columbia federal appellate
court dismissed the case of Stern v. Rosey summarily. Though Justice Seth
Shepard agreed that the issue itself was novel because of the newness of the
phonograph, he ruled that cylinders were not a substitute for sheet music.
Furthermore, he ruled that the cylinders were not even copies as defined
under the law. As Justice Shepard wrote, “We cannot regard the reproduction,
through the agency of a phonograph, of the sounds of music instruments
playing the music composed and published by the appellants, as the copy or
publication within the meaning of the act.”l
52 Key Changes
This legal ruling was simply following the precedent established for piano
rolls some twenty years earlier. William Kennedy and the Automatic Music
Paper Company had sued John McTammany, who proclaimed himself in-
ventor of the player piano, for copyright infringement. McTammany had
made an unauthorized piano roll of a song that Kennedy had licensed ex-
clusively to the Company. The court found for McTammany and ruled that
no infringement had occurred because, under the law, a piano roll was not a
copy; it was simply a part of a machine. It could be rendered as a recognizable
song by the mechanism, but, unlike sheet music, it could not be deciphered
by a person. In 1892, the US Supreme Court declined to hear the appeal of
this case.li
Many music publishers accepted the legal situation where there were no
royalties for piano rolls and recordings because of their promotional value in
increased sales of sheet music. The Automatic Music Paper Company merged
with Mechanical Orguinette Company and was renamed the Aeolian Organ
and Music Company. Aeolian became the largest producer of player pianos
and rolls. Though it seemed contrary to their financial interests, the com-
pany favored a legal regime that required royalty payments to publishers for
those piano rolls. Aeolian’s plan was to corner the market by licensing the
most popular songs from publishers on an exclusive basis and then to sell the
piano rolls to their customers. Even without any retroactive payments and a
35-year term for the exclusivity, the prospect of royalties for piano rolls had
80 publishers signing on the dotted line.
However, the deal was dependent on having a copyright system that
would prevent others from making their own piano rolls of these popular
songs and undermining the value of exclusivity. Behind the scenes, Aeolian
maneuvered to initiate a case that could establish that piano rolls were in-
deed copies deserving of copyright protection. In May 1902, White-Smith
Music Publishing filed suit against the Apollo Company for infringing their
copyrights by making unauthorized piano rolls. It argued that protecting
creators against any form of piracy, including piano rolls, was the very pur-
pose of copyright law.
Aeolian spent heavily to help White-Smith demonstrate that Stern v. Rosey
had been decided wrongly and that piano rolls should be treated as copies.
Aeolian’s lawyers gathered evidence to support two arguments. First, they
tried to show that a piano roll was simply an alternative type of music nota-
tion that could indeed be read by a pianist with practice. However, when they
tried to demonstrate this “fact,” the results were less than successful. Second,
Phonograph 53
they claimed that piano rolls were a commercial substitute for sheet music.
This argument was undercut by the existence of numerous letters from
publishers to Victor, Edison, and Columbia asking them to promote their
songs by producing piano rolls for promotion.lii
The focus on the publishers’ commercial practices opened the door for the
exclusive contracts crafted by Aeolian to be considered as part of the case.
Apollo argued that long-term exclusive rights to the best music would en-
able Aeolian to put their competitors out of business. This proved to be a po-
tent argument, given the climate engendered by President Teddy Roosevelt’s
public posture as a “trust-buster.”
Before White-Smith v. Apollo was decided, the Librarian of Congress,
Herbert Putnam, convened the first of three planned conferences to discuss
revisions to the Copyright Law. President Roosevelt was far more supportive
of enacting a law to cover these new, modern methods of reproduction than
he was of anticompetitive practices. His message to Congress that year stated,
“Our copyright laws are urgently in need of revision.” Putnam invited or-
ganizations representing authors, artists, photographers, and others in the
creative community, but he did not include any representatives of the phono-
graph or piano roll manufacturers. The only group from the music industry
that received an invitation was the Music Publishers Association. Despite the
concerns of popular composers like Victor Herbert, who felt that royalties
should be paid for records and piano rolls made without his permission, the
Association’s representatives at the conference were notably reserved on the
issue of such payments. They were under explicit orders to steer clear of any
issue that might compromise the Aeolian contracts while the case was still
pending.
That strategy proved less than fruitful. Despite the time and resources
invested by Aeolian, the federal district court ruled for Apollo in June of
1905. Once again, the judge found that copyright law dictates that a copy of
a musical composition must be a “tangible object that appeals to the sense
of sight.” Having failed in the courts, the path to protecting piano rolls and
phonograph records as copyrighted works would indeed have to go through
Congress.
When the Library of Congress held the remaining sessions on revisions
to the copyright law beginning in November 1905, Nathan Burkan, a
young attorney who would become one of the most influential advocates
for songwriters as a founder of ASCAP, led the preparations for the Music
Publishers Association. The publishers now argued for a “mechanical
54 Key Changes
reproduction” right that would cover both records and piano rolls going for-
ward. The attendees reached consensus on this and a wide variety of other
issues including lengthening the copyright term and strengthening the civil
penalties for infringement. This commonality of views led the participants
to believe that the revised Copyright Act would sail through to legislative
approval.liii
Yet the first congressional hearing on the draft bill in May 1906 quickly
disabused them of that notion. The publishers were so confident that they
had chosen celebrities rather than experts as their witnesses. The com-
mittee members handled John Phillip Sousa and Victor Herbert with
kid gloves, but the two composers failed to lay out a compelling case for
the benefits of the mechanical right. They did not explain how the com-
pensation would encourage creators and thus benefit the public with
more music.
The manufacturers of players and recordings, on the other hand, who had
been surprised by the fast-tracking of the bill, presented a much more com-
pelling case at the hearing. They built upon many of the same arguments that
had proved persuasive in the Apollo ruling just a few months earlier and had
been affirmed in appellate court just before the hearing. The manufacturers
asserted that the courts’ rulings that records and cylinders were not copies
because they were not readable by people meant that they should not be
considered “writings.” Therefore, Congress did not even have the authority
to craft a law providing copyright protection for these new products because
the Constitution defined copyright as providing authors exclusive rights to
their respective writings.
The exclusive contracts between Aeolian and the music publishers
proved to be even more damning when scrutinized by Congress. The me-
chanical right was portrayed as an essential element of a monopolistic
plan that greedily took money from the public to line the pockets of a few
music publishers. The hearings went so poorly that striking the mechan-
ical right from the proposed bill completely and proceeding without it was
considered as a possible strategy. Burkan kept the literary groups—who
might have chosen to leave the music industry to their own devices—on
board by extending the mechanical right beyond musical compositions to
their artistic outputs. The next time he showed up for hearings on a new cop-
yright bill, in December 1906, he was joined by Sousa and Herbert along with
authors William Dean Howells, known as the “Dean of American Letters,”
and Samuel Clemens, better known as Mark Twain.
Phonograph 55
This time Burkan was well prepared both for the political realities of
Washington and the intricacies of copyright law. Herbert and Sousa joined
him for meetings to schmooze with DC power brokers, including a cour-
tesy call on President Roosevelt himself. At the hearing, Burkan presented
a carefully researched and masterful rebuttal to the claim that Congress had
no authority to create a mechanical right. He argued that “writing” could be
applied to a “perforated roll or a phonographic disc” as they both made the
author’s ideas perceptible to the mind as opposed to the eye. Burkan argued
that the exclusive right to public performances of written works—where the
public would perceive the works with ears, not eyes—was a precedential ex-
ample of the law protecting creators from unauthorized uses.
The members of the committee bought into Burkan’s logic, but there
was still the matter of those Aeolian contracts and their poor optics with
Congress. The phonograph and player piano industries had become political
forces responsible for thousands of jobs for workers and for millions in eco-
nomic activity. They used that increased clout to push Congress to remove
the mechanical right from the bill.
They got their wish when, in January 1907, the drafts from the respective
Patent Committees emerged and the House omitted the mechanical right en-
tirely. The White-Smith case had been appealed to the US Supreme Court,
and Congress was willing to wait for that decision to provide guidance. The
Senate committee, albeit by just a one-vote majority, included a version of the
mechanical right that met with the approval of Burkan and the publishers.
This fundamental difference between the two versions of the bill doomed any
prospects for passage before the end of the session.
The impasse in Congress moved the Supreme Court case back to the
forefront. Copyright was on the judicial agenda for the session, which
commenced the first Monday of October in 1907; oral arguments were
heard at the start of the new year. Burkan advanced the same arguments de-
fining “copy” to include both piano rolls and phonograph records that the
Congressional committees had found compelling. But the justices were un-
convinced. They were willing to maintain the precedents, given that highly
profitable businesses had thrived under those rulings. The Court’s unani-
mous decision issued on February 28, 1908 stated unequivocally that a copy
was defined as “a written or printed record in intelligible notation.” Justice
Oliver Wendell Holmes was unwilling to dissent given the precedents, but
he was sympathetic to the composers’ plight. In his concurring opinion, he
wrote, “On principle, anything that mechanically reproduces that collocation
56 Key Changes
of sounds ought to be held a copy, or, if the statute is too narrow, ought to be
made so by a further act.” The Aeolian contracts included a clause that the
deal would take effect only if there were a favorable decision creating an en-
forceable mechanical right. The Supreme Court ruling put an end to Aeolian’s
hopes of creating an exclusive arrangement with the publishers.liv
The prospects for getting the mechanical right enacted in Congress
were growing dimmer as well. The Senate Patent Committee was now
led by Senator Reed Smoot of Utah, who had previously sided with the
manufacturers. He reintroduced a bill stating that there would be no lia-
bility for copyright infringements from mechanical copies. On the House
side, Patent Committee Chair Frank Currier of New Hampshire and other
members were concerned about the potential for a monopoly in music
rights. The terms of the Aeolian contracts and their exclusivity clauses pro-
vided valid cause for such fears.
Currier introduced a new House bill that adhered to the same hard line
on mechanical rights that the Senate had put forward. The message to the
publishers was clear: you can be remunerated for the copies of your songs
used on piano rolls and records, but only if you give up the ability to control
those uses explicitly. This “compulsory license” was a new concept for cop-
yright law and, in the final joint Congressional committee hearing, Burkan
argued strenuously on behalf of the publishers that the entire concept was
unconstitutional. But without it, the manufacturers would challenge any me-
chanical royalty whatsoever. The publishers and the manufacturers agreed
that a compromise needed to be struck.
The parties agreed to stay in Washington to work out the details and, of
course, the rate for those mechanical copies was the core issue. A two-cent
rate had been suggested at a previous hearing. The number rounded up the
one-and-a-half-cent charge for each sheet-music copy that had proved small
enough to avoid excessive economic harm to consumers. Two cents became
part of the draft agreement among the parties even though many publishers
felt the payment was insufficient. On the other hand, some manufacturers
thought that even this amount was exorbitant. Though the parties did not
formally agree before leaving town, the parameters of the final result had be-
come clear.
A subcommittee was empowered to finalize the legislation when the
lame-duck session of Congress convened in December 1908; it settled the
remaining open items in the bill. The final bill was passed by both houses
of Congress during the closing hours of the lame-duck session. It went to
Phonograph 57
the White House on March 3, 1909, and in one of his final acts as President,
Roosevelt signed it into law. The final bill extended the term for copyright
protection to 56 years, consisting of initial and renewal terms of 28 years
each. The mechanical royalty was indeed fixed at two cents. Once a com-
poser or publisher allowed an initial mechanical reproduction of their work,
any other party could avail themselves of the compulsory mechanical license,
make copies, and pay that fixed rate. This gave rise to “cover versions”: once
there is a recording of a song, anyone else can record and sell their own ver-
sion. This practice has given us both great recordings as well as some that
proved lucrative but poor imitations of the original.23
The political machinations and maneuvering that led to the compulsory
mechanical license gave credence to the cliché that enacting laws was like
making sausage. That’s why it took nearly 70 years for another major revision
of US copyright law. Over that time, the mechanical rate increased to the cur-
rent level of 9.1 cents (or 1.75 cents per minute of playing time for the song).
Though it created a steady, reliable flow of funds to publishers, this statutory
rate consigned them to a small share of the revenues for the millions upon
millions of copies that would be sold over the next century.
The Aeolian agreements showed that publishers were willing to en-
gage in anticompetitive practices to increase their bottom lines. Those
tactics provoked the phonograph manufacturers to push hard for a com-
pulsory license to ensure that there would be no exclusivity whatsoever on
compositions. However, those same companies were not averse to exploiting
their own market power when it served their own financial interests.
As discussed earlier, the phonograph companies tried to control their dis-
tribution channels through both direct and indirect means. They provided
training and information to retailers to assist their educational efforts with
potential buyers. They were particularly concerned that department stores
and other large retailers would discount phonographs and then make up for
the lost margins selling other goods. Victor’s price schedule was included in
its contracts with distributors, and violating the terms could result in suspen-
sion or cancellation.
Although many retailers bought into compulsory pricing, others
objected. The Fair Store sued Victor over their efforts to set retail prices for
phonographs. In Victor Talking Machine Company v. The Fair in 1903, the
US Court of Appeals for the Seventh Circuit ruled in favor of the Victor
Company, setting a legal precedent for price regulation. When later rulings
restricted setting the actual sales prices, Victor and other manufacturers
58 Key Changes
Conclusion
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0003
60 Key Changes
The disruptive factor that brought radio fully into the sphere of the re-
corded music industry was not a station, network, program, or regulation: it
was a product that weighed about a pound and fit handily in a pocket or purse.
The Regency TR-1 from Regency Electronics and Texas Instruments, the first
transistor radio, appeared in November 1954. Pocket-size (or almost pocket-
size) vacuum tube radios existed, but they suffered from long warm-up
times, short battery life, heat from tubes, and poor sound quality. Transistor
radios were smaller and lighter, they turned on instantly, and their batteries
lasted for weeks.
The TR-1 sold for $50 ($480 in today’s dollars) and received poor reviews
for its performance. But it was quickly eclipsed by other models from
Raytheon, Zenith, and several companies from the newly revitalized Japan,
including Sony. These sold—for prices around $15–$20 (about $150–$200
today)—in the millions by the end of the 1950s to the growing number of
baby boomers who were becoming teenagers then.
Transistor radios were among the most important reasons why rock & roll
became big business. For the first time, teenagers didn’t have to listen to their
music on the big console set in the living room that families of the 1940s
Radio 61
Figure 3.1 The Motorola Eight-Sixty car radio from the late 1930s featured
pushbutton station presets. © Motorola, Inc., Legacy Archives Collection.
Reproduced with permission.
Station-switching technology took its next leap forward in the early 1980s
with the invention of phase-locked loop (PLL) electronic tuning. PLL tuning
enabled a frequency to be set fully electronically, without requiring a me-
chanical action to turn a tuning capacitor. With PLL, a radio could have arbi-
trarily many presets and could “seek” and “scan” to other stations.
Streaming audio technology brought radio into the digital age in the mid-
1990s, with technology introduced by the Seattle-based startup Progressive
Networks (later RealNetworks) and a few others. By then, the technology for
storing music in digital files was well understood (see Chapter 8); streaming
audio involved breaking up audio files into small packets, transmitting a
packet at a time, and reassembling the packets at the receiving end into a con-
tinuous stream of audio. Versions of codecs (digital audio encoding schemes)
such as MP3 and MP4-AAC suitable for streaming were invented. Bitrates
Radio 63
for streaming audio had to be chosen to suit the network bandwidth avail-
able. By 1995, Progressive Networks’ RealAudio was capable of streaming
stereo music with adequate sound quality over the 28.8 kbps dialup Internet
connections that were common at the time. Microsoft and America Online
(AOL) introduced or acquired their own streaming audio technologies in the
ensuing years.
The first several uses of streaming audio for music transmission fell into
two categories: simulcasts of AM or FM stations and “pure play” Internet
radio. Early examples of the latter included NetRadio (125 channels of
mostly rock music), HardRadio (over 50 hard rock channels), Sonicwave,
and TheDJ, later renamed Spinner and sold to AOL. Live365 enabled users
to post their own “stations,” with fixed music playlists, for anyone to listen to.
But networks such as the Internet weren’t useful just for sending
preselected audio programs to remote users. Streaming audio services could
send different streams to each user. Entrepreneurs also figured out quickly
that the networks could also be used for listener input that would cause
changes to the programming. They began to experiment with various types
of input schemes. These fell into two categories: those that enabled users to
pick the specific music they wanted to hear, and those that didn’t. The former
became known as interactive or on-demand streaming; we’ll talk about those
services in Chapter 9.
Simpler feedback mechanisms came first. Imagine Radio, which launched
in 1998, enabled users to start stations based on music genres and artists. It
also enabled users to rate both artists and songs on a 0-to-10 scale, which
would determine how often they heard the particular music. LAUNCHCast,
shown in Figure 3.2, started the following year. LAUNCHCast also enabled
users to rate tracks; in addition, it let users “subscribe” to other users so
that the other users’ ratings influenced the subscribing users’ playlists; and
it introduced a “skip” feature. These were all precursors to features of later
services like Pandora and iHeartRadio (such as thumbs-up and thumbs-
down) and the concept of “following” users in social media. Imagine
Radio and LAUNCHCast were eventually acquired by MTV and Yahoo!,
respectively.
Another innovation in streaming radio was aggregation services. By the
early 2000s, hundreds of AM/FM stations had started Internet simulcasts,
and many pure-play Internet radio stations found audiences for certain
niche musical genres (such as Soma FM for ambient music and Aural Moon
for progressive rock). The next logical step was to compile databases of all
64 Key Changes
Figure 3.2 The LAUNCHcast Internet radio player enabled users to rate songs,
artists, and albums.
Figure 3.3 The Kerbango Internet radio by 3Com, the first tabletop Internet
radio, on display at the Consumer Electronics Show in Las Vegas in January,
2001. REUTERS/Alamy Stock Photo.
Startups took two different approaches to this task. The first was based
on the idea that users would contribute this metadata by what we now call
“crowdsourcing.” Last.fm, which launched in 2002, was based on technology
called the Audioscrobbler. This enabled users to assign “tags” (metadata
values) to songs, which would be shared with other users and used to inform
automated musical selection.
The other approach was to pay trained musicologists to assign the meta-
data. That was the approach of the Music Genome Project (MGP), which had
begun by the time Last.fm launched. MGP hired hundreds of musicologists
to listen to songs and assign hundreds of metadata attributes to each one.2
Pandora, based on the MGP, launched in 2005. MGP produced metadata
that was better quality than crowdsourcing efforts such as Last.fm, leading
Pandora to become the most popular streaming music service for a while.
But the technique was expensive and didn’t scale very well: after several years,
Pandora’s music catalog was “only” 1–2 million tracks, while the catalogs of
iTunes and various interactive streaming services, not to mention streaming
radio competitors such as Last.fm and Slacker, went into the tens of millions.
66 Key Changes
Apart from the prevalence of live music in the early days of radio, the
history of the relationship between radio and the music business is best
understood through the history of radio music formats. Formats provide
a top-down view of radio that the music industry can use to help it pro-
mote music.
The first recognizable “music format” on radio was Top 40. Top 40 came
into existence before the transistor radio; it started in 1951 at Todd Storz’s
KOWH in Omaha. But the little device turbocharged the format’s popularity;
it expanded to hundreds of stations nationwide by the end of the 1950s. The
first rock & roll stars—Elvis, Bill Haley, Chuck Berry—came on the scene
in the years immediately following the introduction of the TR-1. Country
radio stations also proliferated throughout the South and Midwest around
the same time.v
Radio became an integral part of the recorded music industry with the
advent of Top 40. Yet record labels understood the value of radio airplay
in promoting sales long before then. The formula for labels was simple: get
radio airplay; sell records; get more airplay; sell more records. It was a vir-
tuous cycle: radio station asks the local record store what’s selling; record
store provides a list, often in exchange for a free ad on the air; station plays
the hot-selling records and tells listeners where to buy them; sales increase;
and the feedback loop continues.
Radio began to target young baby-boom teenagers as television was
displacing radio as the dominant medium for families sitting together in the
living room. As those teens grew older and got jobs and pocket money, the
labels began to understand that they could use the virtuous cycle to their ad-
vantage: to promote musical performers who would benefit from that cycle
and become disproportionately popular. That is, to create pop stars. Record
labels started to figure this out around the outset of the 1960s, leading Top 40
radio and the record industry to grow symbiotically.
Radio 69
Later in the 1960s, two changes affected the relationship between record
labels and radio. One was the separation of FM from AM. In its early days, FM
suffered from a Catch-22 situation: signal ranges were limited, FM receivers
were more expensive, and portable FM radios weren’t available (until the
early 1960s), so consumers didn’t adopt FM; the lack of demand discour-
aged investment in FM transmitters and programming, which in turn inhib-
ited demand even further. So the FCC allowed broadcasters that owned both
AM and FM stations to simulcast their AM signals on FM, to help encourage
adoption. But then as FM started to catch on through the mid-1960s, the
FCC decided to curtail simulcasting in order to increase programming di-
versity. It passed a regulation in 1964 that required AM/FM station owners
in markets larger than a population of 100,000 to limit simulcasting; the reg-
ulation went into effect at the beginning of 1967.vi FM stations had to start
coming up with their own programming.
The other big change around the same time was pop music artists starting
to think of LPs instead of singles as the primary canvases for their output,
as we discuss in the next chapter. LPs were no longer to be thought of as
collections of a few hit singles plus some filler tracks; they were conceived
as integrated artistic creations. And with its superior sound quality and lim-
ited commercial expectations at the time, FM was the natural place to play
groundbreaking albums from artists like the Beatles and Bob Dylan.
Progressive FM radio was born. DJs would play the latest rock records
along with jazz, soul, folk, world music, and spoken word records; and they
would organize the music into thematic sets—by topic, musician, songwriter,
mood, or whatever. WOR-FM in New York and KMPX in San Francisco
started the Progressive FM trend, and it expanded nationwide.
At the same time, the far-less controversial Beautiful Music format (later
called Easy Listening) started to flourish: stations that played light orches-
tral music from the likes of Ray Conniff, Mantovani, and Percy Faith, with
minimal announcements or other interruptions, intended for background
listening and inexpensive to produce. Beautiful Music as a format predated
FM, but—as pioneered by Marlon Taylor’s WDVR in Philadelphia and then
syndicated nationally by Jim Schulke’s Stereo Radio Productions—it was an-
other natural for the high-fidelity medium. Although Beautiful Music didn’t
attract as much attention from record labels, it was important to the growth
of FM as a commercial proposition because of its higher ad revenue potential.
Both Beautiful Music/Easy Listening and Top 40 AM radio continued into
the 1970s much as they had been in the 1950s and 1960s. But FM rock radio
70 Key Changes
Apart from creative considerations that we’ll discuss later in this chapter,
labels found that they could maximize airplay by paying attention to
the “burn factor”—the length of time that radio would play a song before
listeners got tired of it (“burned out”). They would time their efforts to pro-
mote selected tracks on a new album so that by the time one track burned
out, they would already have started pushing another track to radio. A sus-
tained effort of track-by-track radio pushes could result in albums lingering
on the charts for as much as a year, leading to multiplatinum album sales.viii
The precursors of this promotional tactic were 1976–1978 blockbusters
like the Eagles’ Hotel California, Fleetwood Mac’s Rumours, and the Rolling
Stones’ Some Girls. When albums like these came out, FM rock stations
tended to put multiple tracks into heavy rotation at the same time, while
singles were released to Top 40 in sequence. For example, you’d hear four
tracks from Hotel California—“Life in the Fast Lane,” “Victim of Love,” “New
Kid in Town,” and the title track—all in simultaneous heavy rotation on FM
rock stations, even as the latter two tracks were being pushed to Top 40 se-
quentially (both hitting no. 1). Album airplay charts in R&R listed as many as
four tracks from each charting album, though albums with that many tracks
getting heavy airplay were rare.
This was just what Lee Abrams had in mind with the Superstars
format: focus on artists, not individual songs. But by that point in time, labels
were hoping to stretch out albums’ runs on FM rock radio by sequencing
track-by-track promotion just as they did on Top 40 AM. For example, Hotel
California was off the R&R album charts by July 1977, seven months after
it was released. The rule of thumb in the industry at the time was that a hit
single would last a maximum of 12 weeks on the radio. So if Hotel California
contained four songs that got heavy airplay, the album could have lasted a
year on the charts if the songs had been released sequentially.3
Of course, labels couldn’t really control which tracks AOR stations played
from an album and when. So they invented various tactics to get DJs’ atten-
tion and maximize each album’s run on radio. Some labels sent prerelease
records to radio stations, sometimes in exchange for airplay for other acts
on their rosters. United Artists distributed 7-inch EP-format “mini-albums”
containing four songs—the tracks that the label wanted to push to radio—
in packaging that resembled that of the 12-inch LPs.ix In the early 1980s,
Columbia started prereleasing lead-off singles from new albums to radio in
12-inch single format. A tactic used to stretch out an album’s radio presence
72 Key Changes
was to release a less “pop” track as a single first, to establish credibility, and
then release Top 40 tracks as the second or third singles.x
By the mid-1980s, as FM rock radio’s popularity started to peak, tight
formats and consultants had taken over. And the structure of the radio
industry began to change in ways that would affect music formats even
further. As we’ll discuss shortly, in 1984, the FCC began relaxing sta-
tion ownership caps and indicated that it would loosen them further in
the future. This led companies like Clear Channel to bulk up on station
ownership. This in turn led to more homogenous management practices,
including tighter playlists.
In other words, by that time FM was completing its journey from freeform
Progressive to being as tightly formatted as AM since the Todd Storz days. At
the same time, consolidation did lead to a proliferation of music formats that
the industry recognized, even though each of those formats featured tighter
and tighter playlists. As the big chains started buying multiple stations in the
same markets, it made sense to differentiate their formats instead of having
them compete with one another in the same format. As the 1980s wore on,
computers and listenership data enabled stations to focus more narrowly on
certain demographics more effectively.
Although music formats and selection schemes varied, sometimes subtly,
across the thousands of commercial radio stations in the United States, the
music industry recognized only a small number of formats on a nationwide
scale. The arbiter of those formats was R&R, which was the radio industry
Bible from 1973 to 2009. R&R published weekly radio play charts for each
format, based on data collected by Mediabase and subsequently Broadcast
Data Systems, services that monitored broadcast signals and noted the songs
they played.4
R&R played a major role in determining which music formats were im-
portant by deciding which charts to define and publish. If R&R published a
music format chart, it meant both that the format was important to the music
industry and that there were a significant number of stations playing that
format nationwide. Thus, R&R music format charts represented the nexus of
the relationship between the music industry and radio as it evolved over time.
Figure 3.4 shows all of the charts that R&R published throughout its
36 years of operation. When R&R started in the mid-1970s, the only charts
were National Airplay, Country, and Album Airplay (plus the short-lived Pop
40). National Airplay was essentially the radio Top 40 chart. Album Airplay
covered FM Progressive and then AOR. Country radio had always existed in
Radio 73
1974
1979
1984
1989
1994
1999
2004
2009
1974
1979
1984
1989
1994
1999
2004
2009
Figure 3.4 Radio station music format charts that appeared in Radio & Records
from 1975 to 2009. Bill Rosenblatt.
its own world relative to rock and other genres; the Country chart persisted
untouched for the entire 36-year run.
The Album Airplay chart, renamed AOR in 1981, continued intact
through the heyday of the AOR format until 1988. The creation of the New
Rock chart in that year reflected two developments in rock radio during the
1980s. One was the influence that college radio had been building up for sev-
eral years. Many college radio stations had been playing punk and new wave,
music that very few commercial rock stations were playing; yet the records
were selling. Thanks to another industry publication called College Media
Journal (CMJ), the labels figured out that those sales were coming through
college radio airplay and started paying attention.
74 Key Changes
Bob Haber started CMJ in 1978 while he was a student at Brandeis. CMJ
(later renamed CMJ New Music Report) enabled college radio stations—for
the first time ever—to get information on what music other stations were
playing. It got dozens of subscribing stations around the country to report
airplay, and it published charts.
CMJ’s importance rose in tandem with that of the new rock music through
the 1980s. Eventually the CMJ charts and other editorial content became the
equivalent of tip sheets for college radio programmers looking for advice on
what was hip. The idea of a “college radio sound”—including amateurish DJs
and technical glitches, as well as the music—emerged; college radio started to
homogenize in its own way, and in this fashion, CMJ delivered college radio
to the labels (insofar as it could be delivered).
The other factor that caused the bifurcation of commercial rock radio
formats was AOR’s increasing emphasis on older “classic rock”5 material in
order to court the baby-boomer demographic. Younger listeners defected to
New Rock stations, which attracted them with music from the likes of the
Psychedelic Furs, the Replacements, and the Jesus and Mary Chain—all art-
ists that got their initial audiences on college radio.
The next wave of expansion of R&R format charts coincided with the
FCC’s further loosening of ownership caps with the Telecommunications
Act of 1996, which was introduced into Congress the previous year. The 1996
Act eliminated national ownership caps but kept them in place for individual
markets, so that station owners were “limited” to owning eight stations in
each of the largest markets (New York, Los Angeles, etc.), with lower limits
for smaller markets.
The big chains moved quickly to expand ownership based on these new
limits. In doing so, they took on massive amounts of debt that required ag-
gressive revenue growth to repay; this made them even more conservative
and ratings-focused than they had been. Broadcast chains proliferated their
stations’ formats in efforts to reach ever-narrower demographics with ever-
tighter playlists.
In response, R&R initiated several new charts between 1994 and 1996. By
then, R&R had renamed Top 40 as Contemporary Hit Radio (CHR). CHR
bifurcated into CHR/Pop (Hootie and the Blowfish, Bryan Adams, Dave
Matthews Band) and CHR/Rhythmic (Boyz II Men, TLC, Janet Jackson).
Adult Contemporary (AC) spun off Hot AC/Adult CHR (Melissa Etheridge,
Elton John, Sting). Rock (formerly AOR) spun off Progressive (Soul Asylum,
Natalie Merchant, Tears for Fears), which was renamed Adult Alternative,
Radio 75
a.k.a. AAA or Triple-A for Adult Album Alternative. Rock also spun off
Active Rock (metal and grunge). Americana was a short-lived chart that
spun out of AAA and incorporated roots-rock acts (Steve Earle, Delbert
McClinton) with a smattering of country and alt-country artists (Uncle
Tupelo, Dwight Yoakam) who weren’t commercial or traditional enough for
the Country charts.
Industry racism rendered Black radio virtually invisible during the
early years of R&R. The industry had long referred to R&B, soul, and
gospel music as “race records.” A parallel, de facto segregated industry
of African American labels and radio stations existed through the 1940s,
after which white label owners such as Ahmet Ertegun (Atlantic) and
Leonard Chess (Chess) started signing Black artists. Billboard changed
the name of its Race Records chart to Rhythm & Blues in 1949 and then
to Soul in 1969, but R&R did not include a chart for that music during its
first several years.
Black stations were the lifeblood of so much of that music at that time;
they broke many Black artists and hits before they crossed over to Top 40. Yet
they did not gain representation in R&R charts until the height of the disco
boom in the late 1970s.6 R&R started to track disco-format radio stations
largely because of the phenomenal success of the 1977 movie Saturday Night
Fever, which crossed disco over to white audiences and whose double-LP
soundtrack became one of the top-selling albums of all time. The first disco
radio charts were called, euphemistically, “Black Popular Rhythms” and then
“Dance.” It wasn’t until 1983 that Black music stations in general—quickly
renamed “Urban”—got their own charts. (Billboard renamed its Soul chart to
Black in 1982.)
R&R didn’t add Latin charts until 2003. It had acquired the Latino trade
publication Radio y Musica in 2000, and it incorporated that magazine’s
charts and content into its own three years later. It experimented with sev-
eral different music format charts for Latino radio, then gave up in early 2009
and stopped publishing them entirely. (R&R went out of business shortly
thereafter.)
While racism may well have contributed to R&R’s lack of Latin chart cov-
erage until the early 2000s, other factors were involved. A music format
would have to have a national footprint to merit a chart in R&R. Yet Spanish-
language radio resists simple categorization at a national level. Hispanic
populations (both Spanish-and English-speaking) are quite diverse across
different American cities. For example, according to data from the 2009
76 Key Changes
Bob Marley, Miles Davis, and other artists have channels that play their
music 24/7.
Radio achieved its position of power and influence in the music industry
because it was by far the most efficient way to promote recorded music for a
long time. Yet its ability to do that has rested largely on a couple of quirks in
US copyright law.
It’s important to bear in mind that each music track normally has two
copyrights: one for the musical composition, typically owned by songwriters
and music publishers; the other for the sound recording of the composition,
typically owned by labels or artists. Copyright law says that copyright owners
have certain exclusive rights to their creations, but copyrights in sound
recordings are more limited than copyrights in musical compositions. (In the
next chapter, we’ll see how and why this came to be.) Specifically, songwriters
and music publishers have exclusive rights to public performances of their
compositions, but recording artists and labels don’t have exclusive rights to
public performances of their records. Radio broadcasts are considered to be
public performances.
The lack of a performance right in sound recordings means that radio sta-
tions have never had to license records to play them on air. They have only
had to license the musical compositions.
As saw in Chapter 2, arguments over whether copyrights should exist
for sound recordings date back to the age of wax cylinders and player
pianos; those rights include the right of public performance on radio.
Those arguments continued for decades. For example, the president of the
musicians’ union, the American Federation of Musicians, had this to say at a
1961 congressional hearing:
“[I]t is a shocking crime that people like Mr. Leopold Stokowski or Leonard
Bernstein, or Louis Armstrong, or whoever the artist may be, are denied the
right to receive additional fees, when money is made with his product. All
you have to do is put a radio set into this room today and you can listen for
hours and hours to canned music here, records received free by the broad-
caster, if you please, while the men who made them are sitting home trying
to figure out how to pay for their children’s education.”xii
78 Key Changes
As we’ll see in the next chapter, when Congress finally enacted copyrights
in sound recordings with the Sound Recording Act of 1971, it didn’t include a
public performance right, despite recommendations from the US Copyright
Office that it be included.
As far as musical compositions are concerned, court decisions in 1940 es-
tablished that radio stations could purchase records and play them on the air
as many times as they wanted without having to seek permission from ei-
ther labels or music publishers.xiii In other words, radio stations have a com-
pulsory license to the musical compositions performed on the records they
play, analogous to the compulsory license for mechanical reproductions of
compositions that we saw in Chapter 2. Stations do have to pay royalties on
musical compositions, but that is simple to do through PROs (performance
rights organizations), which license public performance rights to musical
compositions on behalf of large numbers of songwriters and music publishers.
As we’ve seen, ASCAP, the first PRO, formed before the start of radio.
Next was BMI (Broadcast Music Inc.), which focused primarily on licensing
music for radio. BMI was formed in 1939 by the National Association of
Broadcasters (NAB), the radio industry’s trade association, to inject price
competition as ASCAP kept raising its royalty rates. ASCAP and BMI re-
ceived antitrust consent decrees from the Department of Justice in 1941;
these are essentially government permission slips to operate a duopoly.
(There was a third PRO, SESAC, which focused on gospel music and
European composers at that time; its market share was negligible compared
to ASCAP and BMI.)
Today, ASCAP and BMI cover most of the market with roughly equal-
sized catalogs, but there are other smaller PROs that also collect royalties
from radio. SESAC has broadened its catalog to include compositions by the
likes of Bob Dylan, Neil Diamond, and Rush. A PRO that licenses broadcast
radio with an even smaller catalog is Global Music Rights (GMR). Veteran
artist manager Irving Azoff founded GMR in 2013 by convincing a group of
big-name songwriters that they could get higher royalties through a separate
PRO than they were getting through ASCAP or BMI. GMR’s current catalog
includes only 90 songwriters, compared to over a million for BMI, but they
include the likes of Bruce Springsteen, Bruno Mars, Prince, Pete Townshend,
and John Lennon. And a new fifth PRO called Pro Music Rights, which
licenses a catalog of mostly hip-hop compositions, has been recognized by
some radio broadcasters.
Radio 79
PROs offer blanket licenses, meaning that a radio station just has to pay
each PRO a license fee and it gets a license to all of the compositions in the
PRO’s repertoire. The PRO then figures out whom to pay for airplay and how
much. (We’ll talk about that in the Cash section later in this chapter.)
In other words, radio stations—unlike streaming services, as we’ll see—
have had an easy path to licensing music, enabling them to play whatever
music they want whenever they want.
Congress had reopened the question of performance rights in sound
recordings in 1978, just two years after the major 1976 revision of the
Copyright Act had been signed into law, but nothing came of it,xiv and the
matter was dropped for a while. But the advent of digital radio in the 1990s
brought the subject back yet again. Although labels had long viewed radio
airplay as promotional, they weren’t inclined to view the new digital radio
formats that way. On the contrary: they were afraid that digital audio tech-
nology, with its ability to make perfect copies, would lead to services that
cannibalized record sales.
So the labels got the Copyright Office to open inquiries on the effects of
digital audio transmission on music piracy in 1991, which led to Congress
passing the Digital Performance Right in Sound Recordings Act (DPRA) in
1995. The DPRA established the performance right for sound recordings, but
only for digital radio; it left intact the lack of a performance right in sound
recordings for traditional AM and FM. The DPRA established a procedure
for setting statutory royalties (i.e., rates set by law instead of by negotiation)
for digital radio services, processes that the US Copyright Office runs every
five years and which resemble litigations in court; we’ll talk more about this
shortly.
The DPRA also established certain rules for AM/FM terrestrial broadcast
stations that simulcast online (most FCC-licensed broadcast stations do) to
prevent people from using them to make digital copies of music. These sta-
tions weren’t allowed, for example, to play more than four songs by the same
artist in a three-hour period, or to preannounce any specific songs.7 The
DPRA also established that interactive streaming services—which wouldn’t
be in existence until several years in the future—would need to negotiate
licenses with record labels and wouldn’t get statutory royalty rates.
The result of the DPRA was that all types of digital radio services—Internet
pure-play, AM/FM simulcast, satellite, the music channels on cable TV serv-
ices, and Muzak—got compulsory licenses to play any sound recordings they
80 Key Changes
wanted as long as they pay the statutory royalties. Separate royalty rates were
eventually set up for the various types of digital radio.
However, as Internet radio services got more customizable, rightsholders
began to argue that they had become tantamount to interactive services.
In 2001, a group of record labels sued LAUNCH Media, the company that
operated LAUNCHcast (which was offered through Yahoo!). Recall that
LAUNCHcast offered features for users to customize their music feeds, such
as by rating music and following other users; but it stopped short of enabling
users to select specific tracks or artists. The labels claimed that this feature set
was “interactive” and therefore that LAUNCHcast should have to negotiate
license terms rather than get a compulsory license with statutory royalties.
The Second Circuit appeals court found for LAUNCH Media in 2009. That
decision paved the way for later customizable Internet radio services like
Pandora and iHeartRadio, which were able to offer any recorded music track
through compulsory licenses and pay statutory royalty rates.
Another loophole regarding sound recording copyrights in digital radio
remained for many years afterwards. The DPRA established copyright pro-
tection for digital broadcasts of sound recordings, but only as far back as
1972—because the Sound Recording Act of 1971 did not make it retroactive
to older recordings. Sound recordings made before February 1972 were sub-
ject to a patchwork of state copyright laws. This meant that digital radio serv-
ices technically could not rely on statutory licenses and needed to negotiate
licenses with labels for playing much of the music featured on oldies, classic
rock, jazz, and classical stations—though of course this never happened.
A series of lawsuits were filed by the likes of Flo & Eddie (lead singers of the
Turtles as well as solo artists) and ABS Entertainment (label of Sam Cooke,
Jackie Wilson, and other classic artists).
This loophole was finally closed in 2018 with the passage of the Music
Modernization Act (MMA), although only for digital radio. The MMA was
primarily focused on fixes to mechanical licensing of musical compositions
for interactive streaming, as we’ll see in Chapter 9. But it also included a pro-
vision that was originally called the CLASSICS Act (Compensating Legacy
Artists for their Songs, Service, and Important Contributions to Society),
which established a performance right in pre-1972 sound recordings for dig-
ital radio.
The final loophole wasn’t closed: there is still no performance right in sound
recordings (of any vintage) for terrestrial radio. The United States continues
to be one of only a handful of countries in the world without this right; others
Radio 81
include Iran, North Korea, and Rwanda.8 AM and FM broadcasters still don’t
have to pay royalties to record labels or recording artists.
Arguments over whether radio should pay royalties to labels and artists
have taken place numerous times since the 1930s. Recent arguments turn
on whether radio still has the promotional value for recorded music that it
did in the predigital era. Whenever this issue comes up inside the Beltway,
both sides—the RIAA for record labels, the NAB for broadcasters—unleash
economists armed with studies purporting to show that it does or doesn’t.
But the real reason why there is still no general performance right in sound
recordings has to do with lobbying at the state level. And that takes us from
copyright law to the FCC.
Radio frequencies are a limited resource, like land and water; so the gov-
ernment has regulated their use through the FCC (originally the Federal
Radio Commission), which was established with the Communications Act
of 1934. As radio stations proliferated around the country, the FCC moved to
establish ownership limits: a single company could not own or operate more
than seven stations nationwide, and no more than one in any given market.
Although ownership caps were eventually relaxed in the 1980s (along with
requirements for public-service programming and other regulations),
they helped shape a market for radio stations that had no nationally dom-
inant players. As we mentioned previously, national ownership caps were
eliminated with the Communications Act of 1996. But by that time no more
new frequencies were available, so if you wanted to own a radio station, you
had to buy the license from someone else, which cost into the tens of millions
for powerful stations in major markets.
When the 1996 Act passed, the big broadcast chains went on buying
sprees, and by the mid-2000s, four chains (Clear Channel, Citadel, Viacom,
and Cumulus) had an aggregate dominant share of audience and revenue,
and the total number of station owners had decreased by one-third.xv But
the total number of stations that those four chains owned was less than
one-fifth of the total number of FCC-licensed stations. Even at its peak in
2003, Clear Channel owned around 1,200 stations out of about 15,000. (Now
iHeartMedia owns about 850.)
As a result, most broadcast stations are owned by small-to-medium-sized
businesses that are based all over the country. In contrast, the recorded music
industry is concentrated heavily in three states: New York (NYC), Tennessee
(Nashville), and California (Los Angeles/ Santa Monica/ Burbank). This
means that whenever Congress considers legislation that pits the interests
82 Key Changes
of radio against those of the music industry, there are members of Congress
in three states being lobbied by the RIAA, while members of Congress in
many of the other states are being lobbied by the NAB. NAB’s nationwide
reach gives them the clout to portray any royalty as a “performance tax,” and
raising taxes tends to be anathema for many politicians.
That’s why there is no performance right in sound recordings today, and
radio pays no royalties to labels or artists. The clash between the labels and
the broadcasting industry over this issue continues to this day. At this writing,
yet another attempt to plug this legislative hole is making its way through
Congress: the American Music Fairness Act, introduced by senators and rep-
resentatives from—you guessed it—New York, California, and Tennessee.9
At the same time, as Internet radio becomes less important with the rise
of interactive streaming services (which have all subsumed Internet radio
features), attention in radio is turning to podcasts. And as we’ll see, copyright
issues attending podcasts could cause radio’s relationship with the music in-
dustry to become more distant.
Podcasts are an increasingly popular media format; currently over 100 mil-
lion Americans listen to podcasts on at least a monthly basis.xvi Podcasts can
be thought of as “talk radio on demand.” Indeed, two of the most popular
publishers of podcast content are two large sources of produced spoken-word
radio programming: National Public Radio (NPR) and iHeartMedia. Both
companies started in podcasting by simply taking their syndicated radio
shows—Rush Limbaugh, Dr. Laura Schlessinger, and Glenn Beck in the case
of iHeart; “Fresh Air,” “Wait, Wait, Don’t Tell Me,” and “Planet Money” in the
case of NPR—and releasing them as podcasts. Both have added many shows
as podcasts since then.
But there’s a problem with this regarding music licensing. When you make
a broadcast program available on demand, you subject it to all the rules for
licensing music for interactive services, which we’ll discuss later in this book.
For example, let’s say you produce a public affairs show that has theme and
background music. If your show airs once a week on an FM station, you
don’t need any specific music licenses; the musical compositions are cov-
ered under the station’s blanket licenses with the relevant PROs,10 and no
licenses are necessary for the sound recordings. But if you make your show
available for download or streaming online, you need several licenses. That
means negotiating with record labels, obtaining licenses for compositions
from music publishers, and generally doing more paperwork and paying
more royalties. Unfortunately, there is neither a simple licensing mechanism
Radio 83
for podcast use nor a statutory or other published royalty rate: every license
has to be negotiated by hand, and the publishers and labels can refuse to li-
cense at all.xvii
As a result of this, not much music is used on podcasts, and there are few
podcasts about music that have music in them. (That’s why, for example,
there is no podcast for iHeartRadio’s “American Top 40” syndicated show.)
As for background or theme music, many podcasters will commission their
own or go to various online sources of royalty-free production music.
The further the radio industry migrates into podcasting, the less it will
rely on music—unless and until changes are made to the way it’s licensed for
that purpose. Podcasting brought in $1.4 billion in ad revenue in 2021 and
is projected to exceed $2 billion in 2022 and $3 billion in 2023.xviii The more
it turns away from licensing commercial music, the more opportunity for
labels and music publishers will be lost. This is a “brave new world” area of
copyright law and is evolving fast.11
Payola didn’t start in the Top 40 era; the payola-like practice of “song plug-
ging”—paying bandleaders to play musical compositions live on the air—
existed in the earliest days of radio. But the practice came into public view
through its association with pop music radio in the late 1950s.
The story of the payola scandals of the late 1950s is well-known: Congress
held hearings and passed a law making paying DJs for record airplay illegal.
The hearings destroyed the career of Freed, arguably the creator of the term
“rock & roll,” while at the same time letting Philadelphia’s Dick Clark of
American Bandstand fame “skate.” (Freed was defiant on the witness stand,
saying that he only took payola for records that he would have played an-
yway; Clark denied taking payments and quickly divested his ownership
stakes in dozens of songs and record labels.)
The 1950s payola scandals have been described on three levels. The most
superficial explanation is that both DJs and record labels were corrupt, and
it took Congress to put a stop to this evil practice that used bribery to sully
the process of—what exactly? Art? Free expression? Serving the needs of
listeners?
Deeper accounts of what happened around that time describe the
payola scandals as an outgrowth of certain segments of society’s revul-
sion to rock & roll as an insidious influence on youth, a revulsion that
was rooted in racist attitudes. (Freed played a lot of music by Black art-
ists on his shows.) But even that doesn’t explain everything. The true
root of the payola scandals was money: not just money paid to DJs, but
also money that radio stations paid—or didn’t pay—to music publishers
and songwriters. And that leads to the deepest level of explanation of the
payola scandals.
As we’ve discussed, broadcast radio stations didn’t (and still don’t) pay
record labels royalties for sound recordings, but they do pay royalties on mu-
sical compositions. In the late 1950s, almost all such royalty payments were
made through ASCAP and BMI. At that time, ASCAP’s stable of songwriters
and music publishers mainly included Tin Pan Alley writers of pop standards
for the likes of Bing Crosby and Frank Sinatra, while BMI had the bulk of
rock & roll songwriters in its catalog. Thus, radio stations that played Top 40
rock & roll were playing BMI compositions.
BMI was ASCAP’s archrival, created by the radio industry itself to com-
pete with ASCAP. ASCAP saw that if it could take down rock & roll, then
radio would have to play compositions in its catalog, and it could dominate
the radio market again. ASCAP figured that it could do this by capitalizing
Radio 85
on the “rock & roll is evil” sentiment and lobbying Congress to go after rock
& roll by going after payola.xix
ASCAP’s gambit worked: Congress enacted a law making payola illegal
in 1960. But it was a finger in the dike. The law defined payola narrowly as
payments to DJs that weren’t disclosed on the air and had the express pur-
pose of getting specific records played. Its penalties were relatively mild: max-
imum fines of $10,000 (about $90,000 today) and maximum prison sentences
of one year. Some have argued not only that the law was toothless but that its
effects were insufficiently considered.xx
Payola has been around in one form or another ever since. In fact, it
has been said that all that the law accomplished was to change where the
payments went: instead of going to individual DJs, they went to station pro-
gram directors, who then influenced or dictated the programming choices of
all DJs on each station. This, ironically, was yet another step in the homog-
enization of music programming on radio that we discussed earlier in this
chapter.xxi
The next attempt to enforce antipayola regulations, in the early 1970s, was
to investigate record labels, the alleged sources of payola, instead of radio sta-
tions. In 1973, CBS abruptly fired and then sued the renowned record exec-
utive Clive Davis, claiming misuse of $94,000 for payola and “drugola.”12,xxii
The US Attorney’s Office in Newark, NJ opened an investigation, code-
named “Project Sound.” Two senators, James Buckley of New York and John
McClellan of Arkansas, called for further investigations.
The Project Sound investigations dragged on for two years, but they
yielded convictions of only a handful of executives from CBS, including
Davis, and the R&B label Brunswick Records—and for tax evasion, not
payola violations.xxiii The prosecutors also investigated Gamble-Huff Record
Co., owners of the Philly Soul label Philadelphia International, but ended up
dropping charges when they couldn’t get anyone to corroborate the evidence.
Once again, this investigation had racist overtones, as most of its targets were
connected to Black music.13,xxiv
Meanwhile, most broadcasters took steps to curtail payola, such as by
making DJs sign no-payola verifications and setting hard limits on the value
of gifts that DJs could accept from labels. Still, labels had various ways to
curry favor with DJs. They would invite jocks to “listening parties” at fancy
restaurants, where they would hear the latest releases over nice meals and
open bars; or they would invite DJs to evenings built around artists’ concerts
or club gigs: nice dinner, limo ride to the venue, and after-party with the artist.
86 Key Changes
Even the DJs who refused drugs, stereos, plane tickets, or other enticements
took part in these events.xxv Radio stations also get copies of new releases
from record labels at no charge; this practice started during World War II14
and became so routine that it’s known as getting “service” from labels.
The labels’ next steps were to insulate themselves from radio stations by
hiring independent promoters. Labels would pay promoters to do “research”;
the promoters would handle payments and other enticements to DJs, pro-
gram directors, and music directors. As the argument went, labels couldn’t
be held liable because they were paying independent promoters flat fees and
had no control over where the money was going after that. Independent
promoters would charge labels tens of thousands of dollars to get a song
added to major-market radio stations’ playlists.xxvi
This led to another round of investigations in the mid-1980s, first in the
House of Representatives, initiated by Rep. John Dingell of Michigan in
1984. A story that aired on NBC Nightly News in 1986 focused on a group of
high-level indie promoters that called itself The Network and included no-
table figures such as Joe Isgro and Fred DiSipio, and insinuated that it had ties
to organized crime.xxvii This prompted an investigation in the Senate, which
featured two future political superstars: a young senator from Tennessee
named Al Gore and a rising federal prosecutor from New York named Rudy
Giuliani.xxviii The following year, a Los Angeles grand jury launched yet an-
other investigation.xxix
Like the Project Sound investigations in the previous decade, these
investigations focused on labels, and they followed on the heels of failed
attempts to get the RIAA to launch its own investigation of labels’ behavior.
Yet none of them found much behavior that would be considered violations
of the laws. The negative exposure from the NBC exposé led the major labels
to stop using the promoters in The Network, though only temporarily.xxx
Labels cut down on independent promoters after the Napster era of the
1990s (see Chapter 8) as both the music and radio businesses began to de-
cline, but then labels allegedly started paying radio stations directly again,
and evidence emerged that labels were at least cognizant of payments that
independent promoters made to individual stations.
The next wave of payola investigations started in 2002 under New York
State Attorney General Eliot Spitzer, who was tipped off by a music business
lawyer named Bob Donnelley.xxxi By June 2006, all of the then four major
labels had agreed to settlements totaling over $30 million.xxxii The notoriety
that Spitzer received from these investigations would help land him in the
Radio 87
New York Governor’s office the following year. Around the same time, the
FCC launched its own investigation of station owners, focusing on four of
the largest: Clear Channel, CBS, Citadel, and Entercom. These resulted in
fines totaling $12.5 million in 2007.xxxiii Labels as well as broadcast chains set
up another round of new procedures designed to eschew payola as the law
defined it.
The most recent attempts to rein in payola came in 2019, when the FCC
launched an investigation on the major labels that was prompted by an article
in Rolling Stone.xxxiv This time the intent of the inquiry was different because
of the antiregulatory bent of the Republican-controlled FCC15 when the in-
vestigation started: its focus appeared to be on the imbalance between payola
regulations on broadcast radio and the lack of such regulations on digital
music services. The result might have been a recommendation to Congress
that it discontinue the anti-payola laws that it made sixty years ago. But after
the presidency shifted to the Democratic Party in 2021, the investigation
faded away.
The money flows on the songwriter and music publisher side are com-
pletely different for broadcast radio. As we discussed, radio pays public per-
formance royalties on musical compositions; PROs handle those payments.
There are three steps to the process of paying radio royalties to music
publishers and songwriters. First, stations pay PROs percentages of their rev-
enue for blanket licenses to the PROs’ repertoire. Each PRO negotiates the
rates with an entity called the Radio Music Licensing Committee (RMLC),
which represents radio broadcasters. Those negotiations generally turn on
how much of the music that stations play is in each PRO’s catalog. Each sta-
tion is free to reject the blanket license and pay royalties per composition
(and negotiate directly with the PRO), but the process would be a nightmare
of paperwork. The PROs usually reach agreements with RMLC, but if they
don’t, the negotiations end up in court.
In the second step, each PRO pays songwriters and music publishers ac-
cording to how much relative airplay their compositions are getting. BMI
used to require all radio stations that took the blanket license to fill out sep-
arate music logs for two weeks out of each year. It would extrapolate those
two weeks to the entire year and pay its rights holders proportionally. For
loosely formatted stations that gave DJs freedom to play what they wanted,
this system would invariably result in less popular “long tail” artists being
paid nothing. Nowadays, the two major PROs use automated music recog-
nition technology, which we’ll discuss in more detail in Chapter 8, to analyze
88 Key Changes
radio services, but if the negotiations break down, federal courts determine
the rates. The Music Modernization Act of 2018, which we’ll discuss in more
detail in Chapter 9, contains tweaks to this process that should result in
higher royalty rates for songwriters and music publishers from digital radio.
In the early days of Top 40, radio airplay was so fundamental to record
sales that tailoring songs for radio airplay was synonymous with tailoring
them to be hits. This was especially true for producers. By the early 1960s,
record producers became known for engineering records to be radio-
friendly—which meant standing out on the portable transistor radios that
were ubiquitous by then.
Two 1960s producers who were especially noted for radio-oriented sound
were Phil Spector, of “wall of sound” fame, and Joe Meek, best known for
producing the Tornados’ 1962 instrumental “Telstar.” But the true home of
radio-friendly production was Motown. Many engineers and producers in
New York and London looked to Motown and tried to emulate its sound.
Noted producer Tony Bongiovi flew from New York to Detroit regularly for
several years to learn his craft before deploying it on records by Jimi Hendrix,
Gloria Gaynor, the Ramones, Talking Heads, Bon Jovi, Aerosmith, and a long
list of others.xxxv
Sometimes artists got involved too: country music legend Buck Owens
had producer Ken Nelson install small mono speakers in the control booth at
Capitol Studios in Los Angeles to approximate AM radio sound quality.xxxvi
Later on, it became standard practice to equip studios with low-power radio
transmitters so that producers and artists could listen to mixes in their cars
in studio parking lots.xxxvii The British rock band Def Leppard had producer
Robert John “Mutt” Lange reach new heights of obsessive perfectionism to
make their 1987 album Hysteria radio-friendly—including recording each
note of each guitar chord on a separate track of analog tape. The album took
three years to produce.xxxviii,18
By the 1990s, producers and engineers found themselves in competition
with radio engineers and equipment makers to produce the loudest, brightest
possible sound—the so-called Loudness Wars. In some cases, this led to
perverse outcomes. For example, the Red Hot Chili Peppers’ 1999 album
Californication was produced by Rick Rubin and mastered by engineer Vlado
90 Key Changes
Meller. Meller’s and radio stations’ arsenals of effects cancelled each other
out, with the result that the album’s lead-off single, “Scar Tissue,” ended up
sounding softer than the surrounding material on the air.xxxix
Engineers embraced the Loudness Wars in the late 1990s and early 2000s
despite the larger dynamic range afforded by the CDs of the day. Dynamic
range has come back, to some degree, since then.xl
The other type of concession to radio airplay that producers and labels
made was song length. As the 1960s wore on, questions started to arise about
how longer or larger pieces could get radio airplay. An eventual answer—
as we discussed earlier in this chapter—was FM, as opposed to Top 40 AM.
But in the meantime, the first solution to this problem was simple: edit long
songs to radio length.
The first “radio edit” was Bob Dylan’s “Like a Rolling Stone.” Columbia
Records was reluctant to release the song at all because of its 6 minute, 13
second length,xli but a copy leaked to DJ Bill “Rosko” Mercer, then of KLBA
in Burbank, who played it at a club in Los Angeles to a rapturous audience.xlii
The song made it onto Dylan’s classic album Highway 61 Revisited, released
in August 1965. But a month earlier, Columbia released a promotional single
to radio that had the first part of the song on side A, with a fadeout at the end,
and the second part on side B.19 At first, radio played only the A side of this
single. But then it bowed to pressure from fans and played the single released
to retail, which contained the full song on the A side (with “Gates of Eden” on
side B). The song reached no. 2 on the Billboard charts.
On the other hand, there was “Light My Fire,” from the Doors’ self-titled
debut album in April 1967. The song ran more than seven minutes long.
DJ Dave Diamond of KHJ, a Top 40 station in Los Angeles, asked Elektra
Records for an edited version for radio airplay; Elektra owner Jac Holzman
responded with a 2 minute, 52 second version that cut the extended instru-
mental break in the middle of the song.xliii It was a no. 1 hit.
Radio edits became more and more commonplace as the years went on.
A radio promotional rep at a major label once said, “We’ll butcher it any way
you like if it gets played.”xliv In some cases, record labels fibbed about the
length of singles, such as labeling a 3 minute, 8 second song as 2 minutes,
58 seconds to get it on the air.xlv In the early 1970s, Top 40 stations started
creating their own edits, sometimes speeding songs up in earlier attempts to
create a brighter sound and to be able to claim that they played more songs
per hour than their competition.
Radio 91
The types of edits made for radio varied. The edit of the instrumental
break out of “Light My Fire” was a precursor to the slashing of Iron Butterfly’s
17-minute acid-rock marathon “In-A-Gadda-Da-Vida” (1968) down to
the same 2 minutes, 52 seconds as “Light My Fire.” Prog-rock epics such as
Yes’s “Roundabout” and Pink Floyd’s “Money” got the same treatment in the
early 1970s. Other types of edits involved cutting or shortening intros, as in
Santana’s “Black Magic Woman,” Steely Dan’s “Do It Again,” and Parliament’s
“Tear the Roof Off the Sucker.” Sometimes verses were dropped (as in Billy
Joel’s “Piano Man,” the Eagles’ “Lyin’ Eyes,” and Dire Straits’ “Money for
Nothing”).
Yet at the same time, stations started making exceptions to the three-
minute rule for certain songs—like the Beatles’ “Hey Jude” and Richard
Harris’s “MacArthur Park,” both of which were released in 1968 and ran past
the seven-minute mark. By 1971, a few songs that got heavy airplay even
went beyond eight minutes, such as Led Zeppelin’s “Stairway to Heaven” and
Don McLean’s “American Pie.” Even some radio edits exceeded 3 minutes,
such as the Temptations’ “Papa Was a Rollin’ Stone” (1972), which ran almost
12 minutes on the album All Directions; the single was only cut down to 6
minutes, 54 seconds but still reached no. 1.
By the 1980s, some “radio edits” were made in the songwriting process.
As electronic tuning became widely available, it became easier for listeners
to change to a greater number of stations, making station-switching more
frequent; this behavior showed up in Arbitron listenership numbers. This led
to a drive to produce singles that got to the chorus or main hook faster—
presaging the way TikTok is influencing songwriting today (see Chapter 10).
And as technology made it easier, artists began releasing more radio edits: for
example, Def Leppard released no less than five radio edits of songs from
Hysteria.
Of course, other radio edits weren’t for length—they were “clean” versions
designed to clear the FCC’s and radio station owners’ rather amorphous
indecency rules. The likely first “clean” version was Aretha Franklin’s 1967
smash hit “Respect,” which had the backing vocalists’ “Sock it to me” chorus
edited out for Top 40 radio.
92 Key Changes
Despite the rise of the Internet, radio has been amazingly resilient. Nielsen
(formerly Arbitron) data shows that the monthly reach of radio only decreased
from nearly 100 percent in mid-1990s pre-Internet days to 93 percent in
2021.xlvi The biggest reason why can be expressed in one word: automobile.
More than half of radio listening takes place in cars.20 Cars are places where
drivers can’t be distracted in fiddling with audio equipment; more than one
person is often listening, meaning that the choice of entertainment has to sat-
isfy everyone; and the skill of weaving in the kinds of nonmusical information
that car riders need (time, weather, traffic, news) has been refined over decades.
Two recent surveys confirm that radio listeners aren’t tuning in for the
music as much anymore. Radio listeners responding to Jacobs Media’s an-
nual Techsurvey answered questions about the main reasons why they like
radio. The top three reasons say it all: no. 1 is “Easiest to listen to in car,” no. 2
is “DJs/hosts/shows,” and no. 3 is “It’s free.” Not until no. 4 do we get “Hear
favorite songs/artists.” After that, none of the reasons relate to music until
no. 16, “Discover new music/new artists,” and no. 17, “Music surprises.” The
other reasons have to do with news, talk, weather, emotional connections
to radio, and force of habit. As Figure 3.5 indicates, past instances of the
Techsurvey indicate that the appeal of on-air personalities overtook the ap-
peal of music on the radio sometime between 2018 and 2019.xlvii
70%
67% 66% 61% 62%
64% 60% 59% 59%
60%
57% 58% 58% 57%
55% 56% 55% 55% 55%
40%
TS 2014 TS 2015 TS 2016 TS 2017 TS 2018 TS 2019 TS 2020 TS 2021 TS 2022
Among those who listen to AM/FM radio,
% who say this is a main reason they listen
Figure 3.5 Percentages of radio listeners who cite music and on-air personalities
as a main reason they listen to radio, 2014–2022. Jacobs Media, Techsurvey 2022.
Radio 93
2010 2021
Satellite, 5%
Satellite, 8%
Other, 2%
Other, 12%
Stream, 32%
Stream, 18%
AM/FM Radio,
AM/FM Radio,
45%
15%
Figure 3.6 Share of weekly music listening hours among US Internet users ages
13 and above, 2010 versus 2020. MusicWatch Audiocensus Study.
Other survey data from MusicWatch indicates that Internet users are
turning away from radio for music. As the Figure 3.6 shows, among the popu-
lation of US Internet users ages 13 and up, the amount of time spent listening
to music on AM/FM radio (including streams of stations) has plummeted
from 45 percent to 15 percent from 2010 to 2020, a steeper drop than for
“owned” music on physical media or downloads.
One reason why streaming has become so much more popular than radio
is that it affords dedicated music fans the tools to select their own music so
much more easily than buying and playing individual albums or songs. In
contrast, radio can offer only a “lean-back” experience that appeals to casual
music fans. At the same time, as we’ll see in Chapter 9, streaming services
have added plenty of lean-back features, further taking music listenership
away from radio.
As a result, the reasons why people still listen to radio are things that
only radio has and other forms of music distribution don’t. Apart from the
nonmusic content, ease of use is the biggest factor. Digital music (apart
from CDs) is too complicated to listen to in cars if you’re driving. That’s even
true today with the rise of Apple CarPlay and Google’s Android Auto, so-
called infotainment platforms that are designed to be seamless extensions
94 Key Changes
Figure 3.7 An adapter that enables portable audio devices to play through car
cassette players via their headphone outputs. Bill Rosenblatt.
Radio 95
The fact that radio is free still matters too. Some digital music services are
free, but very few have the equivalent of DJs or shows; Apple Music is the
only major digital music service that features “personality” DJs. iHeartRadio
is using AI-based technology, from a startup company called SuperHiFi that
we’ll see in Chapter 11, to create custom Internet radio feeds with tight se-
gues from one track to the next, along with prerecorded announcements and
other material between songs—all automatically, scalably, and in real time.
Conclusion
The 12-inch long-playing (LP) record has enjoyed the longest duration as the
dominant recorded music format in the market—more than the 78s and wax
cylinders that preceded it, and more than the cassettes and CDs that came
later. Vinyl ruled for over 30 years, compared with just over 20 for both the
78 and the CD.
Why is that? It’s because the LP put a combination of quality attributes
and unique features together into a package with a value to consumers that
remains unsurpassed, and almost all LP players are compatible with its close
cousin, the 45 rpm single. This helps explain why vinyl has come back from
oblivion. Vinyl brings in more than $1 billion to the music industry today,
and consumer research from MusicWatch has found that more than 17 mil-
lion Americans buy vinyl—fully a third as many as those who purchased
digital downloads at the height of their popularity.i As we’ll explain in this
chapter, we’re unlikely to see a revival in other physical music formats that
rivals that of vinyl.
Attempts to create phonographic discs that played longer than 78s date
back to 1904.ii In 1931, RCA Victor released a 12-inch disc that played for
ten minutes a side at 33 1/3 rpm. But the product suffered from various
problems, including poor durability: discs could only be played a few times
before they wore out. Lack of durability and the Great Depression—players
were expensive—killed it off in 1933. RCA Victor’s arch-rival Columbia
Records, which in 1938 became a division of the CBS broadcasting company
that it co-founded in 1927, tried releasing 10-inch discs at 33 1/3 rpm around
the same time and failed for similar reasons.iii
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0004
98 Key Changes
these for $30 (later reduced to $10) or gave them away with the purchase
of several of their LPs.vi And they offered “starter libraries” of music with
purchases of their more expensive record players.vii
Finally, Columbia offered to license the technology to both other record
labels and electronics manufacturers, to forestall format wars and increase
consumer confidence in the new format. But the tactic didn’t quite work.
While other electronics companies had record players that played LPs avail-
able at the launch in 1948, and almost every label took the license from
Columbia, one didn’t: RCA Victor. RCA expressed interest initially but de-
cided to put out its own competing vinyl microgroove format: a 7-inch disc
that played at 45 rpm and had a bigger spindle hole.
RCA Victor launched the 7-inch 45 in March 1949. RCA had actually
started making 45s in December 1948; more than 100 titles were avail-
able to the public when the format launched.viii The first batch of 45s cov-
ered myriad genres, including R&B (Arthur “Big Boy” Crudup’s “That’s All
Right, Mama”), country (Eddie Arnold’s “Texarkana Baby”), Yiddish (“A
Klein Melamedl (The Little Teacher)” by Cantor Saul Meisels), and children’s
records (“Pee-Wee the Piccolo” by Paul Wing). Initial 45s were pressed on
colored vinyl until labels switched to black for cost reasons.
RCA initially positioned 45s as direct competitors to LPs, releasing clas-
sical music works on multi-45 sets. The 7-inch discs had sound quality
that was actually superior to LPs, thanks to the higher speed and other
factors. But their capacity was similar to that of 78s. RCA also sold a record
changer for 45s, so that users could listen to the equivalent of an LP side
without getting up—and if they were willing to put up with fadeouts and
record-changing silences during classical concertos, symphonies, and other
longer works.
RCA’s creation of a format war caused confusion in the market, which put
the brakes on sales. The record industry’s revenue plummeted from $204 mil-
lion in 1947 to $158 million in 1949—a drop of almost one-quarter in two
years, the worst industry downturn since the Great Depression. Finally, in
1950, RCA gave in and licensed the LP technology from Columbia.ix
Of course, all labels eventually embraced both microgroove formats. The
vast majority of record players could play all three speeds (33, 45, and 78)3
and came with adapters so that 45s could fit the thin spindles for LPs and 78s.
RCA designed a plastic adapter, known as the “spider,” that you could buy
cheaply at the record store and attach to 45s individually when you lost the
adapter that came with your turntable (see Figure 4.1).
100 Key Changes
Figure 4.1 Adapters for playing 45-rpm singles on turntables: A metal adapter
to use with your turntable (L) and a plastic “spider” meant to be attached to the
records themselves (R). Bill Rosenblatt.
Unlike most format wars, which tend to end up with one (or neither)
format succeeding and the other (or both) failing, each of the two vinyl
formats eventually found its own niche: the 45 wasn’t successful for clas-
sical music, but it became popular for pop music singles a few years later. Bill
Haley’s “Rock Around the Clock” was the format’s breakout moment when it
sold 3 million singles in 1955.
RCA also introduced the extended play (EP) format in 1952. The EP used
the same form factor as the 45 but with narrower grooves and more com-
pressed sound levels to achieve a capacity of 7.5 minutes per side, allowing
for two songs on each side. EPs became more popular in Europe than in
America, such as for Beatles records that are now valuable collectors’ items.4
The accelerated release of high-quality titles on vinyl by many labels
after 1948 was enabled by another technology developed around the same
time: tape recording. As we’ll see in Chapter 5, recording onto acetate tape
coated with magnetic iron oxide was first developed in Germany in the 1930s
and brought over to the United States after World War II.
Tape recorders had several major advantages over the disc-cutting tech-
nology then in use in recording studios. Unlike discs, tapes could have
Vinyl 101
arbitrarily long recording times, limited only by the sizes of reels and by the
recording speed, which, as always, traded off duration against sound quality.
They could be edited easily with a razor blade and adhesive tape. They eventu-
ally had multiple tracks, which enabled multiple audio sources (instruments
and vocals) to be recorded separately, yet in sync, and mixed down later to
create a final recording.
Finally, tape recording setups cost much less than disc lathe setups. This
led to an expansion of the recording industry as many entrepreneurs set
up new labels. The list of labels that arose in the late 1940s through early
1950s includes Jac Holzman’s Elektra, Ahmet Ertegun and Herb Abramson’s
Atlantic, Vivian Carter and James Bracken’s Vee-Jay, Sam Phillips’s Sun,
Leonard and Phil Chess’s Chess, and Bob Weinstock’s Prestige. When
Holzman started Elektra as a folk music label, he bought a Magnecord tape
deck and a single Telefunken microphone for a few hundred dollars, built
other equipment himself, and sent his tapes out to major labels for mastering
and pressing. The major labels at that time (Columbia, RCA Victor, Decca,
and Capitol) were eager to rent out their pressing plants to upstart labels: the
shift from 78s to LPs meant that they had far fewer records to press and there-
fore had excess capacity.
The final piece of the format standardization puzzle fell into place in the
mid-1950s: audio equalization. Record-cutting techniques included pro-
cessing of audio signals that attenuated lower frequencies and boosted
higher frequencies. This was done so that groove width could be reduced,
leading to increased recording time, less hiss, lower bass distortion, and less
risk of damage to the vinyl as the stylus traversed it. Equalization was the
process of making those adjustments electronically at recording time and
compensating for them on playback. The scheme for attenuating or boosting
the audio signal by different amounts at different frequencies is called an
equalization curve.
Different record companies originally used different equalization curves.
To get the best sound quality, a consumer had to have playback equipment
that switched among several curves and had to select the one used by the
label of the record being played. For example, the Pilot HF-42 receiver from
1957,5 shown in Figure 4.2, had a knob for choosing among five different
equalization curves. To solve this problem, the industry formed a trade as-
sociation, the Recording Industry Association of America (RIAA), in 1952.
The RIAA designated a standard in 1954 based on RCA Victor’s curve;
this became known as “RIAA equalization.” The RIAA curve became an
102 Key Changes
Figure 4.2 The Pilot HF-42 mono high fidelity receiver from 1957 featured five
different equalization settings.
international standard by 1955,x and it was used for both LPs and 45s. Phono
amplifiers made after the late 1950s supported only RIAA equalization and
no longer had switches for selecting curves.
The last significant and successful innovation in vinyl technology was
stereo. The move to stereo began soon after RIAA equalization became
standard. As we’ll see in Chapter 5, tape was the first consumer music format
to go stereo in the mid-1950s. This led to a drive to implement it in vinyl.
Recording tape in stereo was merely a matter of designing recording and
playback heads that divided the tape up into tracks for left and right stereo
channels. Cutting and playing records in stereo was a bigger technical
challenge.
Yet it only took a few years for the industry to produce a standard stereo
vinyl system. That’s because a few different systems for cutting stereo discs
had been developed, and it was only a matter of choosing one of them. One
was the so-called binaural system, which used cutting heads spaced 1 11/16
(1.6875) inches apart, so that the left channel was recorded on the outer band
and the right channel was recorded on the inner band.xi This required either
two separate standard tonearms or special binaural tonearms, such as the
model shown in Figure 4.3, and it cut in half the amount of recording time
per side. The scheme was rejected as impractical.
Other systems used a single cutting head to record a stereo signal into a
single groove, requiring only one tonearm and cartridge for playback. One
such system, the Vertical-Lateral system, used up-and-down (vertical) motions
for one channel and side-to-side (lateral) motions for the other. It suffered
from differences in sound quality and crosstalk between the two channels, but
it enabled easier compatibility with mono recordings. Another was the 45/45
Figure 4.3 The Livingston Electronic Corp. two-headed tonearm from 1951 for
playing binaural records.
104 Key Changes
system, which was invented by Alan Blumlein at EMI in Britain in 1931 and
developed into a commercial record-cutting system at Westrex, a division of
AT&T. The system worked by making cuts at opposing 45-degree angles (one
at 45 degrees from the top and the other 135 degrees from the top).xii With
its superior sound quality, the Westrex 45/45 system won out and became
standard in 1957.xiii By the following year, most labels had stereo recordings
available.xiv
There was only one more attempt at major innovation in vinyl recording
technology: quadraphonic, or four-channel, sound. It was one of the most
notable technology market failures of the twentieth century.
Quad, introduced in the early 1970s, was an attempt to capture the three-
dimensional ambiance of musical performances by adding a pair of rear
speakers to the usual stereo left/right pair. There was a lot of hype around
quad, but its launch was a chaotic mess compared to the relatively smooth
launch of the LP in 1948.
The biggest problem was that there was no single standard: there were
three competing quad systems for vinyl in the US market.6 Two of them
used standard tonearms and cartridges, while the third promised the best
four-channel audio experience but required consumers to buy new equip-
ment. None of the three systems was compatible with the others, and
different labels released records in different quad systems. Some stereo
amplifiers and receivers could play all three formats, while others could
only play a subset.
It was a train wreck. Consumers were alienated, confused, and not seduced
enough by the sound experience or by gimmicks in playback equipment—
such as the joystick in the receiver shown in Figure 4.4—to lay out the money
for new receivers, speakers, and possibly turntables. Quad evaporated by
1977–1978. Of course, multichannel audio lives on today in the form of
surround-sound systems for home theater;7 but few serious audiophiles used
quad equipment or discs for music listening.
Otherwise, the technologies used to create vinyl records—Columbia’s mi-
crogroove LP, RCA’s 7-inch 45, RIAA equalization, and 45/45 stereo—were
established by the late 1950s and have remained largely the same since. Apart
from quad, other variations over the years have been minor, such as 12-inch
45 rpm singles or EPs with “hotter” sound quality released to radio stations
and club DJs, and various types of high-quality manufacturing processes and
vinyl substances for audiophiles.8
Vinyl 105
The industry’s transition to LPs and 45s caused a dramatic expansion of dis-
tribution channels to reach wider audiences beyond music aficionados and
electronics hobbyists to the general public. One could almost say that the LP did
for recorded music what big-box retail did for all sorts of consumer products
starting in the early 1960s (when Target, Kmart, and Walmart all started).
The first new outlets were mass-market retail stores beyond the individually-
owned record shops that sold 78s. Vinyl began to be sold in variety stores
(a.k.a. 5 & 10 cent stores), drugstores, and supermarkets in the early 1950s.
Independent subdistributors, or “rack jobbers,” would select titles to fill racks
at those stores and would handle payments and returns to record labels.
The first rack jobber to distribute records was Elliot Wexler’s Music
Merchants in Philadelphia in 1952; Music Merchants had been a distributor
of sheet music. Next was David Handleman’s Handleman Company, a dis-
tributor of drugstore items in Detroit. Handleman’s business took off; the
company kept going as a distributor of CDs as well as videos and computer
software until it shut down in 2008.xv
106 Key Changes
The labels soon found themselves at odds with rack jobbers, given their
lack of cooperation in pushing the products that the labels wanted them to
and the liberal return policies they negotiated on records that didn’t sell.9,xvii
Label executives would assail rack jobbers for their low risk tolerance for sel-
ling titles that weren’t big hits by established artists.
One tactic that labels used to regain some measure of control was
“cutouts”: records that weren’t selling well and that labels would delete from
their catalogs. As hard as it may be to grasp today, with digital music serv-
ices building up vast catalogs of “every record ever recorded,”10 record labels
started doing this routinely in the early 1970s. They would sometimes liter-
ally cut notches into the jackets of deleted albums and ship them into retail
channels at deep discounts. This caused all sorts of logistical and book-
keeping headaches throughout the supply chain, not to mention the stigma
that artists would feel when their albums were deleted.xviii Labels’ other
supply-chain tactics included throwing in titles by lesser-known artists for
free along with supplies of hit records to rack jobbers, in addition to the kinds
of inducements to rack jobbers that radio DJs and program directors were
enjoying, as we saw in Chapter 3.xix
Soon after the rise of rack jobbers came discount record store chains.
Sam Gutowitz started his Sam Goody store shortly after the launch of the
LP as both a retail store on 49th Street in Manhattan and a mail-order
business that took advantage of vinyl’s lighter weight and smaller package
size than 78s. Sam Goody sold records at 30 percent discount from retail
prices, a move that enraged record labels.xx Courts ruled that the labels
were powerless to dictate Goody’s prices, so they tried to stop him through
tactics such as refusing him return privileges and credit. Nevertheless,
Gutowitz expanded to a chain of retail stores in the early 1950s; his busi-
ness grew rapidly and soon accounted for 5 percent of the total record
business.
The second discount chain was Grover Cleveland Sayre’s Musicland, which
opened its first store in Minneapolis in 1955. With their lower prices and big
selection, the two chains expanded the market for vinyl even further, despite
the labels’ protestations. (The two chains would consolidate under the same
ownership in 1978, using the Sam Goody name.)
The next new retail channel for vinyl was mail- order record clubs.
Columbia Records started the Columbia Record Club in 1955; RCA Victor
(partnering with both Reader’s Digest and Book-of-the-Month Club) and
Capitol followed with their own clubs soon thereafter.xxi
108 Key Changes
Labels had several reasons to start record clubs. The main one was to com-
pete with emerging retail channels. Other entities besides labels had started
record clubs by mail, such as the Music Appreciation Club, an offshoot of
the Book-of-the-Month Club.xxii The clubs enabled labels to expand their re-
tail distribution to rural areas and compete with the discounters’ mail-order
businesses. Otherwise, in a sense, labels set up record clubs to compete with
their own existing wholesale channels—and to be able to charge full retail
price while eliminating middlemen.
Record clubs became famous for their “twelve records for a penny”
deals: you filled out a postage-paid postcard (which probably fell out of a
magazine), taped a penny to it, and put it in a mailbox; a box of LPs arrived
at your front door a couple of weeks later. But each month thereafter, for the
next couple of years, you got another record in the mail, for which you were
billed at full retail price, unless you wanted to pay for postage to return it.
This wasn’t the labels’ original idea; Book-of-the-Month Club had started it
in the 1920s.xxiii This practice, known as “negative option billing,” became
controversial and was outlawed in some countries outside the United States.
Each of the labels initially only made their own records available through
the clubs, but the labels began—grudgingly—cross-licensing to one another’s
clubs in 1958 to stave off third-party competition. Cross-licensing increased
through the 1960s until almost every club offered records from the other
clubs’ labels—with the notable exception that BMG Music Service (formerly
RCA Victor Record Club) and Columbia House (formerly Columbia Record
Club) did not cross-license their label siblings’ music to each other.11
Record clubs were moderate successes for the labels for a long time, until
the internet came along to replace them. By 1955, there were a dozen clubs
with a million members among them.xxiv By the early 1960s, record club
sales accounted for 15 percent of total industry sales volumes, compared to
27.4 percent through NARM-affiliated rack jobbers. By the mid-1970s, the
clubs’ membership had grown to 3 million, with their demographic skewing
older than typical retail buyers.xxv RCA Victor and Columbia were the two
largest clubs by far—large enough that the Federal Trade Commission, at the
behest of the other major labels, started to investigate them for anticompeti-
tive business practices.xxvi
In general, the expansion of record sales to the mass retail market led to a
profusion of channels that filled the supply chain with various convoluted and
contentious relationships—often to the benefit of consumers. Discounters
drove prices down and led labels to seek ways to circumvent them; this led
Vinyl 109
Figure 4.5 The Seeburg HF100G jukebox, from 1957, at the Silverball Retro
Arcade in Asbury Park, NJ. It was one of the first jukeboxes to play 45-rpm
records. Bill Rosenblatt.
sales growth slowed down in 2022 compared with their rapid rise through
2021,xxxviii that’s still much bigger than anyone looking at the market in the
late 2000s would have predicted.
to $730 (about $7,300 today) for a phono system with preamp, power amp,
a turntable, tonearm and cartridge, speakers, and speaker cabinets. Adding
an AM/FM tuner would have brought the price up over $900 ($9,000 today).
With hi-fi components, there was more emphasis on sound quality than
on furniture style, more knobs and dials, lighter tonearms, and heavier
platters, as well as escalating price tags. Aficionados attended hi-fi trade
shows and constructed listening rooms in their houses with optimal place-
ment of speakers and chairs, and acoustic wall treatments. “Demonstration
records” were released featuring sound recordings and effects that showed off
hi-fi systems’ capabilities. High Fidelity magazine debuted in 1951, followed
by Audio (formerly Audio Engineering) in 1954 and HiFi and Music Review in
1958 (later Stereo Review). As explained in the previous chapter, transistors
came along in the mid-1950s; they eventually replaced vacuum tubes in hi-fi
equipment, although a faction of audiophiles insisted—and still insist to this
day—that tubes sound better.
As Figure 4.6 shows, once the record industry recovered from the LP/45
format war in the mid-1950s, the hi-fi component industry was about half as
large as recorded music by revenue and grew at about the same rate. It con-
tinued to grow at more than 20 percent per year into the 1960s. The audio
component industry continued in this vein until 1980, when it peaked at $1.8
billion,xli and then started to decline.
$450
$400
$350
$300
$250
$200
$150
$100
$50
$–
1950 1951 1952 1953 1954 1955 1956 1957
Records Hi-Fi Components
Figure 4.6 Annual revenues from recorded music and hi-fi components,
millions of dollars, 1950–1957. Data sources: RIAA (recorded music), Institute
of High Fidelity Manufacturers (hi-fi components). Bill Rosenblatt.
114 Key Changes
Figure 4.7 GE Wildcat portable stereo record player, circa 1970. CC BY 2.0,
Bradley Stemke.
didn’t sound that great. But nowadays, encoding techniques have improved,
bitrates have risen, and audiophile-grade digital–analog converters have
been introduced. Digital data storage has become cheap enough so that users
who care about sound quality can store large collections of digital music in
“lossless” codecs such as FLAC on garden-variety PCs. In other words, dig-
ital files can sound superb today. And most young fans’ playback equipment
or earbuds aren’t good enough to tell the difference anyway. Finally, much
recent vinyl was recorded digitally in the first place and had to be converted
back to analog to make LPs.
The factor that most likely explains the renewed interest in vinyl is that
it restores a sense of ownership to music fans. In the 2016 book The End of
Ownership, two legal scholars, Aaron Perzanowski of Case Western Reserve
and Jason Schultz of NYU, explain that digital files bought online aren’t really
“owned” in the legal sense, and they advocate for reforms to establish own-
ership of digital content in law. They present consumer research suggesting
that people expect ownership rights, even though they might not actually get
them.xlii
116 Key Changes
In fact, vinyl is far more popular than CDs in the used market, even though
LPs’ sound quality deteriorates with use a lot faster than CDs. Discogs, for
example, lists over 46 million vinyl items in its marketplace at this time of
writing, compared to 21 million CDs, and the average price of a used LP
is considerably higher than the average price of a used CD. It also bears
mentioning that many fans buy vinyl without even necessarily intending to
Vinyl 117
play it: one study showed that only 50 percent of vinyl buyers in 2022 actually
owned record players.xliv
Finally, it’s worth noting that while the vinyl renaissance was originally
mostly confined to classic rock and a bit of jazz, recent Discogs monthly
sales charts have seen more current pop, R&B, and hip-hop titles. This also
suggests that vinyl’s renewed appeal is broadening beyond hipsters. Today’s
vinyl fans have already picked up their copies of Dark Side of the Moon,
Rumours, and Thriller;16 now they’re ready for something new, and record
labels are meeting that demand with new releases.17 And the interest in vinyl
has reached the mainstream: in 2022, Taylor Swift’s Midnights sold 945,000
copies on vinyl, almost enough to go Platinum on vinyl sales alone.18,xlv
Blonde on Blonde was released in May of that year, followed by “The Return
of the Son of Monster Magnet” on Frank Zappa’s Freak Out! (another double
LP) a month later. But these were only 11 and 12 minutes long respectively.
The first true side-long rock track was “Revelation,” a 19-minute jam on
Da Capo by the LA psychedelic band Love, which followed in November.
The first rock album with a track taking up both sides of an LP was an acid-
soaked jam on the self-titled debut album by an obscure Southern California
band called The Beat of the Earth, on the tiny Radish label, in 1967.
As the 1960s rolled into the 1970s, artists and record producers started
focusing on constructing albums with the right songs in the right running
order. Album-side sequencing began to take on a life of its own as a producer’s
art form. There were few iron-clad rules, though many 1970s rock albums
had up-tempo rockers opening each side, followed by ballads. In addition,
producers knew that the last tracks on each side couldn’t have heavy bass,
because the linear speed of a turntable decreases as the needle moves toward
the center, meaning that the frequency response range would be limited and
too much bass could cause the stylus to be knocked out of the groove.20 One
label head noted that it was common to include a long track on every LP “so
that the radio DJ could get laid or go to the bathroom.”
And then there was progressive rock, where bands would record heavily
composed, suite-like tracks of album-side length. The first of these was the
title suite of Ars Longa Vita Brevis by The Nice, featuring future ELP key-
board player Keith Emerson, in 1968. After that, album-side tracks became
de rigueur in the early 1970s among prog-rock bands like Yes (“Close to the
Edge”), ELP (“Tarkus”), and Pink Floyd (“Echoes”). The high point of the
genre—or nadir, depending on your point of view—was Yes’s 1973 Tales from
Topographic Oceans, a double LP consisting of one track per side.
Finally, the use of the LP format in rock led to an explosion of creativity
in album cover designs. One important step was to print song lyrics on the
album cover; the first album to do that was the Beatles’ Sgt. Pepper. The deci-
sion to print lyrics on album covers may seem obvious in hindsight, but it re-
quired overcoming objections from music publishers, who feared that doing
so would cut into sheet music sales.
Another important step in the evolution of album covers was to hire
“name” artists to do cover designs. The Velvet Underground and Nico from
1967, with the “banana” cover by Andy Warhol, is often cited as the first ex-
ample of that (followed by his “zipper” cover for the Rolling Stones’ Sticky
Fingers in 1971). But Warhol had been designing album covers since 1949;
120 Key Changes
he had designed over a dozen covers, mostly of jazz albums, before doing the
Velvet Underground’s. The earliest instance of an already-famous artist doing
an album cover is Salvador Dali’s 1955 cover of Jackie Gleason’s Lonesome
Echo. Other early examples include ELP’s Brain Salad Surgery (H. R. Giger)
and Patti Smith’s Horses (Robert Mapplethorpe), both from the mid-1970s.
The converse also happened, as a few artists became famous for their rock
album cover designs. The pioneers here were Storm Thorgerson and Aubrey
Powell of the British design firm Hipgnosis,21 whose cover for Pink Floyd’s A
Saucerful of Secrets in 1968 led to hundreds of other covers for dozens of mu-
sical artists and many other projects beyond music. Roger Dean also started
designing album covers in 1968; his iconic fantasy cover designs for Yes from
1971 onward led him to design work for dozens of album covers as well as
video games.
As Chuck D of Public Enemy once said, “If you don’t own the master, the
master owns you.” “Master rights” are rights to sound recordings, and as
we’ll see later in this chapter, sound recordings didn’t carry copyrights until
1972—more than 20 years into the vinyl era. Yet it was almost unheard of for
artists during the vinyl era to own their master rights; they assigned them to
record labels in recording contracts. The reason was simple: when you made
a deal with a label, the label advanced you the money to make the record and
most likely owned the studio and the manufacturing plant. Later on, artists
might record in independent studios, but the label still paid the artist an ad-
vance to cover the studio costs.
In other words, artists didn’t have much leverage against their labels unless
and until they became big stars—and as we saw in the previous chapter, there
were very few big stars in recorded music until at least the late 1950s. Labels
typically signed artists to contracts that required them to deliver a certain
number of albums over a period of years.
Yet certain artists’ leverage increased as they became famous. In the late
1960s, the major labels found a new way to keep their stars happy: give them
subsidiary labels of their own. These became known as vanity labels.22 The
vanity label would typically do A&R and other creative work, while the parent
label would handle manufacturing and distribution. The first of these were
the Beach Boys’ short-lived Brother, created by Columbia in 1966; Bizarre
Vinyl 121
and Straight, created by MGM for Frank Zappa in 1967; and Apple, spun off
from Capitol for the Beatles in 1968. This became more common practice
as time went on. While most vanity labels were dedicated to the artists who
founded them, a few of the early ones had their own successful signings of
new artists, such as Straight’s signing of Alice Cooper and Captain Beefheart,
Apple’s signing of Badfinger and James Taylor (plus all four Beatles as solo
artists), and Led Zeppelin’s Swan Song label (under Atlantic) signing Bad
Company and Dave Edmunds. Vanity labels became even more successful in
the 1990s, such as Maverick, which Madonna started within Warner Music
Group in 1992, and Dr. Dre’s Aftermath in 1996 within Universal/Interscope.
Otherwise, little happened during the vinyl era that would affect the
industry’s royalty structure for labels and artists. Two changes to the
industry’s royalty structure that took place in the 1970s involved public
performances of sound recordings, but they resulted in new royalties for
songwriters and music publishers only. As we’ll see later in this chapter, these
were for jukeboxes and cable television.
Apart from these changes to the royalty structure of the music industry,
the company structure of the industry—specifically, the nature of the rela-
tionship between record labels and music publishers—changed substan-
tially during the vinyl era. Put simply, labels got bigger than publishers and
proceeded to eat up much of the music publishing industry.
Thanks to pop music and vinyl records as mass- market consumer
products, recorded music became a major American industry during the
vinyl period. Total annual revenue had grown from $157 million in 1949,xlvi
the year after the introduction of the LP, to $2.2 billion by 1974,xlvii 25 years
later. Even when adjusted for inflation, this was almost a sevenfold increase
in sales.
It is likely that the recorded music industry had grown bigger than the
music publishing industry by gross revenue by the early 1970s.23 Record
labels had taken over from music publishers in distributing and promoting
music by then. As RIAA President Stanley Gortikov said at a congressional
hearing on copyright reform in 1975, music publishers had become “heavily
administrative and clerical; they are largely service entities, conduits for
the processing of income and paper transactions. They don’t promote as
they used to. They don’t advertise as they used to. They don’t help create de-
mand as they used to. They don’t employ field representatives as they used to.
These promotional functions necessarily have been taken over by recording
companies.”xlviii
122 Key Changes
The major labels started buying up music publishers in the 1960s and
ended up owning many of the major ones. A few examples: Philips (later
Polygram) bought Chappell publishing in 1968; Polygram sold it to private
investors, who in turn sold it to Warner Bros. in 1987, creating Warner-
Chappell Music, the publishing arm of Warner Music Group. MCA Records
made its first of several publishing acquisitions with Leeds Music in 1964;
MCA went on to merge with Polygram to become Universal Music Group,
of which Universal Music Publishing—the largest music publisher in the
world—is a subsidiary. Sony Music bought ATV Music from Michael Jackson
in a multi-year process that began in 1995; ATV Music included the Beatles’
song catalog, which Jackson had bought in 1984. The resulting Sony/ATV
Music, the publishing arm of Sony Music Entertainment, was renamed Sony
Music Publishing in February 2021. In fact, the big three major labels’ pub-
lishing businesses account for over half of total music publishing industry
revenue today.xlix
One of the reasons why record labels became bigger businesses than music
publishers by the 1970s may be that they finally got copyrights on their sound
recordings—although the economic implications of music recordings’ lack
of federal copyright protection until 1972 have been the subject of intense
debate for decades. Nowadays it seems obvious that music recordings are
copyrighted, but that was not at all the case when vinyl formats came into
being. And it’s not entirely the case today either: US copyright law still confers
fewer rights on sound recordings than it does on musical compositions.
The better part of a century elapsed between the first audio recordings and
copyright being established in sound recordings in US federal law. As we saw
in Chapter 2, arguments over whether sound recordings should be eligible
for copyright date back to the days of cylinders and piano rolls. During the
period between the Copyright Act of 1909 and the 1960s, several litigations
were brought, and more than a dozen separate bills were put forward in the
House and the Senate, all in hopes of establishing copyright protection for
sound recordings. None of these efforts succeeded.l
Myriad reasons were given for why sound recordings shouldn’t get
copyrights. In addition to those we encountered in Chapter 2, those reasons
included that sound recordings weren’t sufficiently creative compared to
Vinyl 123
carrying. This changed when cable operators began to offer their own pro-
gramming; the 1976 Copyright Act included a compulsory license for mu-
sical compositions performances on cable TV. As we’ll see in Chapter 6,
music licensing for cable television would take on a whole new meaning a
few years later.
The second change involved jukeboxes. Jukebox operators bought
records for the machines but didn’t pay royalties for their use. The PROs, the
Copyright Office, and others had long argued that jukebox plays were public
performances that enriched jukebox operators—which collectively took in
$447 million in 1959liv—but paid nothing to the copyright holders. Those
arguments failed for decades.lv
Congressman Emanuel Celler, from music-industry-heavy New York,
started an effort to end the “jukebox exemption” and establish a levy on
jukeboxes in 1952. Yet it wasn’t until 1978 that jukebox operators had to pay
performance royalties. At first, they paid $8 annually per machine for com-
pulsory blanket licenses—flat one-time fees that covered all musical reper-
toire, administered by the Copyright Office. The Copyright Office collected
over $100 million from more than 140,000 jukeboxes during the license’s first
full year of operation.lvi The fee was raised to $50 by 1984. In 1989, the com-
pulsory license changed from one with flat rates to one that required jukebox
operators to negotiate rates with each PRO. Jukebox operators set up the
Jukebox Licensing Office in 1990 to handle these negotiations,lvii analogously
to the Radio Music Licensing Committee that we saw in Chapter 3.
Yet this was largely a pointless exercise: by the time the levy was enacted,
the jukebox industry had gone into a nosedive. Jukebox manufacturing had
plummeted from 700,000 to 25,000 annually, and by 1980, there were fewer
than 5,000 jukebox operators. The soda fountains and chrome dinettes of the
1950s had given way to jukebox-free fast-food chains.
Conclusion
The collapse of the jukebox market didn’t hurt the music industry. On the
contrary, it’s best seen as part of a migration of music consumption to new
and bigger use cases enabled by new devices. Vinyl helped recorded music
become a multibillion-dollar industry, one that by the 1970s was led by large
companies looking for growth. Growth would come from new ways to con-
sume music, and those weren’t going to come from bars and restaurants. Fans
Vinyl 127
were happy for opportunities to listen to the music they wanted in the settings
they wanted, whether those were at home, in the car, inside, outside, while
traveling, or while hanging out with friends; and electronics companies were
happy to supply gadgets and media formats to meet those occasions. As we’ll
see in the next chapter, the next phase of the music industry was one in which
it used a format as old as vinyl to make that happen.
5
Tapes
Rhymin’ and Stealin’—Beastie Boys
When we think of inventions made by and for the Nazis and exported
to the United States after World War II, we usually think of rockets or
Volkswagens. But another one was arguably just as important: magnetic tape.
Two companies that would be associated with Nazi atrocities developed the
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0005
Tapes 129
technology, and its initial use was as a Nazi propaganda tool to record and
broadcast Hitler’s speeches over the German state broadcasting network.
Fritz Pfleumer, an Austrian-born engineer living in the German city of
Dresden, invented a scheme for coating paper tape with magnetic particles
in 1928 as a cheaper alternative to wire recording. He sold the rights to his
invention to BASF, a division of the German chemical company IG Farben.
The electric company AEG then produced the first commercially available
magnetic tape recording device, the Magnetophon (see Figure 5.1), and
demonstrated it at the Internationale Funkausstellung (International Radio
Exhibition) in Berlin in 1935.i
The first known music recording on magnetic tape was made in 1936.
A Magnetophon, with tape provided by BASF, was used to record a concert
by the London Philharmonic Orchestra under the direction of Sir Thomas
Beecham at the BASF-Feierabendhaus in Ludwigshafen, while the orchestra
was on tour in Germany. Yet the sound quality of tape recordings made in the
mid-1930s was markedly inferior to existing wire and disc technologies.
Three developments in magnetic tape technology through World War II
made it more suitable for recording music.ii One was discovered by the RRG
Figure 5.1 AEG Magnetophon K4 reel-to-reel tape recorder, circa 1940. YLE
Archives (public domain).
130 Key Changes
tapes could play easily on mono tape machines: the single playback head
simply read both tracks at once. (And vice versa: mono tapes could readily
play on stereo machines.) For a while it was even thought that tape was the
only practical stereo medium.viii Of course, dividing up recording tape into
multiple tracks took on a life of its own in studios as multitrack recording
came into being in the late 1950s.ix
The first prerecorded stereo tapes came in 1953, although the major labels
didn’t start to issue titles on tape in stereo until 1957, the year when vinyl
records went stereo.x A standard format for commercially recorded tapes
emerged in 1959: the tape was a quarter-inch wide and divided into four
tracks, two left/right stereo pairs, analogous to the two sides of an LP.xi
The sound quality of tape is proportional to the area of tape that passes a
playback head every second, which is a function of the width of the tape and
the recording speed. The standard for commercial releases called for tape to
be recorded at 7.5 inches per second (IPS). This was slower than the 15 or 30
IPS speeds typically used for mastering in recording studios, where the track
widths were usually double that of consumer prerecorded tapes.2 This meant
that the amount of standard prerecorded tape passing a playback head was
as little as one-eighth the area of the master tape. Sound quality is also pro-
portional to tape speed in the sense that slower speeds are more susceptible
to speed variations that cause distortions known as wow and flutter. Still, 7.5
IPS four-track stereo was considered sufficient for high-fidelity sound when
played through quality home equipment. As the 1960s progressed, lower-
cost prerecorded tapes with lesser audio quality became available at 3.75 IPS
(using half as much tape) as well.
Magnetic tape also made recording more accessible to consumers at home.
Disc-recording equipment was expensive, designed to be integrated into re-
cording studios by professionals, and rarely used in homes. Magnetic tape
recorders in the early 1950s came in complete packages with microphones,
amps, and speakers included, at half the cost of disc setups. Disc recording
equipment had disappeared from electronics retailers’ catalogs by the end
of the 1950s, while tape recording equipment got cheaper and models
proliferated.
As far as sound quality was concerned, by the mid-1950s prerecorded
open-reel tape was considered to have slightly better sound quality than
vinyl because prerecorded tapes were closer copies of original master tapes.xii
Tape also wasn’t susceptible to scratches and dust as vinyl was, so it tended to
hold its sound quality better over time.
132 Key Changes
Figure 5.2 A Fidelipac tape cartridge used in radio stations (L) and a
Stereo-8 a.k.a. 8-track tape cartridge (R). Bill Rosenblatt; Fidelipac cartridge
courtesy WPRB.
four tracks configured as two stereo pairs. It ran at 3.75 IPS and had a total
playing time of 40 minutes.xvi
The Stereo-Pak system debuted in 1962, but it didn’t last very long as a
commercial music format. It was eclipsed a few years later by another car-
tridge system, which was designed by Bill Lear of Lear Jet fame. Lear had gone
into business distributing Muntz’s Stereo-Pak players so that he could install
them in his private jets. He developed relationships with Ampex, Motorola,
Ford Motor Company, General Motors, and RCA Victor to create a new ver-
sion called Stereo-8. Stereo-8 had four stereo pairs (eight tracks) instead of
two, increasing playing time to a maximum of 80 minutes (20 minutes times
four stereo programs), though some releases used only 10–11 minutes per
track (for a total of 40–44 minutes). Stereo-8 became known simply as 8-
track. As Figure 5.2 shows, the 8-track differed from the Fidelipac in certain
ways, which enabled Lear to avoid having to license patents from Cousino
and Eashxvii while also leading to cheaper and smaller players5 and even
more ease of use.6
Ford was the first automaker to install 8-track players (made by Motorola)
into its cars; it did so in late 1965 as a factory-installed option for the 1966
model year. Lear also produced aftermarket auto as well as home and port-
able 8-track players under the Lear Jet brand. Third-party electronics makers
followed with their own 8-track players the next year. Home component 8-
track players (designed to be connected to amps or receivers) came in 1968
with the introduction of the TC-8 from Sony.xviii
For the first couple of years, units were available that played both 4-track
Stereo-Pak and 8-track cartridges. But 8-tracks took over in popularity, be-
cause Lear made more arrangements with automakers, and because more
music could fit onto 8-tracks. By 1970, Muntz had lost the format war and
abandoned the Stereo-Pak.7
The 8-track format had its limitations. One was inferior sound quality.
Like open-reel and carts, 8-tracks also used quarter-inch tape, but they
had twice as many tracks and (like Stereo-Paks) ran at the slower 3.75 IPS
speed. Cramming eight tracks onto quarter-inch tape not only meant less
tape area per track, it also made the format susceptible to head misalign-
ment problems: the playback head had to move up and down to switch
tracks, and if it wasn’t aligned exactly correctly, you’d hear an adjacent track
in the background. Yet car audio systems of the late 1960s through 1970s
never had great sound quality, so these issues weren’t so important for most
consumers.
Tapes 135
Another limitation was track length, especially when the shorter 10–
11-minute-per-track version was used. Songs would be chopped in half or
reordered to fit. This was a particular problem for classical music, as well as
for the side-long pop music tracks that were emerging in the 1970s. For ex-
ample, Kraftwerk’s “Autobahn” and Todd Rundgren’s “A Treatise on Cosmic
Fire” were each sliced up into three tracks on the 8-track versions of the
albums Autobahn and Initiation.
The third problem with 8-tracks was that they didn’t provide random
access—the ability to start at specific points in a recording. Whereas it was at
least possible to fast forward or rewind open-reel tape and navigate to values
on tape decks’ counters, 8-track players had nothing but a button to push to
go to the next track. This made sense in the car (like pushbuttons on radios)
but not in the home. In most cases, the next track in an 8-track tape was the
next album side.
These limitations added up to ensure that 8-track was never a very popular
music format anywhere other than the automobile.8
The next tape format to come along was by far the most successful of
all: the Compact Cassette. It was invented at Philips in the Netherlands and
introduced at the Internationale Funkausstellung Berlin in 1963; introduc-
tion in the United States under the Norelco brand came the following year.
The designer of the cassette was Lou Ottens, a Philips engineer who had
designed Philips’s first battery-operated portable reel-to-reel tape recorder
a few years earlier. He based the design on a twin-sprocket tape cartridge
that RCA had released (to little success) in 1958 and shrank the size so that
it would fit in his jacket pocket. (Ottens would go on to lead the Philips team
that worked with Sony in designing the CD.)xix
Philips quickly found itself in a format war with the German electronics
company Grundig, then the largest manufacturer of tape recorders in the
world. In 1965, Grundig introduced the DC-International, another two-
sprocket cartridge design that was about 20 percent larger than Philips’s
Compact Cassette, yet smaller than the earlier RCA design. Philips won the
format war, mainly by agreeing to license its intellectual property royalty-
free to other electronics makers such as Sony and Panasonic.
So, the cassette was introduced two years before the 8-track. But it wasn’t
intended as a recorded music format; it was meant as a convenient pocket-
sized medium for voice recording and dictation. Cassettes are less than half
the size of 8-tracks. They use 0.15-inch tape at a speed of 1.875 IPS (half of 8-
tracks’ 3.75 IPS) in two-track mono or two-stereo-pair configurations. In the
136 Key Changes
Figure 5.3 The Philips EL 3300, the first cassette recorder, from 1963. Used with
permission of Koninklijke Philips N.V.
Tapes 137
durability.xxiii Once again, and just as with 8-track over open-reel, conven-
ience won out over quality: by 1968, cassette recorders were outselling both
open-reel and 8-track machines, a trend that would only accelerate in the
years to come.xxiv
Cassettes’ fast-growing popularity motivated the industry to improve their
sound quality. Two technological innovations came together at the outset of
the 1970s to help make this happen.
The first of these was noise reduction to reduce the hiss that was endemic
to cassettes. A general technique for reducing noise is to compress certain
frequency ranges of the audio signal at recording time and then expand
them again on playback, a process known as compansion (a portmanteau of
compression and expansion). The manufacturing facility uses a compression
scheme to make prerecorded cassettes, and then the user’s player has to have
compatible expansion circuitry to play them back properly.
Ray Dolby brought audio compansion to tape recording. A media tech-
nology prodigy, Dolby worked with Ampex during his undergrad years at
Stanford to develop videotape in the mid-1950s. He started his eponymous
company in 1965; his first product was a compansion-based noise reduc-
tion system that he called Dolby A. It reduced high-frequency noise by 20
dB, which a listener would perceive as a 75 percent reduction. Dolby A came
into wide use in recording studios. But it required high degrees of precision
in matching the compression and expansion processes to work correctly—
precision that would be too expensive to build into low-priced consumer
players.
So Dolby designed a successor technology three years later, which he
named Dolby B. Dolby B traded off less noise reduction in favor of simpler
implementation and listenable sound quality in players that did not have the
decoding circuitry. Dolby B became the de facto standard noise reduction
scheme for cassettes. It provided 10 dB of noise reduction, meaning that it
cut the perceived volume of hiss in half.9
The other innovation was new types of magnetic particles for embedding
onto tape that had better signal retention properties, which would increase
the tape’s frequency response range beyond that of standard ferric oxide.
The first and most successful of these new formulations was chromium di-
oxide (CrO2).10 Du Pont had developed a process for synthesizing CrO2 in
1956. It began licensing the technology to tape makers in the early 1970s.
CrO2, or “chrome,” became the standard for premium-quality cassettes.11
Tape machines needed electronics that were adapted to these new tape
138 Key Changes
recorders,” and found that it held up favorably against several more expen-
sive reel-to-reel decks. The magazine used it as a reference machine for meas-
uring the audio quality of blank cassettes.xxvi The 201 stayed in Advent’s
product line for several years.12 It wasn’t the smash hit product that the
Advent Loudspeaker was—it’s more like a cult classic—but it put the in-
dustry on notice that hi-fi quality from cassettes was possible and reasonably
affordable.
The next step in the “hi-fi-zation” of cassettes came in 1973. The Japanese
company Nakamichi—which ironically had built the balky tape transport for
the Advent 200—took the concept to an extreme with the introduction of the
Nakamichi 1000. The idea of the 1000 was to create a cassette deck with audio
quality to rival the best open-reel tape. In doing so, Nakamichi borrowed
various features from studio-quality reel-to-reel machines13 and added a few
innovations of its own, including some that required users to record cassettes
on the machine to have the best playback quality.14 The 1000 achieved
Nakamichi’s goals, though at a price of four times that of the Advent 201.xxvii
It set a benchmark for the legitimacy of cassettes as a high-quality medium
for music—one that has arguably been exceeded only by the company’s own
subsequent top-of-the-line models. Though few could afford the Nakamichi
100015—it sold for the equivalent of $8,000 in today’s money—audiophiles
knew about it and understood its implications.
The cassette was the last successful consumer tape format for commer-
cial music reproduction. Subsequent successful tape formats were digital in-
stead of analog, and as we’ll see in Chapter 7, they weren’t used as consumer
audio formats.16 Other analog tape formats continued to thrive for nonmusic
applications such as telephone answering machines and children’s toys.
Tapes were the first medium that enabled everyday consumers to make
copies of recorded music products easily. Whether they should make those
copies became a vexed question. Tape eventually became the medium that
shifted the objects of the music industry’s concerns about copyright infringe-
ment from bootleggers to ordinary consumers.
Hi-fi magazines of the 1950s contained articles instructing readers on
how to make open-reel tape copies of their LPs and suggesting they do so
to preserve the vinyl.xxviii Yet as we’ve discussed, open-reel tapes were never
140 Key Changes
very popular, so this wasn’t a major concern to record labels. Home 8-track
recorders and blank tapes became available, but 8-track was never popular
as a home recording format. Nevertheless, around 1967 some record label
executives were known to have tried to block the availability of home compo-
nent cassette and 8-track recorders for fear that home taping would cut into
record sales.xxix
People could record music onto cassettes directly from radio broadcasts—
without using microphones—starting in 1966 when the radio-cassette
portable was introduced. That didn’t particularly concern the record labels
either. But then component cassette and 8-track decks, which enabled
people to duplicate albums onto tape, finally came out in 1968. This led to
questions about what was and wasn’t legal to do, and the answers weren’t
exactly clear.
The reality at that time was that recording any copyrighted music at
home—whether from the radio or your own records—was nominally a vio-
lation of copyright law in the United States.17 But the practical advice given
out in trade magazines was that you could tape off the radio or your own
records and get away with it as long as you kept the tapes for your own per-
sonal use only.
For example, in its Tape Recorder Annual 1968, HiFi/Stereo Review maga-
zine said:
Yet elsewhere in that same magazine, there was an article called “How to
Record Off the Air” that offered the (then)18 dubious advice that “it isn’t il-
legal to record radio programs” as long as you don’t “(1) play them before a
public gathering, (2) charge admission to listen to them, or (3) copy them for
resale purposes.”xxx
At that point, the only way to make decent-sounding copies of LPs or radio
broadcasts was on open-reel tape with an expensive deck. But by the late
Tapes 141
1970s, it was possible to make copies that were very close to the originals
with moderate-priced cassette decks19 and premium-quality blank cassettes.
In fact, it became fairly common practice for audiophiles to make cassette
copies of LPs immediately after purchasing them, because cassettes held
their sound quality a lot longer than vinyl (with typical handling).
The music industry’s concern about home taping heightened at the end of
the 1970s, when total industry revenues started to drop for the first time since
the LP versus 45 format war of the late 1940s. Despite the huge increases in
opportunities to play recorded music that portable cassette devices afforded,
overall industry revenues plummeted from 1979 through the early 1980s
and didn’t recover until the introduction of the CD in the mid-1980s. Home
taping was surely a factor in this.20
Yet the US recorded music industry didn’t make the biggest noises against
home taping; those came from the United Kingdom. The British music in-
dustry was galvanized into action by the arrival in Britain in 1981 of a new
device that made home taping even easier: the dubbing cassette deck.
Dubbing decks contained two tape transports, typically one that recorded
and played, and another that only played. Some models could be set up to
duplicate a tape at double speed, meaning that it took half the playback time
to create a copy (with a slight loss in sound quality). A few even had bidirec-
tional double-speed dubbing functionality, so that an entire two-sided tape
could be copied in a quarter of the playback time.xxxi
The dubbing deck was the brainchild of Alan Sugar, a self-made entrepre-
neur who grew up on a council estate in London and rose to become a bil-
lionaire, a media personality, and a baron, being named a life peer in 2009.
His company Amstrad (for Alan Michael Sugar Trading) produced a line of
consumer electronics and, eventually, personal computers. While on a visit
to Toyko’s Akihabara electronics district, Sugar saw a dubbing cassette re-
corder made by Sharp and designed for professional studios, and he decided
to build a consumer version of the product.
The Amstrad TS55 wasn’t an expensive audiophile product; it was a low-
priced all-in-one stereo system with a dual cassette deck, turntable, tuner,
amp, and two speakers. It debuted in October 1981. Advertisements for the
system contained notices warning against duplicating copyrighted mate-
rial. In his autobiography, Sugar admitted that he did this as a nudge-nudge-
wink-wink to give people the idea to use it for just that purpose.xxxii The BPI
(British Phonographic Industry, the United Kingdom’s analog to the RIAA
in the United States) tried suing Amstrad to keep the product off the market.
142 Key Changes
Although the BPI won its initial lawsuit, Amstrad appealed to the House of
Lords and ended up prevailing in 1988.21
Around the same time that the TS55 was launched, the BPI launched a
publicity campaign called Home Taping Is Killing Music, which included a
skull-and-crossbones logo that it got some record labels to put on the inner
sleeves of LPs, as shown in Figure 5.4. The campaign got a lot of exposure,
including widespread ridicule and parody that probably did not help to drive
home its message.
The BPI also tried to get Parliament to enact a levy of 1 penny sterling
(about 2 cents) per minute on blank cassettes, which would have raised the
price of a chrome C-90 (90-minute tape) by about 60 percent (based on a typ-
ical “street price” of $3 at the time).xxxiii This effort failed.22
On the other hand, a few artists and labels decided to use the growing
fervor for home taping as a way of ingratiating themselves with fans. The
indie rock bands Bow Wow Wow and the Dead Kennedys released cassette
versions of their albums as C-90s with music on the A side and the B side
Figure 5.4 The Home Taping Is Killing Music logo on the inner sleeve of an
album released in the United Kingdom. CC BY-ND 2.0, blondetpatrice.
Tapes 143
left blank, with recommendations to use the B sides to record other albums.
Island Records even launched a cassette series called One Plus One that fea-
tured the label’s albums on both sides of a high-quality chrome tape with
an invitation to record “whatever you like” on one of the sides; artists with
albums in the One Plus One program included U2, Tom Petty, Robert
Palmer, and Cat Stevens.
In the United States, meanwhile, evidence was mounting that home taping
of music was increasing dramatically. The Copyright Royalty Tribunal, a
body established by Congress in 1976 to administer compulsory licenses,
carried out a study that found that 20 percent of the US population ages
10 and up taped music at home. Sales of blank cassettes skyrocketed from
1976 to 1978.xxxiv Dubbing decks began to appear in the US market after
the Amstrad TS55 was introduced, including both high-end components
(with prices to match) and portables.23 By the mid-1980s most of the major
Japanese audio brands had introduced moderately priced component dub-
bing decks.xxxv
Yet the music industry’s options for legislative remedies were limited.
In 1978, a California district court ruled that home videotaping of tele-
vision broadcasts was fair use. This was in a case that pitted a Hollywood
studio against Sony, which had been marketing a videocassette format
called Betamax. The case was appealed all the way to the Supreme Court, in
what became Sony Corp. of America v. Universal City Studios, known as the
Betamax case. The Supreme Court agreed with the lower court24 and found
for Sony in 1984—though, ironically, by then Sony had lost a videocassette
format war to JVC’s VHS. Betamax was a landmark decision that would go
on to be cited in virtually every subsequent case involving new technologies
used for copying content.25
The RIAA did not launch a large-scale public propaganda campaign as the
BPI did, nor did it facilitate a lawsuit to get dubbing decks pulled off store
shelves. Instead, the music industry engaged in various behind-the-scenes
tactics to quash unauthorized taping. Labels threatened to pull ads from
radio stations that played albums in their entirety and from publications
that put blank tape ads on the same pages as ads for music. They got into
arguments with electronics makers and record retailers, both of which ac-
cused the labels of selling products that were inferior to home-recorded
cassettes for the same prices as vinyl. In fact the majors’ typical practice
by that time was to use cheap ferric oxide tape in high-speed duplication
144 Key Changes
processes that compromised sound quality, for unit costs that were lower
than LPs.xxxvi Labels also sued retail chains such as Sam Goody for allegedly
making and selling millions of unauthorized copies of commercial cassettes;
Goody pleaded no contest and was fined $10,000, and one of its executives
was sentenced to a year in prison.xxxvii
The US music industry did emulate the BPI in attempting to get a royalty
on home taping enacted—with similar results. Bills that were introduced in
the House and Senate in 1982 would have made home taping legal while im-
posing royalty fees on tape recorders of up to 25 percent of wholesale prices
and a levy on blank tapes of a penny a minute (resulting in a price hike of
about 30 percent on chrome C-90s).xxxviii
As these bills made their way through Congress, the music industry en-
gaged in what may have been one of the most bizarre26 quid-pro-quo polit-
ical deals in its history. A group of women in Washington, including Tipper
Gore (wife of Senator Al Gore) and Susan Baker (wife of Treasury Secretary
James Baker), founded an organization called the Parents Music Resource
Center (PMRC). The PMRC was concerned with what they considered to
be a rising tide of harm to children from hearing explicit song lyrics. They
pushed record labels to put warning stickers on album jackets with labels
such as X for explicit sex or profanity, V for violence, D/A for drugs or al-
cohol, and O for occult content.
The PMRC used its influence to get hearings scheduled in Congress
in September 1985. Witnesses included singer-songwriter John Denver,
Dee Snider of the metal band Twisted Sister, and—most famously—Frank
Zappa. In his withering testimony, Zappa accused the RIAA of agreeing to
warning stickers in exchange for favorable consideration of its home-taping
royalty in Congress. The bill ultimately failed, although several labels had
already agreed to put stickers on album shrink wrap anyway.xxxix Zappa
memorialized the events in the track “Porn Wars,” from his 1985 album
Frank Zappa Meets the Mothers of Prevention, which included snippets of
speakers at the hearings.
In summary, none of the music industry’s efforts—in the United States
or elsewhere—had much impact on home taping during the heyday of the
cassette.
Copyright also had implications for creators who used the cassette format
to establish one of the most popular music genres in history, as we’ll see later
in this chapter.
Tapes 145
Tape in its various forms coexisted with vinyl for the entirety of its 50-
year aggregate run as a commercial music format. Schwann catalogs of
recordings listed vinyl as the default format and noted where albums also
came out on cassette or 8-track. And for the most part, they were sold in
the same ways that vinyl was. This parallelism even extended to the LP/45
dichotomy: “cassingles” (cassette singles) were released, often with paper
sleeves instead of plastic boxes, starting in 1980 with Bow Wow Wow’s paean
to home taping, “C·30 C·60 C·90 Go.”
There were some differences in target markets among formats, such as
by music genre. By 1971, the vast majority of pop music releases were made
on both 8-track and cassette (in addition to vinyl), slightly more on the
former than the latter. A very small number of releases came out on 8-track
only, mostly compilations and greatest hits collections, with country music
dominating (Country Hits Parade, Queens of Country Music, etc.).xl
Classical and jazz were a completely different story. In the early 1970s, at
most about ten percent of classical albums were released on cassette, even
fewer—unsurprisingly—on 8-track. Perhaps 15–20 percent of jazz albums
were released on cassette, slightly more on 8-track. This situation changed
somewhat by the mid-1970s, when good sound quality had made it to
moderate-priced cassette equipment: jazz cassette releases had grown to cover
about one-third of vinyl titles, where they stayed throughout that decade;27 and
classical releases on tape never made it past 20 percent or so of vinyl releases.xli
Retail outlets for prerecorded tapes were initially the same as for vinyl. But
products that were much smaller than LPs and designed for portable listening
engendered new retail outlets, which in turn led to new distributors getting
into the business. Cassettes and 8-tracks appeared in drugstores, conven-
ience stores, and various other types of retailers. Albums that were released
on 8-track tape only, such as the country music compilations mentioned pre-
viously, were often sold at truck stops.
A key figure in the expansion of music retail to tape was Larry Finley, an-
other entrepreneur in the land of the eternal freeway. Finley (born Lawrence
Finkelstein) was the co-owner of a San Diego music venue, so he had a foot in
the door of the music business. And as a radio broadcaster and station owner,
he knew carts.
146 Key Changes
Finley got in on the ground floor of prerecorded tapes. He started his first
label, International Tape Cartridge Corp. (ITCC), in 1965. He purchased a
million blank cartridges and released titles in both Stereo-Pak and 8-track
formats.xlii In addition to releasing his own material, Finley acquired tape
rights to albums from dozens of record labels.
Finley saw the accretive effect that tapes had on overall music industry rev-
enue and became a cheerleader for the entire tape segment of the industry; his
“bully pulpit” was the Tape Cartridge Tips advertorial column that he wrote in
Billboard magazine through 1968. Finley went on to start a trade association
called ITA (International Tape Association) to set standards for things like
tape length and manufacturing quality in prerecorded tape releases.xliii ITA
subsequently branched out into videotape and then digital distribution.28
Yet cassette sales weren’t limited to the same channels as vinyl. Because
cassettes were originally a home recording format, they blurred the lines be-
tween creators and distributors to a greater extent than ever before. While
major labels were using facilities with high-speed commercial tape duplica-
tion machines that sold for hundreds of thousands of dollars,xliv anyone with
two cassette decks—or a dubbing deck—could get into the duplication busi-
ness. Smaller-scale tape duplicators that handled up to four tapes at a time,
at up to 16 times cassette speed, became available for a few hundred dollars
or less. And manufacturing costs for cassettes in small volumes were reason-
able, whereas they were prohibitively high for vinyl.
As a result, cassettes powered the rise of various independent music distri-
bution networks that wouldn’t have been possible without them. The earliest of
these were tape-trading networks, in which fans would duplicate and swap tapes.
Most of these were bootlegs of concerts and studio outtakes rather than dubs of
commercial albums. The first was the First Free Underground Grateful Dead
Tape Exchange, which Les Kippel started in 1971—using open-reel tape at first—
for trading tapes of the Dead’s epic live shows that the band allowed fans to make.
The tape-trading ethos was to swap tapes for free (or only for “B&P” or blanks-
and-postage charges) and to shun those who would sell the tapes for profit, al-
though inevitably many of the tapes fell into the hands of for-profit bootleggers.
The pioneer in this area was the New Jersey-based singer, songwriter, and
multiinstrumentalist R. Stevie Moore. Moore is one of the godfathers of indie
rock and one of the most prolific recording artists of all time, with over 150
studio albums. Releasing all that material to his modest-sized but dedicated
fan base on vinyl would have been impracticable. Moore began releasing
music on cassettes in 1973, just when hi-fi sound quality from the medium
was starting to be available. In 1982 he started the R. Stevie Moore Cassette
Club, a mail-order service for his albums. The vast majority of Moore’s music
is self-released.29
An informal movement known as the cassette underground emerged in the
1980s. Several tiny cassette-only labels were born, most of them specializing
in various flavors of postpunk, indie rock, and experimental music. Some
music magazines also distributed sampler cassettes of new artists with their
issues. For many artists, the ability to record on cassettes and distribute in
low volumes made the difference between not releasing music and releasing
music.xlv Some of these labels didn’t even distribute prerecorded cassettes;
instead they asked fans to send them blank cassettes and returned them with
albums recorded on them.xlvi The cassette underground produced a handful
of “stars” such as Moore, avant-garde guitarist Eugene Chadbourne, and the
British indie bands Cleaners from Venus and Danny and the Dressmakers.
But otherwise the cassette underground’s activity generally took place in
lieu of vinyl and other traditional aspects of recorded music distribution.
Although the “lo-fi” movement in indie music was often associated with
cassettes, the music on cassette-only releases was created in much the same
way as music released on LPs and 45s.
The same cannot be said of the other sphere of creators’ activity based on
cassettes, one that helped usher in one of the biggest movements in popular
music ever: hip-hop. It’s no exaggeration to say that the cassette was to hip-
hop as the transistor radio was to rock & roll; the genre probably would not
have flourished without it.
Hip-hop got its start during the 1970s in the South Bronx. DJs and rappers
like Kool Herc, Grandmaster Flash, Lovebug Starski, Grand Wizzard
Theodore, and Afrika Bambaataa would entertain crowds at parties, armed
with DJ rigs and PA systems. They would play records, repeat sections of
songs that got people dancing, play records over each other, and rap over the
music. They invented turntable scratching techniques that added spice to the
mix. In other words, they used three basic tools—turntables, microphones,
and mixers—in ways that were so innovative and distinctive that the results
became a new form of audio expression.
148 Key Changes
Grand Upright Music v. Warner Bros. Records in 1991, in which the publisher
of Gilbert O’Sullivan’s 1970s soft-rock hit “Alone Again (Naturally)” sued the
record label of rapper Biz Markie for his “Alone Again,” which sampled the
track; it was found to be infringing. Another important case was Campbell
v. Acuff-Rose in 1994, involving the sampling of Roy Orbison’s “Oh, Pretty
Woman” on 2 Live Crew’s “Pretty Woman,” which was found to be fair use.
Still, mixtapes kept on going in the hip-hop underground, even retaining
the word “mixtape” after the transition to CDs.
Yet to many everyday music listeners during the 1980s and 1990s, “mix-
tape” didn’t mean hip-hop DJs and MCs. A mixtape was a selection of
songs that you recorded onto a cassette to give to someone else—a friend,
a lover, a would-be lover, or someone whom you wanted to educate about
music. Unlike today’s streaming music playlists, cassette mixtapes had the
all-important element of effort involved in making them and personalizing
them for the recipient.
Mixtapes were meant to be one-offs, not duplicated. You constructed your
mixtapes in real time from your vinyl collection, and you created “cover art”
and “liner notes” on J-cards (the paper inserts to cassette boxes) by hand.
Just as radio DJs did in the days of progressive FM, you considered the
order of and transitions between songs just as carefully as you did the songs
themselves.
But mainly, consumers used cassettes to open up new spaces and use
cases for music listening, and new devices emerged that enabled those new
use cases.
One of those new use cases was music on demand in cars. The 8-track
tape was invented for this, but cassettes began to overtake them there. By
the end of the 1970s, the leading car audio systems included cassette rather
than 8-track players, and automakers tended to include factory-installed
cassette players in their cars. Auto-reverse—the ability to sense the end of
the tape and reverse playback direction—came to cassette players in cars
before it came to home players. Cassettes had also displaced 8-tracks in the
home: although a few component 8-track record/playback decks were avail-
able, none of them were audiophile-grade machines, let alone ultrahigh-end
equivalents of the Nakamichi 1000.
150 Key Changes
The other new use cases were about personal portability. Two new types of
devices, introduced in close succession, significantly enhanced the cassette’s
value to users on the go in different ways.
The first of these was the boombox, the combination radio and cassette
recorder/player. The boombox wasn’t a technological breakthrough; it was a
nickname given to an evolution of the combination radio-cassette recorder,
which dated back to 1966. But these were compact devices with unimpres-
sive sound quality. Portables that had more powerful amplifiers and stereo
speakers with decent (or exaggerated) bass response, arguably the first
true boomboxes, came from Japanese electronics makers and appeared in
Western markets starting around 1978. Boomboxes offered “big” sound at
prices considerably lower than home component stereo systems.
By the mid-1980s, boomboxes had metastasized into mammoth units the
size of suitcases, some of which had detachable speakers. The unit that was
supposedly the largest one-piece boombox ever produced was a Taiwan-
manufactured unit, shown in Figure 5.5, that was marketed under var-
ious no-name brands and model numbers and known among boombox
aficionados as the “Master Blaster.”33 It was almost 3 feet wide and 18 inches
high; it included side carrying handles, rolling wheels on the bottom, dual
Figure 5.5 The Elta 6930 “Master Blaster” from the mid-1980s, also badged as
various other brands and model numbers, the largest one-piece boombox ever
made. Max Poser/My Radio Berlin.
Tapes 151
cassette decks, a four-band radio, graphic equalizer, and speakers with 10-
inch woofers.
The boombox’s signature use case was sharing your music with a group
of friends—and/or involuntary passers-by—in public, wherever you went.
Boomboxes began to be associated—rightly or wrongly—with urban Black
and Latino youth, who would carry them around on the streets. They would
be used for impromptu parties. They were known as “ghetto blasters” before
the more acceptable term “boombox” emerged. By the late 1980s, New York
and other cities had banned their use in public places as part of “sweep laws”
that some considered to be racially motivated.xlix Perhaps the pinnacle of the
boombox’s cultural significance was in Spike Lee’s 1989 movie Do the Right
Thing, where Radio Raheem uses it to blast Public Enemy’s “Fight the Power”
and the device becomes a catalyst for racial violence and police brutality.34
The next late-1970s innovative device addressed a different use case: lis-
tening to music with high-quality audio in private wherever you went. That
product launched in July 1979: the Sony Walkman, one of the most iconic
devices in the history of consumer electronics.
The Walkman was also an evolution—of previous Sony pocket-size cas-
sette machines such as the TCM-100 (a.k.a. Pressman). But it was the first
such device devoted solely to stereo playback, and the first to ship with high-
quality stereo headphones.
The idea for the Walkman came from Sony cofounder Masaru Ibuka. An
opera buff, Ibuka challenged his engineers to find a way for him to listen to
operas with good sound quality on his many long plane rides. The engineers
essentially took the TCM-100 design, removed the recording functionality,
and added stereo playback to create the TPS-L2, shown in Figure 5.6.35 The
product, including lightweight headphones, sold for the equivalent of about
$500 today; it did so in volumes beyond Sony’s wildest expectations.
The Sony Walkman led to countless other personal portable cassette
players from various manufacturers in ever-shrinking sizes. Sony’s own
WM-20, introduced in 1983, was the size of a cassette box. “Walkman” came
dangerously (for Sony) close to becoming a generic for “personal portable
cassette player,” as “Kleenex” did for “facial tissue.” (Sony would subsequently
use the name “Discman” for its portable CD players.) Most other electronics
makers simply used model numbers for their small portable cassette players
rather than use Sony’s trademark or make up their own names,36 although
searching for “Walkman” on eBay today yields vintage portable cassette
players from other brands.
152 Key Changes
Figure 5.6 The Sony Walkman TPS-L2 portable cassette player, the first
Walkman model, from 1979. Used by permission of Sony Electronics Inc. All
Rights Reserved.
Tape became the dominant format for recorded music for a while, but
not until 30 years after it was first used for that purpose. And for the first
half of that time, it was a minuscule part of the recorded music industry.
By 1957, prerecorded open-reel tape was a $7 million market in the United
States,l which was less than 2 percent of overall annual recorded music rev-
enue at that time.li Not only were tapes harder to use than LPs (and required
equipment that was more obtrusive in the living room), but they cost more
too: in 1963, just before cassettes were introduced, typical retail prices for
Tapes 153
prerecorded open-reel tapes were $7–$9, versus $4–$6 for LPs and lower for
budget releases.
The RIAA began tracking total annual US recorded music revenues con-
sistently in 1973; Figure 5.7 shows how revenue from vinyl and tape formats
changed from then until 1990, the peak year for tape sales. Sales of 8-track
tapes were an order of magnitude higher than open-reel tape, but they never
approached the level of vinyl sales, despite the availability of almost all pop
music releases on 8-track. In 1973, as the figure shows, 8-track had become
a mainstream format. Revenue from 8-tracks hovered around 40 percent of
revenue from LPs through the late 1970s as the two formats grew in tandem.
Then 8-track sales began plummeting in the late 1970s and disappeared en-
tirely by 1982.
Cassettes eventually became the best-selling music format, but that took
a while. Cassette sales lagged behind 8-tracks for several years. In 1973, they
were only 16 percent of 8-track sales (and only 6 percent of LP sales), pulling
in a mere $76 million. Cassettes gradually gained on both vinyl and 8-tracks
through the mid-1970s, then continued growing as 8-tracks started to de-
cline. In 1979, sales of prerecorded cassettes surpassed 8-tracks.
Cassette sales finally exceeded vinyl sales in 1984, despite the rise of home
taping. The obvious reasons for this were cassettes’ small size, exciting new
devices that enabled portable use, and potential sound quality to rival vinyl.
$4,000
$3,500
$3,000
$2,500
$2,000
Vinyl
$1,500
$1,000
8-Track
$500
Cassette
$0
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
Figure 5.7 Recorded music revenue from vinyl, 8-track tapes, and cassettes,
1973–1990, millions of dollars. Data source: RIAA. Bill Rosenblatt.
154 Key Changes
But there was another important reason for the timing of cassettes’ ascend-
ancy: price.
When cassettes and 8-tracks first came out, their list prices tended to be
a dollar or two higher than vinyl LPs, if lower than open-reel. But the price
gap narrowed steadily through the 1970s and eventually disappeared. When
labels raised mainstream LP prices from $5.98 to $6.98 in 1976, they raised
cassettes and 8-tracks to $7.98 (and phased out open-reel). But then when
the labels raised LP prices to $7.98 in 1978, they held tape prices steady. By
the start of the 1980s, the big retail chains—which typically discounted eve-
rything and priced each type of item according to its own profit margin—
were pricing LPs and tapes equally.lii
Cassettes leapt passed vinyl’s peak revenue of $2.7 billion in 1987. They
remained the dominant format by revenue for a short time until CDs
overtook them in 1990. And although cassette revenues started to decline
after that, cassettes continued to outsell vinyl until the early 2000s, when
both formats faded away.
Cassette sales started up again in the mid-2010s. A catalyst for this was
the 2014 Disney sci-fi film Guardians of the Galaxy. In the movie, the hero’s
dying mother gives him a mixtape of 1960s and 1970s hits, which he plays
on a Sony Walkman to help him stay connected to his home on Earth.
Disney created a soundtrack album based on the mixtape and released it
on cassette, complete with faux-handwritten J-card, in an exclusive ar-
rangement with indie record stores. It was the first cassette that the studio’s
record label had released since 2003. The soundtrack (which was subse-
quently released on CD and digitally) hit the top of the Billboard album
charts that year. Prices for used original Sony Walkman models on eBay
spiked.
But this phenomenon has been nowhere near the size of the vinyl renais-
sance we discussed in the previous chapter. The RIAA does not track new
cassette sales anymore—other than to include it in the tiny “Other Physical”
category,37 which constitutes less than 0.1 percent of industry revenue.
Discogs.com, the premier marketplace for used records, lists 1.4 million
cassettes on its website at this time of writing, compared to over 46 million
vinyl records. And the video for Adele’s “Easy on Me,” the first single from
her long-awaited 2021 album 30, featured a cassette as “product placement”;
but the actual cassette of the album only sold 2,000 copies,liii compared to (at
this writing) about two billion streams.
Tapes 155
Conclusion
Still, unlike other tape formats, cassettes have a place in music fans’ hearts.
There has been enough new cassette activity since the mid-2010s that a few
cultural commentators, particularly in Britain,liv proclaimed a revival and
made various attempts to explain why it’s happening.38,lv Many of these
explanations could apply to vinyl as well, but not all of them. For one thing,
unlike vinyl, no one thinks that cassettes sound superior to today’s digital
audio. Whereas various audiophile-grade turntables are being designed
today with price tags reaching up into Rolex watch territory, the few cassette
machines that are sold today are cheap portables and retro nostalgia models.
Instead, the special appeal of cassettes nowadays is that of the cassette
underground of the 1980s. Artists who release cassette-only albums today
could certainly put their music out faster and cheaper online. But to these
artists’ fans, cassettes evoke an era when they signified closer connections
to the people whose music is on them—who may have been personally in-
volved in loading blank tapes into duplicators, packaging the tapes in boxes
with J-cards they designed, and sending them off in the mail. Never mind
that some of the artists who used to do this in the 1970s and 1980s now say
that they used cassettes because they had no other reasonable choices and
despite their mediocre sound quality and poor capacity for cover art.lvi That’s
the DIY ethos that today’s cassette artists embrace and seek to maintain.
Meanwhile, as the popularity of both cassettes and home taping began to
rise and that of vinyl began to fall, the music industry plotted its next move. It
decided to address tape’s tradeoffs against vinyl by introducing a new format,
one that combined the portability of cassettes with sound quality that was ar-
guably better than vinyl but that didn’t support duplicating records at home.
As we’ll see, this would work out extremely well for the industry—at least
until the last of those attributes ceased to be true. Yet just before the CD hit
the market, another set of outside forces came back in to disrupt the music
industry as it did back in the 1930s: broadcast media.
6
Television
Television Rules the Nation—Daft Punk
When MTV launched in August 1981, its success was far from assured. In
fact, the network had a list of substantial shortcomings. Its entire library of
videos consisted of 125 songs. With 24 hours of airtime to fill seven days
per week, every single track would need to be played three times per day, no
matter the quality or popularity. That presented an issue, because many of
those videos were from British new wave bands who were not particularly
well-known in the United States at the time. And the ones from American
artists included acts like Devo, who had developed a cult following but had
not achieved mainstream acceptance. Very few households in the United
States could experience the fledgling network because most cable networks
simply chose not to carry it. And the networks that did offer MTV at first were
in limited media markets such as Columbus, Ohio, and Tulsa, Oklahoma.
This represented a total addressable viewership of only 1 million households
at launch. With such a limited audience, advertisers were not lining up to
purchase commercial time.
Given those constraints, the nascent MTV of 1981 could certainly be
deemed a “Minimal Viable Product” for a music channel. Although they
were not a format that was distributed to consumers in the same way as vinyl
or tapes, music videos became so important to MTV’s audience that devoted
fans became known as the “MTV Generation.” MTV unleashed a wave of
creativity and turbocharged the popularity of superstars such as Michael
Jackson and Madonna, whose videos were often as memorable as the melody
or lyrics of their songs.
We’ll review in detail how MTV overcame its initial deficiencies to be-
come a cultural phenomenon and an enormous financial success. MTV
did have one crucial asset right from the start that had been true for more
than 50 years: people loved to watch, and not just listen to, recorded musical
performances.
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0006
Television 157
The idea of combining audio and visual performances was not new; it
dates all the way back to the early days of the phonograph, although that new
invention did not play a role at first. Music publishers wanted to sell sheet
music and needed to expose new songs to as broad an audience as possible.
Getting a big name on the vaudeville circuit to perform a new song in their
act was highly desirable, but there was no guarantee that a particular artist
would like the tune. A surefire technique was to hire “song pluggers” to play
the new material at cafes, restaurants, and even department stores.1
The pluggers would repeat their performances in the evenings in
nickelodeons and movie theaters. Before the show and during the intermis-
sion, the pluggers would play the new song while hand-colored pictures on
painted canvases or glass slides that related to the lyrics would be shown be-
hind them. These “illustrated songs” were popular with audiences as addi-
tional free entertainment. They proved to be an effective promotional tool
because they made popular songs into “visual spectacles” some 90 years be-
fore MTV.
The inventors of the day were pursuing a different path to meld music and
images. In 1895, Edison introduced the Kinetoscope, which combined his
cylinder phonograph with a motion picture player in a single physical cab-
inet. The two media were not synchronized, and the ability to exhibit films
to an audience in a large screen soon superseded this “peep-show” approach
that served only a single viewer at a time.
Efforts to perfect the audio-visual experience continued throughout the
silent film era, and the age of the “talkie” arrived in October 1926. Al Jolson
was already one of the most popular American musical artists when he
starred in the first feature film with sound, The Jazz Singer. Jolson’s musical
performances were recorded on the set and the advertisements highlighted
that “You’ll See Him and Hear Him on Vitaphone,” the latter being the tech-
nology that synchronized the audio with the film. The movie was a major box
office hit earning a total of over $2.5 million, or almost $40 million in 2022
dollars, cementing the value of sound as more than a mere gimmick to the
motion picture industry.
Movie musicals became a staple in cinema as the extravagant produc-
tion numbers crafted by Busby Berkeley and the intricate choreography
of Fred Astaire and Ginger Rogers entertained millions during the Great
Depression. Musical numbers were so popular with the public that the
Mills Novelty Company of Chicago updated Edison’s Kinetoscope idea for
the 1940s and created a “movie jukebox” called the Panoram. Each unit
158 Key Changes
Broadcast television took note of this success and of the growing and
rabid fanbase of a certain young singer from Memphis. Both Steve Allen and
Milton Berle invited Elvis Presley to appear on their shows. Ed Sullivan, who
hosted the top-rated TV Variety program, was not a fan; he felt that Elvis was
unfit for family viewing. But the ratings for the appearances on his rivals’
shows were too large to ignore. Sullivan featured Presley on September 9,
1956 and sixty million people tuned in. To avoid offending anyone with the
gyrations of Elvis’s pelvis, they shot his performance from the waist up.iii
One year later, the same Baby Boomers who screamed for Elvis began
tuning in every afternoon to a show that played just the hits they cared
about: American Bandstand. Hosted by former radio DJ Dick Clark, the
show featured artists such as Paul Anka and Chubby Checker lip-synching
their hits while teenagers in the studio danced along. The program had
started on a local Philadelphia station in 1950 as a TV version of the radio
show “Bandstand.” Though there were occasional studio guests on that local
show, much of the airtime was filled by short films of prerecorded musical
performances. By 1959, American Bandstand, propelled by rock & roll as the
soundtrack of teen culture, reached an audience of 20 million each day.iv,v
The musical show that appealed to millions of older music fans in the early
1960s was considerably more sedate. Sing Along with Mitch, hosted by Mitch
Miller, featured a male chorus performing pop standards with the lyrics in-
cluded at the bottom of the screen.4 The karaoke-style show of its day proved
so popular that Miller recorded and produced albums with songs from the
show complete with tear-out lyrics. The weekly show was canceled in 1964 as
musical tastes shifted even more dramatically toward rock.
Before that seismic shift, the early 1960s brought one more attempt to
build a business around the concept of a video jukebox. Scopitone machines,
originally offered in Germany, were deployed in adult establishments, in-
cluding cocktail lounges, to avoid competing with the jukeboxes that were
popular with teens.vi These units improved on the Soundies by offering short
films in color, and patrons could choose which one they wanted to watch
rather than being restricted to watching the next one on a loop. Over 500
machines were installed by 1964, with films featuring performances by art-
ists like Neil Sedaka and Bobby Vee. The business failed to convince many
of the next wave of music superstars, including those who were part of the
British Invasion, to include their songs; this fatally wounded one more video
jukebox platform.
160 Key Changes
Of course, the band leading that invasion was the Beatles, who began
dominating the US charts in 1964. On a trip to London, Ed Sullivan happened
to be at the airport the same day as the Beatles returned to England from a
European trip and witnessed their screaming fans first-hand. He booked the
band for their first live TV appearance in the United States. On February 9,
1964, at 8 p.m., 60 percent of the households in the United States tuned in
to watch the Beatles shake their mop-top haircuts and try to be heard over
a crowd of screaming teenage girls. The Sullivan show had received 50,000
requests for the approximately 700 seats in their theater. The estimated
72 million viewers handily topped Elvis’s record, and TV fueled Beatlemania
and the British Invasion.vii
The broadcast networks, seeing the drawing power of rock & roll music
on television, followed a tried-and-true business practice: create imitations.
Shindig premiered on ABC in September 1964, and Hullabaloo followed
on NBC in January 1965. Shindig had replaced Hootenany, another music-
oriented show that focused on folk music, whose ratings had plummeted as
rock & roll began to top the charts. The guest lists for the two shows read
like the Billboard charts, including Sam Cooke, the Byrds, the Supremes,
the Lovin’ Spoonful, and Simon & Garfunkel. But the networks were truly
broadcasters: they tried to balance appealing to younger viewers while
attracting the Mitch Miller crowd as well. They featured the artists that would
draw teens but added hosts like Jerry Lewis and Zsa Zsa Gabor who could
make the show feel less risqué and more palatable for older viewers. This
middle-of-the road combination proved less than compelling for either au-
dience segment, and within two years both shows were off the primetime
schedule.
NBC had much more critical success with the Monkees television show,
which first aired in 1966. Clearly inspired by the successful Beatles film
A Hard Day’s Night, which captured the Fab Four’s frantic lives chased by
rabid fans, the producers chose to put together a TV-ready foursome who
could portray an up-and-coming long-haired rock band. Davy Jones,5 Peter
Tork, Micky Dolenz, and Mike Nesmith were selected after a lengthy au-
dition process. They sang catchy songs6 but, at least initially, did not play
the instruments on their recorded tracks. The combination of the comedic
elements with cleverly filmed music numbers earned them the Emmy for
Best Comedy series in 1967, and the band sold 8 million records in the first
four months of the show.
Television 161
During the 1970s, the programming of rock music on television shifted its
focus from primetime to late night. The Midnight Special and Don Kirshner’s
Rock Concert both featured live performances, in contrast to the typical
appearances on television of lip-synching to prerecorded tracks. To compen-
sate for the monaural sound of broadcast TV, Rock Concert performances
were recorded in stereo and then simulcast via FM stereo stations to create
a superior audio experience. Although more diverse artists such as Earth,
Wind & Fire were sometimes featured on these late-night programs, R&B
and soul artists commanded the stage on Soul Train, a syndicated music pro-
gram airing weekly beginning in 1971. At that time, it was the first commer-
cial TV program aiming for a Black audience that was produced by Black
people, led by the Chicago-based radio news reporter turned entrepreneur
Don Cornelius. There were other syndicated variety shows that were geared
towards fans of specific music genres: country fans watched Hee Haw while
more traditional music (and polka) fans chose the Lawrence Welk Show.
Despite the well- established appeal for prerecorded audiovisual
performances, there was no regular outlet for the short-form music videos
being created as promotional tools for radio stations in the late 1970s.
Bob Pittman, who later became one of the founders of MTV, cohosted a
music video and news show that ran on some NBC stations in 1978. Mike
Nesmith, the former member of the Monkees, crafted a new show produced
by Jac Holzman, the founder of Elektra Records, to promote Warner
Communications’ records. Nesmith recognized that the short music videos,
like the one for his song “Rio” that was produced as a marketing tool for
other territories, could promote the song within the United States too.
Calling his show “PopClips,” after a similar program airing in Australia
and New Zealand, he pitched his pilot to broadcast networks without suc-
cess. In 1979, John Lack, an executive for the catchily named Warner-Amex
Satellite Entertainment Company (WASEC), the entity charged with pro-
gramming for Warner Communications’ cable systems, met with Nesmith.
Lack liked the PopClips idea and contracted for 13 episodes, which played
several times per day on Nickelodeon, a channel that specifically targeted
young audiences with its programming. Homes in Columbus, Ohio, that
subscribed to Warner Communications’ nascent QUBE network7 watched
those music videos over cable television. PopClips had comedians as
announcers, including Howie Mandel, though Lack felt that they took away
from the focus on music.viii
162 Key Changes
There is some dispute about how much the PopClips experience influenced
the team that launched MTV. Lack hired Bob Pittman as head of Pay-TV at
Warner Cable; Pittman brought his own broadcasting experience, which in-
cluded the local NBC music video show. John Sykes, who was hired as MTV
Promotion Director, thought that Nesmith did not deserve any credit and
that he could “go jump in a lake.”
Regardless of the exact heritage, it is clear that creating “radio on TV” by
expanding from limited length programs to an entire channel built around
music videos was the seed of an enormously successful franchise. Creating
a platform for advertisers to connect with 25–34-year-olds who had become
more and more difficult to reach via traditional TV was the important new
element. The music videos themselves were not particularly innovative—at
least at first.
MTV kicked off on August 1, 1981 by playing “Video Killed the Radio
Star” by the Buggles. The prophetic title has become a popular music trivia
question, but Steve Casey, an MTV programming executive, said that it was
the second video they aired that really foreshadowed the effects on the music
industry: “You Better Run” by Pat Benatar.
Television signals broadcast over the air require a “line of sight” be-
tween the transmitter and the receiver. That allows the airwaves carrying
the signal to reach the viewer’s antenna. In the early days of television, the
only broadcast stations were in cities, and people in more rural or moun-
tainous locations often had difficulty picking up those over-the-air broad-
cast signals. To solve this problem, a single large and more expensive antenna
was constructed in each of these more remote areas to receive the broadcast
signal. Then, coaxial cables from that “community antenna” were run to in-
dividual homes to convey the programming via radio frequency signals. To
compensate for the signals weakening as they traversed the cable, amplifiers
were deployed along the way to boost the signals. This architecture became
known as “Community Antenna Television” or CATV.
By the mid-1970s, several advances in technology improved both the
quality of CATV service and the costs of delivery. Geosynchronous satel-
lites were used to beam programming such as HBO to that master antenna
or “cable headend,” thereby eliminating the line-of-sight restriction. When
fiber optics replaced coaxial cables, signal loss decreased, which reduced the
number of amplifiers needed to maintain transmission quality. Those same
fiber optic cables were capable of carrying much more data than the older
coaxial cables. That higher bandwidth meant that fiber-optic cable systems
could supply many more channels to their subscribers.
The greater channel capacity allowed companies like Warner Cable to
focus on a variety of niche audiences rather than programming to appeal
to as broad a population as possible, as was necessary with broadcast televi-
sion. There was room for channels targeting golf enthusiasts, animal lovers,
and bargain hunters who wanted to shop from their couches—and even one
that showed music videos round the clock. Fortune magazine named MTV
as one of its 1981 “Products of the Year,” citing its ability to “serve up specific
audiences for advertisers, as specialized magazines do.”ix
When Warner surveyed potential MTV viewers prior to launch, 85 per-
cent said they would watch music videos. An even higher percentage said that
they owned stereo systems and that most of those systems were located in the
same room as their television set (which could deliver only monaural sound).
When a cable system agreed to carry the channel, MTV required it to pur-
chase a “stereo transmission processor” from a company they recommended.
The $1400 unit, developed by Dolby, modulated the stereo soundtrack and
allowed it to be transmitted through the cable TV network along with the
video signal. By connecting the cable-TV output to an FM receiver’s antenna
164 Key Changes
input and tuning the receiver to the right frequency, customers could listen
to the stereo sound. Many cable operators were able to charge subscribers
about $1.50 extra per month above their standard fees simply for providing
the ability to listen to programming in stereo.x The FCC adopted a standard
for stereo audio in television transmission in 1984, and manufacturers even-
tually built the capability to receive stereo into televisions.
We have all come to expect that every music video displays the name of
the song and the artist. But the ability to overlay textual information on a
video was invented not long before MTV’s launch, by Systems Resources
Corporation in the 1970s. This higher-tech process replaced filming dec-
orative cards or cranking credit rolls by hand. The company was renamed
Chyron, and “chyron” became the generic term for “text or graphics overlayed
on video.” MTV made it a policy to use a chyron to add the name of the song
and the artist on every single music video at the beginning and the end to in-
crease its promotional effectiveness.
Not only was MTV a channel in the programming sense, but it was the
new distribution channel for the music video format. MTV did not create
music videos, but it also had to persuade artists and labels to let the channel
air them. They did not decide which cable operators would offer the channel
to viewers, but they had to persuade businesspeople at those operators that
MTV could attract a youthful fanbase.
MTV’s leaders learned early on that convincing these gatekeepers was no
small task. Billboard magazine held the first Video Music Conference in the
fall of 1979. At that point, MTV’s founders had not even decided on a name
for their channel. Nevertheless, John Lack attended the event and, during
one of the panels, he announced the plan for a new network showing music
videos around the clock. Many of the label executives in the room found the
concept, shall we say, less than compelling. Sidney Sheinberg, the President
of MCA Records, stood up and declared, “We won’t give you our fucking
music.”
Those sentiments exemplified the labels’ reluctance to furnish their
copyrights to establish a new format. Of course, that reluctance can some-
times be overcome through substantial upfront payments specified in
licensing agreements. MTV’s launch plans did not include making any such
Television 165
payments. Instead, they were asking the labels to offer their catalog of music
videos for free. As Bob Pittman put it, “This was not a business model. That’s
chutzpah.”xi
MTV’s founders knew that one of the most common complaints about
radio was that DJs could not be relied on to announce the details of the songs
being played. How could radio play provide promotional value if they failed
to declare who or what was being promoted? MTV’s use of chyron tech-
nology to overlay the title of the song and the performer’s name on each
clip enabled the new network to argue that every single video would serve
as a three-minute advertisement for the music. Addressing radio’s weakness
head-on was a positive for MTV, but it would not have been enough by itself
to persuade the labels to take a chance on supplying their music videos.
Yet MTV’s timing was propitious for it, if not for the music industry itself.
As we saw in Chapter 5, in the early 1980s, the record business was in the
midst of a multiyear slump. Over the ensuing four years, the number of units
sold and associated revenues both dropped almost 20 percent from 726 mil-
lion and $4.3 billion respectively. In addition to the reasons discussed in
Chapter 5, the shift was also caused by demographic trends. Younger music
fans had been the core of the economic engine through the rock era, but
those Baby Boomers were aging out of their prime music purchasing years,
and younger music fans were not picking up the slack. Fifteen-to-24-year-
olds represented 45 percent of the industry’s revenues from 1979 through
1981, but that number was heading downward toward 39 percent by 1982.
Furthermore, the industry was broadly dissatisfied with radio’s performance
in breaking new artists because of increasingly restricted playlists.xii
Given the industry slump, many labels were willing to take a chance on
MTV, although both MCA and Polygram remained on the sidelines. The
record companies had a limited number of music videos on hand at that time,
and giving them to MTV seemed like a risk worth taking. As Al Teller, then
President of CBS and Columbia Records, explained, “The timing was perfect.
The music industry was in the doldrums.”xiii MTV, if successful, would boost
those lagging results by activating a younger demographic. The thinking at
the time was that once MTV had established itself as a viable channel (in
both senses of the word), there would certainly be opportunities to revisit the
financial arrangements and to extract payments for labels and artists.
It is worth noting that it is riskier than it might seem for the labels to ac-
cept terms that would never have been contemplated if not for the industry’s
declining fortunes. The more successful the channel becomes, the more
166 Key Changes
Sheinberg’s of MCA, but the bottom line was the same. He escorted Lack to
the door and threw him out.
To realize its full potential, MTV had to get Malone and other equally con-
servative cable operators—as well as the recalcitrant labels—to revisit their
opposition. The initial video catalog may have been limited in size, but the
young viewers who received the channel liked what they saw. Those fans be-
came the catalysts to increase the size of both the video catalog and the ad-
dressable audience.
more cutting-edge and less mainstream. These acts chose to make avant-
garde music videos to serve their own artistic purposes.
There were some concerns in the industry that fans would not enjoy
watching music videos of bands simply lip- synching over prerecorded
songs. The productions in the early 1980s already featured sounds, such as
synths and drum loops, that were “manufactured” rather than played live.
Recreating those recordings accurately during live performances required
the use of those same techniques in venues. Fans had become accustomed
to more “artificial” elements within performances of their favorite songs
whether live or on TV.xv
As we’ve seen with the transistor radio and the personal tape player,
younger, more passionate music fans are often the first to adopt new formats
and to make them their own. MTV was no different: it quickly became
“must-see TV” for the youthful, suburban fans in communities with access
to the channel. By October of 1982, a Nielsen survey showed that 85 percent
in the key demographic group for the channel were watching the station be-
tween four and five hours per week. In a survey of 600 high school students,
80 percent were watching MTV two hours per day on average.
Filling the network’s schedule seven days per week and 24 hours a day re-
quired repeated plays of their constrained catalog. This had the unintended
consequence of creating a limited playlist like most radio stations. And, just
like those stations, the constant drumbeat of promoting a limited number of
songs to viewers made an impression.
The new American fans of bands in heavy MTV rotation, such as A Flock
of Seagulls and Duran Duran, bought those bands’ albums at unprecedented
levels. Columbia Records had very low expectations for sales in the United
States for the Australian band Men at Work and pressed fewer than 8,000
copies of their debut album initially. The video of the band’s single “Who Can
It Be Now” cost $5,000. But once MTV started airing it, the album climbed to
no. 1 and stayed there for several months, much to the surprise of the label.xvi
As one would expect, the increased popularity and sales volumes of
the artists whose videos were being shown by MTV were concentrated in
markets that aired the channel. Having some cable operators offer the station
while others did not created a “natural experiment” that demonstrated the
impact of the network’s impact on sales. In October 1981, Billboard maga-
zine surveyed record stores in those cities where MTV was being offered to
viewers. Artists whose videos were being aired showed increased sales when
compared to other cities and other artists.xvii,xviii
Television 169
Another example of MTV’s impact was Stray Cats, a rockabilly trio from
Long Island that had no radio airplay and a low profile before a video of
its song “Stray Cat Strut” went into heavy rotation on the channel. When
the band’s tour hit MTV markets, they sold out all the clubs where they
appeared. As singer/guitarist Brian Setzer noted, “We’d go to a place like Des
Moines and play for a thousand people.” Fans purchased 200,000 copies of
the band’s album. They would have topped the charts if not for the even more
successful record from Men at Work that was boosted by MTV as well. Tulsa,
Oklahoma, was one of those MTV cities and, before long, Sound Warehouse,
a local retailer with three locations, began to stock its shelves based on the
videos in rotation.xix
Retailers were not the only ones noticing that fans were making buying
decisions for tickets and albums based on the artists they enjoyed on MTV.
The labels that had been withholding videos over the lack of payments from
the channel overcame their reluctance. Jordan Rost, then VP for Research
at the parent company Warner/AMEX, discussed the results and said, “It
did not take a lot of convincing as this was a 3-minute ad. It was just better
airplay.”xx
The eagerness to build a motivated fanbase brought Polygram on board,
followed shortly thereafter by Universal. And the marketing teams at all
the record companies cited the economic impacts of being aired on MTV
to justify the investment in new videos covering a broader array of priority
releases. Video budgets increased and the promotional impact grew as well.
The video for “Every Breath You Take” by the Police cost close to $100,000.
Jeff Ayeroff, the Creative Director for Warner Brothers Records, pointed to
the 5 million albums sold and labeled the returns as “phenomenal.”xxi In May
of 1981, only 23 of the top singles in the Hot 100 chart had videos. Two years
later, that number was up to 59; and one year after that, more than three-
quarters of the top tracks had videos.xxii
The labels expanded the flow of content to the channel to trigger more
sales. But the cable operators who controlled access to a larger audience had
a different set of priorities. According to Bob Pittman, advertisers were re-
luctant to be associated with “sex, drugs, and rock & roll,” and without those
ad revenues it was even more difficult to convince the cable operators of the
channel’s potential. MTV hired an ad agency to create a campaign that would
send a message to their potential viewers in what Pittman referred to as a
“Hail Mary pass to obtain the distribution the channel needed to survive.”
The first advertising message to explain the network to a broader audience
170 Key Changes
was “Cable Brats,” with a sub-tagline of “Rock ‘n roll wasn’t enough for them,
now they want their MTV.” Whenever the MTV executives would show up
to make their case to the cable operators, they were referred to as the “cable
brats.”
Needless to say, that campaign did not last long; the message was soon
boiled down to “I want my MTV.”xxiii Commercials featuring that catch-
phrase blanketed the airwaves in specific markets where the channel was not
available. MTV ran the ads in Denver, and the fans began calling TCI, their
cable operator, almost immediately. The number of calls climbed into the
thousands by the second day, and soon the local newspapers were covering
the story. John Malone, the executive who had once refused to even take calls
from MTV, caved within a week and added the channel to the TCI lineup.
And MTV would then move on to the next city and prompt the public there
to besiege their cable operator with demands for “their MTV.”
As a format, MTV changed the relationship between the industry and
its fans. Every recorded music shift required mass-market acceptance to be
successful. But in the case of music videos, the public’s role was more than
just enjoying their favorite songs in a new configuration. They played a far
more active role by expressing their preferences loudly and repeatedly. The
Generation X teens who loved the network even came to be referred to as
the “MTV Generation.” They believed their opinions mattered and that the
entrenched business interests should pay heed to their collective voices. We’ll
discuss in Chapter 8 how improved network connectivity and the powerful
software of the Internet era would prove to be even more potent catalysts in
upending the business.
“I want my MTV” was certainly catchy, but the words themselves would
not have been enough to persuade music fans to advocate for a channel they
may have heard about but had never experienced personally. Motivating the
public to take action required communication that commanded their atten-
tion. And no one could elicit a greater response from fans than their mu-
sical idols.
others would consent to such arrangements, but only for the right (very
large) price. MTV’s challenge was the fact that their talent budget for ads
equaled their acquisition budget for music videos: zero. Getting even one su-
perstar artist to make a commercial without writing a check with lots of zeros
seemed like an impossible task.
Nevertheless, Les Garland, Executive VP of Programming for the channel,
took on that mission and flew to France to meet with Mick Jagger of the
Rolling Stones. Garland told Jagger that he wanted him to go on camera and
say “I want my MTV” to promote the channel and its music videos. Jagger
explained that the Stones did not make commercials. Garland pointed out
that the Stones did indeed accept tour sponsorship from companies. Jagger
agreed but said the Stones got “paid a lot of money for that.” Of course,
Garland had no such budget to draw on, and he told that to Jagger. Instead,
Garland put a single dollar bill on the table. Surprisingly, Jagger agreed. The
commercial was shot the next day.
Getting Jagger on board not only gave MTV a rock legend to pitch the
channel to fans, it gave the entire advertising effort the credibility to attract
other artists under the exact same arrangement: no compensation. The net-
work reached out to Pat Benatar and told her Mick Jagger was doing the
ads; she followed suit. The same argument worked for David Bowie, Pete
Townshend, The Police, and Cyndi Lauper, to name just a few. Sony Music
CEO Tommy Mottola said, “The campaign gave MTV integrity and credi-
bility with the audience; their favorite band or icon was endorsing this brand
and telling them this was the thing they should watch.” The fans took those
instructions to heart and loudly demanded their MTV in phone calls and
letters. The cable operators complied. Group W, the operator covering the
northern half of Manhattan, added MTV in time to air the 1983 New Year’s
Eve Rock n’ Roll Ball. That same company also controlled the Southern
California market; three weeks later, they flipped the switch there.xxiv Music
fans in the major media centers in United States, along with millions of
others nationwide, had done as their idols had commanded, and now they
had their MTV.xxv
Although music videos are an essential element of virtually every record
release today, at the start of the MTV era many artists resisted the pres-
sure to appear in them. Billy Joel saw himself as a piano player and not an
actor; he hated making videos. Elton John might have been the consummate
showman on stage, but he felt that he photographed poorly, so he avoided
shooting videos as well.
172 Key Changes
Van Halen was another band that held negative views about MTV. They felt
that music videos were simply beneath them, but rather than refusing to par-
ticipate, they elected to undermine the process. When they needed a video
for their cover version of Roy Orbison’s “Pretty Woman,” the band chose to
take on the characters of a cowboy, a samurai warrior, Tarzan, and Napoleon.
That could have been the basis for an entertaining video. However, the band
chose to build the story around two little people reaching up the skirt of a
young woman whose hands were tied to two posts while the events were
observed by a hunchback. Van Halen pushed the envelope further by ending
the video with the girl removing her wig to reveal that she was a man. What
was most surprising for the early 1980s is that the network actually aired that
video once or twice before banning it completely.12
Bruce Springsteen wanted to focus on “writing songs, making records, and
doing concerts,” and there was no room within those three priorities for music
videos. When his album Nebraska was released in 1982, Columbia Records
put together a music video for “Atlantic City.” The video met with Springsteen’s
approval because he did not appear at all in the roughly three-and-a-half-
minute run time. By 1984, MTV’s power to promote new songs became
too potent to resist, even for the Boss, and he can be seen pulling a young
Courteney Cox from the audience to join him on stage in the video for
“Dancing in the Dark.” That song became Springsteen’s highest charting
single up to that point in time.
Not all creators viewed making music videos as a burden or a process to
be subverted. Some bands used the production budgets to live up to the rock
stereotype. They enjoyed casting bikini models to appear as eye candy in
the videos and made sure that everyone on set was well supplied with a va-
riety of illegal substances. That’s not to minimize the use of attractive women
in music videos to appeal to much of the young and male MTV audience.
Whitesnake furnished just such an example later in the 1980s. The British
hard rock band had built a successful following in Europe, but it had been
unable to achieve meaningful sales in the United States. The music video for
“Here I Go Again” featured actress Tawny Kitaen—the future wife of singer/
bandleader David Coverdale—doing front handsprings across several Jaguar
XKEs. MTV put the video into heavy rotation, and the album featuring the
song broke through in America. It made it to no. 2 on the Billboard charts
and eventually went eight times Platinum.xxvi
Of course, there were other artists who saw videos as opportunities to
make more meaningful artistic statements. Devo is a compelling case study
Television 173
of “rock videos,” and soon scenes were recut to become exactly that.xxix
“Maniac,” by Michael Sembello, featuring Alex in a training montage, was
shown repeatedly on MTV; it climbed to the top spot on Billboard’s Hot
100 chart as well. From that point on, music videos based on songs from
soundtracks became an integral part of MTV’s on-air inventory. In addi-
tion, the advertisements for those same films became a major source of the
network’s revenues.
The one artist whose career epitomizes music videos that marry amazing
vocal performances with ground-breaking choreography is Michael Jackson.
Yet he, like many other Black artists, was rarely seen on MTV in its early
years. Opinions differ on the cause of the channel skewing its airtime toward
white acts. MTV executives denied any claims of prejudice. They pointed to
their background in Album-Oriented Radio (AOR) and their reliance on au-
dience research guiding the decision to focus on rock music, which excluded
Black artists as a result. They believed that pushing the channel into R&B (or
country or folk, for that matter) would alienate a significant portion of their
audience that preferred rock. That was the rationale that Mark Goodman,
one of the MTV on-air hosts (known as VJs), put forward when David Bowie
turned the tables on him during an interview and asked why there was a pau-
city of Black artists on the channel.
Some artists were vociferous in their objections to MTV’s decisions to pass
on videos starring Black artists. After the network turned down the video for
“Super Freak,” Rick James complained publicly that excluding him and his
peers such as Stevie Wonder, Smokey Robinson, and Earth, Wind & Fire was
damaging to their record sales. He called the network “racist.” Soundies had
already demonstrated the appeal of short films of musical numbers featuring
talented Black performers. But like the Hollywood studios of that earlier era,
MTV chose not to take advantage of those artists.
Michael Jackson’s Thriller was the top-selling album, and “Billie Jean” was
dominating the singles charts, in 1982. Despite these impressive results,
MTV refused CBS Records’ requests to air the video for two full months
after its release. The label had the contractual right to pull its music videos
from the channel with 24-hour notice, and, according to David Benjamin,
VP of Business Affairs for the label at that time, they decided to exercise
that clause unless MTV changed its decision. With 25 percent market share,
CBS Records possessed a big hammer. Both Pittman and Garland refused
Benjamin’s call, but he did get through to John Sykes, VP of Programming
at MTV, and instructed him to remove all the label’s videos. Benjamin
Television 175
immediately headed over to CBS Records CEO Walter Yetnikoff ’s office to in-
form him of the step he had taken. By the time he walked around the hallway
and reached his boss’s door, the secretary was waving him in. Yetnikoff was
on the phone, and he told Benjamin that he was talking to his “best friend”
Bob Pittman who had just said that MTV was going to start playing “Billie
Jean.”xxx MTV executives deny that this showdown ever occurred and say
they always planned to add the video to their rotation. Regardless, Michael
Jackson’s “Billie Jean,” followed by “Beat It,” were both enormous successes
for MTV. The popularity of Jackson’s videos proved that the rock-centric
focus was far too narrow. He cleared the path for other Black artists to be
included on the channel, and despite its recalcitrance, MTV benefited enor-
mously as well.
It was Jackson’s next video, however, that demonstrated the artistic po-
tential for the medium most clearly. Intrigued by John Landis’s direction
of American Werewolf in London, Jackson contacted him to make the video
for the title track from Thriller. Jackson and Landis together crafted a nearly
14-minute-long musical horror film that required a budget of $900,000, far
higher than any other music video up to that time. Yetnikoff begrudgingly
agreed that CBS Records would help to fund the video but capped its con-
tribution at $100,000. The creators decided to film a “Making of ” documen-
tary along with the video, creating a one-hour special that could be promoted
as an event. They convinced the cable network Showtime to pay $300,000
for pay-cable rights; and MTV, in a significant exception to its policy of not
paying for the creation of videos, kicked in $250,000 for exclusive rights on
basic cable. Jackson covered the remaining budget personally.
Given those investments, the expectations for “Thriller” were high, yet the
reactions managed to exceed them all. MTV and many other media outlets
rate it as the greatest music video of all time. It created a template for music
videos to become more sophisticated in terms of both storytelling and orig-
inal choreography. The “zombie dance” became a cultural touchstone with
fans and is still imitated around the world, particularly on Halloween. Sales
of Thriller were reinvigorated after the video release, more than doubling to
reach its current position as the top-selling album of all time worldwide.13
Its sales were ultimately twice as large as those of Saturday Night Fever, an-
other movie soundtrack whose sales were propelled by the disco craze and
the videos of John Travolta dancing to the Bee Gees’ music. The Thriller VHS
tape, which combined the music video and the documentary, sold over a mil-
lion copies. On the other hand, the acclaim for Thriller was not universal: the
176 Key Changes
inspired by the Fritz Lang film Metropolis, had a total budget of $5 million.
It was the most expensive music video ever to that point in time (and re-
mains, today, the third most costly music video ever). “Vogue,” with its black-
and-white fashion magazine glossiness and dance club vibe, cost only $5,000.
David Fincher directed both and established his credentials as a directorial
talent, which ultimately led to Fight Club, The Social Network, and three
Oscar nominations.
Although we have focused on Michael Jackson and Madonna, many
other artists expanded the boundaries of what could be expressed via music
videos in genres from pop to hip-hop, including a-ha, Peter Gabriel, Jay-Z,
Dr. Dre, and, of course, Beyoncé. Bob Pittman pointed out that the creativity
embodied in music videos changed the live performance business as well as
audiences. For the stars on MTV, a concert could no longer consist of the
band members on some risers surrounded by speakers with a few spotlights
and dry ice. Instead, they needed to lift the production values to a level where
the concert experience compared favorably with the images ingrained in the
audience’s memory by the video.
Yet in spite of some artists’ creativity, there was an underlying concern in
some quarters that the growing importance of MTV emphasized surface ap-
pearance and style to the detriment of more “serious” musicianship. There
were certainly artists such as Paula Abdul and Jennifer Lopez who initially
relied more heavily on their dance skills rather than their musicianship as a
calling card to build a reputation. Perhaps the most extreme example of style
winning out over musical substance was when Milli Vanilli, with MTV’s un-
stinting support of their single “Girl You Know It’s True.” The song won the
Best New Artist Grammy in 1990, and the album sold millions of copies. It
all came crashing down when the press reported that the photogenic duo had
not recorded their own vocals on the record. It was the first and only time
that the Grammys ever rescinded an award.
At its inception, the “cost of goods” for music videos, MTV’s “raw
materials,” was ostensibly zero. (Although it did not pay the labels for the
content, it did cost MTV about $1,000 for each track to clean up the audio
and to transfer to the one-inch tape needed for broadcast.)xxxii The more suc-
cessful MTV became, the greater the likelihood the labels would seek to alter
178 Key Changes
what they increasingly believed were overly generous business terms. Yet it
would take a couple of years for the right moment to come to seek changes.
By the beginning of 1983, with the NYC and LA markets finally covered,
MTV garnered a new degree of respect and influence. At the end of January
of that year, Billboard added music video plays to the industry metrics in
its magazine, making it clear that the format was an important element in
promoting and marketing music and artists. Like many Internet companies
of a later era, MTV started out by building a recognizable consumer brand
and cultural currency using unpaid content and did not reach profitability
for quite some time. In its first two years, the network lost $34 million.
Several critical success factors fell into place in 1983. MTV came into the
year reaching 18 million homes; that number, paired with those desirable
youthful demographics, made advertisers sit up and take notice. Soon 140
companies were buying ads on the station pitching more than 240 different
consumer products. (Sixteen of those companies were advertising jeans.) The
tables had turned: instead of pleading with the cable operators to offer MTV,
the parent company WASEC was now able to start demanding payments
from those same operators for the right to carry the station.
Meanwhile, the music industry slump that began in the late 1970s
subjected every expense to greater scrutiny. As we’ve discussed, improved
sales driven by video play demonstrated that the labels could reach record
buyers more cost-effectively via MTV than by supporting a tour. In addition,
the labels had no influence over artists’ performances on the road, but they
could exert at least some control over the music video budget, content, and
director.xxxiii
We’ve already cited several examples of videos that required extraor-
dinary investments to realize a particularly creative vision; these were less
cost-effective despite the huge sales numbers that the videos helped the songs
pull in. As artists began to see what others were doing, even typical music
videos grew more costly as they embodied more ambitious stories and relied
on higher production values. In 1981, the average music video cost approxi-
mately $15,000 to produce. Two years later, that increased to $25–35,000 and
yet again to $40–50,000 in 1984. And despite the rising costs, the labels were
producing so many videos that MTV could become more selective and de-
cide which tracks received the heaviest rotation and which ones did not air
at all.
As the costs climbed, the labels found a way to reduce the bottom-line
impact: they modified their accounting practices to treat video budgets
Television 179
as recoupable expenses, meaning that the costs were deducted from gross
revenues before profits were paid to the band. This effectively raised the
break-even point for any record release and delayed the time when roy-
alty payments would reach artists. The bands that pushed for a party at-
mosphere at the video shoot were effectively footing the bill for their own
entertainment.
Even with these accounting machinations and the positive returns from
video play, the labels felt the bargain they had struck with MTV to display
their content without compensation was fundamentally unfair. Tommy
Mottola, former Chairman of Sony Music, said of MTV, “They built the
biggest music enterprise in the history of the world off of our backs, off of our
money, off of our sweat.”xxxiv When the channel turned a profit in 1984, the
labels demanded a new deal.
Larger corporate issues were motivating MTV to finalize deals for access
to the music videos that were the lifeblood of its business. The Warner-Amex
joint venture that owned MTV had lost close to $100 million in 1983. To raise
more money, it needed to offer stock, and MTV’s newly profitable posture
had become a linchpin of a successful offering. To make Wall Street com-
fortable with its long-term prospects, MTV needed predictable economics—
which translated to having a deal with the labels.xxxv
Pittman knew that with MTV’s healthier financials, the free ride was over.
The channel would have to pay the labels to air the videos that had become
increasingly costly to produce. However, the network’s demonstrated power
in building artist brands and driving sales had made it invaluable to the mar-
keting teams at the labels. The record companies had become addicted to the
promotional value of MTV, and that gave the network significant leverage in
the negotiations. MTV used that position of power to focus on garnering ex-
clusive rights as a way to fend off any potential competitors.
The first proposed agreement with Capitol/EMI would have required
MTV to pay the label $1.25 million over three years. In exchange, 35 percent
of the label’s videos shown on the network would be exclusive to MTV for
30 days. MTV would select two-thirds of the videos covered by the exclu-
sivity clause, and Capitol would choose the remainder. This clause gave the
channel the ability to ensure that the most desirable songs would be aired on
MTV, and only MTV, for a month. At the same time, MTV’s demonstrated
promotional value made the labels anxious to have the right to select some
of the videos that made it into MTV’s rotation. The labels’ own marketing
priorities took precedence over following the preferences of the audience.
180 Key Changes
Cable television did not have the same type of payola landscape as radio did,
as we saw back in Chapter 3.
Capitol/EMI ultimately rejected these terms, but they established the
framework for later deals with the other record companies.xxxvi By September
1984, contracts had been signed with CBS, RCA, MCA, Geffen, Polygram,
and the Warner Music labels. These provided exclusivity for “television exhi-
bition” for one to two months covering 20–40 percent of each label’s videos
in exchange for differing cash amounts. For example, market share leader
CBS received $8 million over two years, a sum that the label projected would
cover the production expenses for the 200 videos they made per year. MTV
Networks was incorporated as a separate entity, and that new company for-
mally acquired both MTV and Nickelodeon from WASEC. Warner-Amex
gave up some financial control through a successful stock offering while re-
taining 90 percent of the voting control.
Al Teller, then President of CBS and Columbia Records, had recommended
to his boss, Walter Yetnikoff, that the label should pass, but he was overruled.
The labels got the cash and secured the right to place some videos on MTV.
In exchange, they gave MTV the exclusivity that would cement its market-
leading position and make it even more powerful down the road. As Teller
said, “The music industry has a long history of doing incredibly stupid things
at important moments in its history.”
MTV’s clout would be reduced with competition. And the obvious pop-
ularity of music videos attracted interested parties. The creators of the
Financial News Network, which included Wall Street powerhouse Merrill
Lynch, started efforts to launch the Discovery Music Network (DMN) to air
on cable stations nationwide. NBC launched the show Friday Night Videos,
and WTBS, Ted Turner’s Superstation, introduced a weekend show called
Night Tracks in the summer of 1983. The success of the latter led Ted Turner
to plan for a new station dedicated to music videos to go head-to-head
with MTV named, unimaginatively, the Cable Music Channel (CMC). The
channel would be offered for free to cable operators to undermine MTV’s
efforts to extract carriage fees from those same operators. All these initiatives
saw MTV’s narrow focus on a teen audience as an opportunity and intended
to offer a broader range of music along with tighter standards on the display
of sexuality and violence.xxxvii
CMC launched late in October 1984, and, in one of the quickest flame-
outs in the history of media, it was shut down approximately one month
later. Not enough cable providers signed up to carry the channel, and it was
Television 181
bleeding buckets of red ink. Many operators simply did not have room for
yet another music video channel, while others resented what they felt were
Turner’s strongarm tactics in promulgating his cable news channel, CNN.
Turner sold the assets of CMC to MTV’s parent company for $1 million and a
commitment to purchase $500,000 worth of ads on other Turner properties.
The CMC assets included satellite bandwidth that gave MTV the technical
platform to launch a second music channel. That new station was created
in part to head off the competitors with a focus on a broader range of music
for an older demographic. That sister station launched in January 1985 as
Video Hits One, or VH-1 (shortened to VH1 in 1994). The other potential
competitors were not much more successful in terms of market launches, but
they were very active in the courts, as we will discuss later in this chapter.xxxviii
Despite a successful stock offering and MTV’s momentum, Warner and
American Express began looking for a way to sell their holdings in MTV
Networks. Warner was in a deep hole because of the ill-advised purchase of
the video game company Atari. AMEX’s rationale for investing to get into in-
teractive services had proved to be illusory in the 1980s, and they found cable
programming to be an expensive proposition. In 1985, Viacom purchased
the two companies’ interests in MTV Networks in a complicated transaction
for $525 million.xxxix The founders had been hoping for a leveraged buyout
that would grant them major equity stakes in the company, but instead they
simply became employees at another large media company. MTV, of course,
continued to thrive, and its valuation climbed into the billions. Steve Ross,
the head of Warner Communications said selling MTV was “the biggest re-
gret of his career.”xl,14
The strength of the MTV brand and the cultural cachet it held with young
people were incredibly valuable assets that the network used repeatedly to its
financial advantage. They allowed it to expand to international markets. As
Bill Roedy, Chairman of MTV International, said, “ ‘A-wop-bop-a-do-bop,
a-wop-bam-boom’ [sic] means the same in every language.”xli The network
created valuable, repeatable programming that was an extension of its brand,
such as MTV Spring Break, Total Request Live, and the MTV Music Video
Awards. Labels competed to get slots for their artists on these platforms.
The logical next step was to create programming that reached that same
audience but did not rely on music videos. MTV could own the intellectual
property itself and improve its margins. Reality shows such as Real World
and animated programs such as Beavis and Butt-Head proved popular and
did not require any payments to labels. The original premise of the network
182 Key Changes
as music videos all the time disappeared. As Nick Rhodes of Duran Duran
said, “At some point, the M in MTV changed from Music to Money.”xlii
The copyrights associated with music videos are complicated and entail
several different permissions and payments. Using music in a timed relation-
ship with visual elements in a television show, motion picture, commercial,
or music video requires a synchronization or “synch” license.15 The license
requirement covers both copyrights embodied in a particular rendition of a
song: the recording and the composition. The performer on the recording,
or the label holding the rights to that version, must authorize the new crea-
tion, as must the composer who wrote the song or the music publisher that
holds those rights. When an entity performs that resulting creation publicly,
including airing it on a cable station, it must pay a performance royalty to the
songwriter or composer as compensation for that use.xliii
Synchronization rights are not covered by any blanket arrangements or
statutory (government-set) licenses, nor are they even covered by any widely
accepted industry conventions. As a result, synch licenses typically require
separate negotiations with each of the rights holders whenever someone
wants to create a new audio-visual work. Given the instant recognizability of
a specific certain recording, getting the approval to use it for a film or a com-
mercial can be an expensive and time-consuming proposition. For example,
when Microsoft launched Windows 95, it decided that the company’s very
first television commercial should have music that emphasized the start of a
new era as well as the start button on the launch screen. Bill Gates wanted to
use the song “Start Me Up” by the Rolling Stones, and he negotiated the synch
rights personally with Mick Jagger. It required a payment of $3 million for
the 60-second commercial.16 In cases like this, the hit song’s inherent value
created a “halo effect” for the advertised product.xliv
However, for a new release, even by an established artist, the music video
was intended to create demand for a song that was not yet known. In this
case, when the label decided to make a music video for an act on its roster, it
treated the video as a “promotional” use. The label’s contract with the artist
typically included the authority to synchronize any recordings in that manner
without additional compensation. Unlike a commercial use in an advertise-
ment or a movie, that promotional designation obviated payments for the
Television 183
the case was settled out of court and, after buying out Wodlinger, Viacom
closed Hit Video USA altogether.
Exclusivity with record labels was an ongoing source of concern to other
channels offering music videos even when it was not the main focus for their
programming. In 1991, Black Entertainment Television (BET) programmed
music videos into its schedule but was frustrated by those clauses. It began a
boycott of all MCA Records artists to protest MTV’s exclusive rights to “Now
That We’ve Found Love” by Heavy D and the Boyz. Though the terms were
kept confidential, BET reached a settlement with MCA Records, and Heavy
D’s new video continued to be aired by MTV alone.
MTV was not done fighting antitrust battles, but when the next one arose
in the early 1990s, it was the one complaining about unfair treatment. As in
the United States, airing music videos in the United Kingdom and Europe re-
quired both rights to the videos and the payment of performance royalties. In
the United Kingdom, Phonographic Performance Limited (PPL), owned by
the major and independent record companies, licenses music to broadcasters
on their behalf. As MTV began its international expansion in mid-1984, PPL
formed a sister company called Video Performance Ltd (VPL) to license
music videos for broadcast on television. It also had reciprocal arrangements
with other European countries.
In the United States, MTV had started out without making any payments
to the labels, and when the time came to change that posture, they had
the advantage of “perfect information” on the deal terms with each of the
labels. The labels had no intention of providing that same sweetheart deal
to the now much more powerful MTV in Europe. Under UK law, the labels
were permitted to empower VPL to negotiate as a single representative of
all the member companies. VPL’s deal with MTV required 20 percent of
the channel’s top-line revenues to be paid to the labels for the use of music
videos; that does not include the payments for performance royalties. MTV
reaped the benefits of its dominant market posture in the United States, but
when the labels followed a similar playbook in Europe, the network did not
appreciate being on the receiving end. The head of MTV Europe considered
the label deals untenable and described them as a “holdup.”xlvi
Despite concerted efforts to get just one of the labels to break rank and
to negotiate individually, MTV was unable to do so. Ultimately, it decided
it had no choice but to take legal action against what it called the labels’
oligopoly. First, it lodged a complaint with the European Commission ac-
cusing VPL of monopolistic practices and overcharging for music video
186 Key Changes
rights. The labels held firm, and MTV decided to file a lawsuit in the United
Kingdom. A preliminary order from the European Commission required
the major labels to negotiate separate deals and made it clear that this would
be its final decision as well. The labels came to the table individually, and
the court case was rendered moot. But while the legal battle was underway,
four of the major labels created a local music video channel in Germany,
called Viva, to compete with MTV’s pan-European channel. The German-
language Viva proved popular and created the first meaningful competition
to MTV in Europe, which reacted by offering more targeted channels in the
local language.xlvii
Conclusion
MTV did not fundamentally change how the industry made money, but it
proved to be an accelerant. MTV’s promotional value became an alterna-
tive to radio and touring in building an artist’s image and brand. It boosted
revenue streams from music purchases and ticket sales. In fact, including
visuals helped to establish artists’ careers more quickly and effectively than
songs alone.
Nor did MTV directly cause any significant shifts in copyright law. But the
business and legal dynamics between the labels and MTV had repercussions
down the road. The music industry learned what it thought were valuable
lessons, but they did not play out as hoped. This would not be the last time
that the labels would enable a fledgling company to use their artists’ music
to build a powerful and valuable enterprise that could then use its newfound
leverage to its advantage.
When the labels faced a similar scenario again, the MTV experiences
led them to several major shifts in their approach. First, they decided that
startups needing music licenses, even those with minimal financial re-
sources, would be expected to pay artists from the start. Second, the labels
expected those payments to include a substantial upfront guarantee to pro-
tect against any potential downsides. Third, they would request, and often
receive, an equity stake in any new corporate entity.
However, these practices had unintended consequences. They created
disincentives for investors in new companies that slowed down the de-
velopment and growth of innovative copyright-respecting services. That,
in turn, created openings for services that launched without licenses,
Television 187
The Audio Compact Disc (CD) is the biggest-selling audio format in terms
of revenues—by a wide margin—of all time. That much is beyond dispute.
But what’s less clear is how it got its size and recording length. The CD’s tech-
nical specifications are spelled out in the “Red Book” standard, one of a series
of technical standards related to the format.1 Sony and Philips collaborated
to author that standard, which was issued in 1980. The document spelled
out that the CD would have a 122 mm diameter and would contain a max-
imum of 74 minutes of music. Researching what led to this specification
yielded multiple different versions of the “true” story. Like the Japanese film
Rashomon,2 the participants each have their own version of the events.
The “official” history on the Philips website explains that the Sony vice
president in charge of the project, Norio Ohga, insisted on that length to
guarantee that Beethoven’s Ninth Symphony as conducted by Herbert von
Karajan would fit on the disc in its entirety because that was his wife’s favorite
musical selection. According to Kees Immink, a key innovator in digital
media who served as a leader on the Philips team, this account is a reminder
not to believe everything you read on the Internet. In one article, Immink
states the top Philips executives picked the size to be only marginally larger
than the successful Compact Cassette to strengthen its appeal to consumers.i
Rather than putting the fans or the art first, two other explanations sug-
gest that mundane business considerations ruled the day. Polygram, owned
by Philips, already had an experimental manufacturing plant capable of
producing large volumes of 115 mm discs containing up to an hour of music,
that hour being the original target spec for the format. Sony lacked a sim-
ilar production capability. By insisting on the longer playing time—which,
in turn, required a larger disc—Sony effectively blunted Philips’s competitive
advantage in production capacity.ii
Marc Finer, who was part of the Sony Electronics communications team
that supported the launch of the CD, recalls yet another explanation. Philips’s
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0007
190 Key Changes
preference was based on the 14-bit word for each digital sample of the analog
audio because that was the capability of the digital to audio (D-to-A) con-
verter they had available. Sony had a 16-bit D-to-A converter that provided
superior quality but required additional data, and it insisted on a larger disc
as a result. According to Finer, Sony’s view won out and the larger disc was
the result.iii
How the companies arrived at the final CD size may remain open to de-
bate, but the exact number of minutes of music on the disc turned out to be
far less crucial than it seemed at the time. Other factors led to the format’s
unprecedented success, and technology improvements eventually extended
the disc capacity to achieve an 80-minute maximum.
The first 50 commercially released CDs reached the Japanese market in
October 1982. The list of titles was led by Billy Joel’s 52nd Street, a chart-
topping record that won the Grammy award for Best Album in 1980 and, not
coincidentally, was issued by Columbia Records, a subsidiary of Sony Music.
That introduction represented the dawn of the most popular physical music
format ever. The financial rewards reaped by Philips, Sony, and every other
music company were beyond anything that the industry had ever seen be-
fore or since. During the 1990s, as the CD format contributed billions to in-
dustry revenues, the struggle was not how to make money but simply how to
count it all. That success may seem inevitable in hindsight, but it required a
confluence of technological innovations and collaboration across the various
players in the industry to make it happen.
Figure 7.1 J. P. Sinjou Introduces the Compact Disc, March 1979. NBC News.
Minimum value
is –32,768 CD audio has
44,100 samples per second
microns wide, which is less than 1/100th the thickness of a human hair or 1/
25th the width of the groove in an LP. Needless to say, controlling a laser so
precisely was itself one of the engineering advances required to make optical
media even feasible. The “write” laser is far more powerful than the laser in-
cluded in every CD player. It must be capable of modifying the material to
create the pits and lands, while the “read laser” simply bounces the light over
the surface of the disc to detect them.
Manufacturing CDs in volume requires tightly controlled processes
as well. Like vinyl records, the master copy is used to “stamp out” the final
products. The high-grade plastic used for those final products needs to be
exceptionally pure and the production facility clean enough to ensure that
dust particles are not introduced. A reflective layer is added so that the laser
bounces back when the data is later read off the disc. Then a lacquer layer is
added to prevent oxidation and to allow for artwork to be printed on top.
The processes to extract the data at playback are just as carefully designed.
The “read laser” included in each player must be precise in terms of locating
the beam properly as the disc spins to obtain an accurate reading of the infor-
mation. The beam comes from beneath the spinning disc and reflects straight
back when it encounters a land but scatters when it sees a pit, which yields the
binary information needed.
Playback of optical discs is, in many ways, the antithesis of the process of
playing vinyl records. For optical discs, instead of starting at the edge and
spiraling inward toward the spindle, the CD laser reads the data starting at
the inside edge and moves outwards. And instead of rotating at a constant
speed of 33 1/3 or 45 rpm, the CD spins at a dizzying speed of 500 rpm but
then gradually decreases to 200 by the time the laser is reading the data
from the outermost edge. That guarantees that the disc moves at a constant
linear rate, which ensures constant frequency response throughout the en-
tire playing time—as opposed to LPs, which, as we saw in Chapter 4, limit
frequency response as the stylus makes its way toward the center. The data
extracted from the disc flows through the rest of the decoding process at an
unvarying rate, simplifying the conversion back to analog signals.
These stringent manufacturing procedures yielded an exceptionally low
error rate for the information stored on the disc, but given the hundreds of
millions of bits involved, the end result could not be flawless. To compen-
sate for imperfections, two additional pieces of data were placed on the disc
at every sampled instance. These bits could be examined at playback to de-
termine the correct information when the audio information went missing
Compact Discs 195
or was damaged. In addition, the data was recorded out of exact chronolog-
ical order to make the results robust to even larger defects, such as scratches.
Sony contributed these sophisticated error-correction techniques, known
as Reed-Solomon coding, to the project. Reed-Solomon coding was first
deployed by NASA a few years earlier to ensure reliable communications
with the Voyager spacecraft as it explored the outer planets of the solar
system. The CD was the first consumer product to take advantage of these
techniques.v
CDs initially targeted a one-hour playback time of audio, compared to
two hours of audio and video for LDs. The smaller data capacity meant that
the disc could fit comfortably in one hand, in contrast to the larger LP-sized
platter of the LD. This choice meant that as technology improved, the sizes of
players could be reduced to allow for automobile and portable players. As we
saw in Chapter 5, the success of Sony’s Walkman demonstrated how valuable
portability could be to the music experience; Sony kept working quietly to
miniaturize the electronics.
Sony released the first commercial CD player, the CDP-101, in Japan in
November 1982; worldwide launch followed in March 1983. As Figure 7.3
shows, this was a stereo component, a bit smaller than cassette decks of the
time, designed to be connected to a stereo amplifier or receiver.vi
The Sony D5, the first CD player that could be deployed in vehicles, hit the
market late in 1985.vii It was only a little larger than the jewel case, though the
external battery was the size of a shoebox. Further improvements meant that
Figure 7.3 Sony CDP-101, the first commercial CD player. Used by permission
of Sony Electronics Inc. All Rights Reserved.
196 Key Changes
The CD logo did more than simply name the product. The logo conveyed
that the product was more than “compact” like the cassette already familiar to
music fans. It also highlighted that the product contained “Digital Audio.” In
the early 1980s, as the world experienced the burgeoning personal computer
revolution, the word “digital” conveyed the idea of advanced technology and
modernity all by itself. It also supported the superior quality message that
was the core of the value proposition to the average music fan.
The tagline that accompanied the logo was “Pure, Perfect Sound Forever.”
That message targeted the precise pain points that the average music fan
had become all too familiar with when dealing with their vinyl records or
cassettes. The new format was not only about the perfection of sound but also
about durability and robustness over the life of the product.
Listening on the go with a cassette on a Sony Walkman was convenient,
but the quality of the sound was clearly inferior to vinyl. There were issues
with the tape snarling or tearing, and rewinding loose tape back into the car-
tridge using a pencil placed in the hole in the cassette was a “fix” that most
consumers found necessary on many occasions. Meanwhile, vinyl records
skipped, warped, scratched, and broke. Simply listening repeatedly to the
same album gradually degraded the sound because of the friction between
the needle and the record. With the CD, the laser never touched the disc;
it read the data encoded in the underlying material without physical con-
tact. The laser performed the same way every time and did not wear down
like the phonograph needle stylus that needed (though rarely got) frequent
Compact Discs 197
replacement. Every play produced the exact same sonic experience as the
first listen.
Unlike vinyl, dirt on the disc did not interfere with the playback, though
you did need to be careful about fingerprints. In addition, the physical struc-
ture of the disc, though not indestructible, was far more resistant to scratches
and did not warp or break easily. Even the hissing sound present in magnetic
tape or created by the needle moving over the surface of the grooves on a
vinyl record was eliminated. The CD’s ability to play all the tracks sequen-
tially from start to finish eliminated the need to get up from the couch after
25 minutes or so to flip the vinyl from Side 1 to Side 2. For couch potatoes,
this was the best enhancement to an entertainment product since the inven-
tion of the TV remote.
The zeros and ones that constitute the music on the CD are imprinted as a
continuous stream of information from the first bit to the last. There are no
physical breaks or spaces between the songs. Instead, the data begins with
a “Table of Contents” (ToC), a set of metadata that includes the title of the
disc as well as the length of every track. That allows consumers to play any
track they choose in any order. The hardware uses the ToC data to find the
exact spot for the laser to begin. Of course, you could play any song on a vinyl
record if you were willing to make the effort—and had the skill—to drop the
needle on exactly the right spot without damaging the record. Cassettes were
even less amenable to random access: although making mixtapes from songs
on the radio or your record collection was popular, it required a significant
investment of time and effort.
Empowering fans to select the playing order of the songs they want to hear
with the press of a button seems like just one more small selling point on
a lengthy list of consumer benefits for the CD. Many fans didn’t even care
enough about this feature to bother using it. The fan always made the de-
cision to choose the recording from their collection that they wanted to
place on their turntable, in their Walkman, or in their CD player. But once
they made that choice, the audio journey they took was, for all intents and
purposes, pre-determined by the album’s creative team.
Yet this new feature was a harbinger of the complete dismantling of the
album as a unified entity controlled entirely by the creators. The ability to
craft and to release an end-to-end thematic experience, as exemplified by
the concept albums that we saw in Chapter 4, did not disappear. But the
artist’s control over the final experience had been compromised. The fan’s
ability to define their own preferred version of that audio experience was a
198 Key Changes
One of the first issues to resolve was the packaging for this new product.
Record retailers had invested millions in display fixtures to show off those
12-inch vinyl records. They were not anxious to spend a substantial amount
to change their shelves to fit the smaller size of a new unproven format.
Not to mention that those 5-inch CDs would be much easier to pocket and
steal, increasing inventory “shrinkage.” After several months of negotiations
among the labels and the retailers, a clever but wasteful compromise was
reached: the “longbox” package, which measured 6 by 12 inches and was
made of plastic and cardboard. It contained a space for the disc, stored in a
“jewel case” in the lower half of the box, and another space for the booklet in
the upper half. Two longboxes could sit side by side in the existing spaces for
albums, making it easy for shoppers to look through the selection of discs
available. And making the package much larger than actual physical media
acted as a deterrent to theft.xi
The packaging created a public uproar because of the large amount
of extra trash generated as CD sales climbed. To quell the uproar over the
waste, the packaging eventually evolved to a plastic box of the same size that
contained the CD-sized case, which was opened at the cash register and
reused. Shoppers who were playing by the rules did not see the extra step of
unlocking the case as a particularly friendly experience. Ultimately, record
stores revamped their displays to fit the CD jewel case by itself and attached a
magnetized detector to each one for security.
Enabling record stores to display the CD without a major investment was
a good first step, but making potential buyers flip through longboxes did
little to convey the benefits of the new format. At electronics stores, TVs and
VCRs were the priorities while audio products were just not considered cool.
Beginning in 1983 and carrying over into 1984, the Compact Disc Group
put together a display program that featured Sony CD players and bundles
of discs to demonstrate this new experience for potential buyers. In addition,
it instituted a program for 50 radio stations around the country to promote
the product. Every time these stations played a song from a CD on the air,
they would announce that the music was coming from a Sony CD player. Of
course, the music sounded exactly the same as it always had over the radio
airwaves,5 but the marketing message was delivered just the same.
As sales of players and CDs started to grow, the promotions to support
the channels and to stoke interest scaled up as well. They obtained “product
placement” in movies and TV shows to show off the new format. In 1985,
the music companies and Bose worked with MTV to enable a giveaway that
200 Key Changes
included hundreds of players and thousands of discs. MTV “VJs” were re-
quired to hype the giveaways, so there was extensive on-air coverage to a na-
tional audience of young consumers. The program was tied into live MTV
broadcasts from Daytona Beach for Spring Break for the first time; it was so
successful that the annual event became synonymous with the music net-
work. And in echoes of Edison’s phonograph demonstrations decades earlier,
the CD proponents had bands surreptitiously shift from live performance to
lip-synching to a digitally recorded performance to prove audiences could
not hear the difference.xii
The early years of the CD coincided with the creation of the electronics
superstore. Wards Company, which had started as a TV and appliance chain
in Virginia, expanded its warehouse concept and morphed into Circuit City
in 1981.xiii The Sound of Music, a small chain of high-end audio retail stores
based in Minnesota, expanded its product line to include home appliances
and VCRs, and in 1983 rebranded itself as Best Buy. As the CD’s popularity
grew, these stores found that popular new releases drew customers who
might be persuaded to purchase other items. To fuel that traffic, the stores
created advertising inserts in the weekend papers highlighting the extensive
selection of CDs at discounted prices.
Circuit City and Best Buy were not alone. The draw of those discounted
CDs was so compelling that other major retailers, such as Walmart and
Target, which did not primarily focus on the sale of music, adopted the same
approach. The CD became a loss leader for these retail chains: selling music
for less than the wholesale price made sense if the CD shoppers filled up their
carts with other items. The race to the bottom in CD pricing became an issue
for the labels and ultimately led to a significant legal battle that we will dis-
cuss later in this chapter.
Before 1991, as we saw in Chapter 5, the cassette was the top revenue pro-
ducer for the industry. But in that year, the sales of CDs made it the top-selling
format, a position it would not relinquish for almost 20 years. However, an
even more impactful change occurred at the cash registers of those stores
in 1991: Soundscan, a New York-based research firm, began collecting data
based on scanning the bar codes on CDs (and other music formats). Before
then, Billboard compiled its charts based on phone calls to stores to find out
which titles were the hot sellers; the process managed to be both inaccurate
and subject to bribes and other forms of manipulation. Charts based on the
Soundscan data revealed that genres such as hip-hop, country, and grunge
Compact Discs 201
were far more popular than the industry believed. This set the stage for artists
as diverse as Garth Brooks, Ice Cube, and Nirvana to reach superstar status.xiv
Record stores tried to compete more effectively against these giant retail
chains by enhancing their shopping experiences. One solution was to return
to what had worked so well in the 1950s. At that time, record store patrons
had the option of entering isolation booths to hear the latest singles to help
them decide which songs to buy. The heyday of the CD revived the concept,
as retail stores installed listening stations with selections as large as 14,000
CDs. Fans could once more get acquainted with new music they had not
heard previously to help them to decide what they should purchase.xv
At the CD’s launch in 1983, the catalog of available titles in the format in
the United States numbered around 20. Rights issues hampered the list from
matching the 50 unique items introduced in Japan. This was not the most
auspicious beginning for the format that, as we know now, became the source
of phenomenal growth.
Polygram Records was a subsidiary of Philips, and Sony/CBS was a
joint venture between CBS Inc. and Sony Japan Records. Even with those
relationships, getting all the labels on board required effort. Jan Timmer,
the head of Polygram at the time, described the initial meetings as “Hostile.
Very hostile.” The industry was in the midst of a slump in both vinyl record
and cassette sales. The labels had made significant investments in the pro-
duction facilities for those formats, and the prospect of spending upwards of
$20 million to set up a production line for the unproven CD because of the
“clean room” conditions required for quality control seemed like a very risky
proposition.xvi Several top executives at the labels, including Walter Yetnikoff
of CBS Records, were adamantly opposed to the compact disc. In addition to
the investment required, he saw the CD as providing a source for even higher
quality bootleg cassettes than those that, as we saw in Chapter 5, had plagued
the industry.
Meanwhile, Sony and Philips, after the respective failures of Betamax and
LaserDisc, had a lot riding on the success of the CD. They were willing to in-
vest resources and political capital to overcome the naysayers. Sony’s Ohga
went over Yetnikoff ’s head to the head of the CBS Corporation to ensure
202 Key Changes
15,000
$13,200
$12,800
$11,400
$9,900 $9,900
10,000 $9,400
$8,500
$ (M)
$6,500
$5,300
5,000 $4,300
$3,500
$2,600
$2,100
$1,600
$930
$390
$17 $103
0
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Year
Figure 7.4 CD unit revenues by year, 1983–2000. Source: RIAA. Howie Singer.
century, CD volumes approached 950 million units and accounted for a re-
markable 87 percent of the total album units sold in the US.
The revenue impact of the CD (see Figure 7.4) exceeded its volume im-
pact, because each disc sale put more money in the cash register than the
comparable album or cassette. LPs and cassettes were priced at $7–$12 on av-
erage in the 1980s. Retail prices for CDs were nearly double that at $15–$20
retail. The $100 million realized in 1984 was only 2.4 percent of the industry’s
annual revenue. In 1989, the CDs accounted for $2.6 billion, or about 40 per-
cent of total revenues. Four years after that, almost two-thirds of total in-
dustry revenues were earned by the CD. More than 92 percent of revenues in
2000, or over $13 billion, were brought in by the CD. In the span of 17 years,
CDs had gone from a footnote to being virtually the entire recorded music
business.
LP sales dropped off from roughly 300 million units in the late 1970s to al-
most nothing by 1990. CD sales at their peak tripled the pre-CD vinyl num-
bers. (Cassette volumes remained relatively stable until the late 90s.) That
growth in revenues from the CD was attributable to some of the best-selling
artists of all time at the peak of their creativity, including Michael Jackson,
Madonna, and Prince. But it also was caused by millions of fans preferring
the sound quality, convenience, and reliability of the CD to the possibly
scratched and worn vinyl records that they already had on their shelves. The
labels observed this behavior and dug into their catalogs to release older
albums that were past their peak but held special places in the hearts of music
204 Key Changes
fans. CD reissues of the best music of the 1960s, including the Rolling Stones,
Bob Dylan, and the Motown catalog, flowed into the market.
One glaring initial exception to this flow of music was the Beatles; EMI
made news when it decided to release the Fab Four on CD at last.6 Reports
at the time suggested that financial disagreements with the band (including
the estate of John Lennon) over payments of past royalties had caused the
label to delay the release as a negotiating tactic. It is an indication of the CD’s
financial power by the late 1980s that this leverage was even considered as
a factor. EMI denied this was the case and stated that it had been waiting
until it had enough manufacturing capacity to meet the anticipated de-
mand for the biggest selling band of all time. Regardless, the first four Beatles
albums were reissued in February 1987—in monaural sound like the orig-
inal releases. Eight more albums followed by October of that year, including
Help!, Revolver, Rubber Soul, and Abbey Road.xvii
The enormous CD sales volumes drove top-line revenues to new heights
for the labels. And the labels’ royalty payments and practices ensured that the
margins they obtained from each were in their favor as well. Manufacturing
CDs was far more expensive than making cassettes when the product
launched. In recognition of this disparity, the record companies used
reduced percentages of the cassette royalty rate to calculate CD payments.
This difference could have been 15–25 percent less depending on the specific
artist and label. Even when CD volumes increased and manufacturing costs
declined, this reduction in royalties for “new technology” persisted for sev-
eral years.xviii
That was not the only way that the record companies structured their deals
to maximize their profits on the CD. Most deals included a packaging deduc-
tion to reflect that the music itself should generate payments for the artist,
but not the plastic case, cover art, or booklet. This deduction typically ranged
from 20 to 25 percent on CDs; and yet again, these levels persisted even as the
CD packaging costs declined with volume.
From the 1960s on, the music industry preferred the album over the single
as the best way to experience an artist’s creativity and for its superior eco-
nomic returns. Most albums had just one or two hit singles if you were lucky.
Michael Jackson’s Thriller album released in 1983 changed that pattern: it set
the record with seven top-ten singles from a single album, including “Beat
It” and “Billie Jean.” The sheer number of hits convinced fans they had to
own a copy of the album; it sold 32 million copies by the end of the year.7
Others followed suit with both Bruce Springsteen’s Born in the USA and
Compact Discs 205
Janet Jackson’s Rhythm Nation matching the record of seven top-ten singles
each in 1984 and 1989 respectively.
By the 1990s, the labels shifted gears and chose to cap the number of sin-
gles pressed or n
ot to release certain singles at retail at all while promoting
those songs on the radio. The aim was clear: persuade fans to buy that far
more profitable album on CD. Instead of the purchase decision being driven
by overall entertainment value, let a ubiquitous radio hit be the driver. For
example, Chumbawamba’s song “Tubthumping” never topped the Billboard
Hot 100 chart despite its enormous popularity. The label manufactured only
100,000 singles, effectively limiting its performance on the chart.xix It is
doubtful that anyone at Chumbawamba’s labels was particularly upset with
this outcome, as the album the song was on, Tubthumper, sold 15 million
copies worldwide. That single song, in essence, generated close to $100 mil-
lion in revenues from CD sales in the late 1990s. The so-called “War on the
Single” yielded lucrative returns in the short term, though it fed the fans’
frustration with paying a lot of money for a CD to get access to the one song
they really cared about.xx
Although typical listeners were impressed with the quality of the sound,
many artists, record producers, and engineers were somewhat less enamored
with the CD. A group was formed called “Music Against Digital,” and one of
its main spokespeople was Neil Young.xxi In 1992, Young wrote an opinion
piece for Harper’s magazine entitled “The CD and the Damage Done.”8 He
referred to the CD era as the “darkest time ever for recorded music”9 and
wrote that “listening to a CD is like looking at the world through a screen
window.”xxii
Young was not alone in his disdain for the CD. Many artists felt (and still
feel) that vinyl records produce a “warmth” that is missing from the dig-
ital versions. Many audiophiles feel that way too, even though both CD
mastering and playback technologies had advanced since the early days of
the CD to ameliorate the “brittleness” of the sound. Nick Rhodes, a founding
member of the New Wave band Duran Duran, recalls his reactions when
the CEO of EMI played him a CD of the band’s second studio album Rio,
which had peaked at no. 2 on the UK album charts and no. 6 in the United
States. Rhodes had mastered several of the tracks for that album in addition
206 Key Changes
to writing and performing the songs with his bandmates. He said, “I almost
wanted to cry hearing it. It was not my idea of progress.” The sound was
“cleaner” than the vinyl record, but he “liked all the noise” on the original
because it offered the “atmosphere” the band wanted and had worked so hard
to create in the studio. And he thought the “packaging was horrible” too. The
broad canvas created by the vinyl album packaging with photographs, art-
work, and lyrics had been reduced to a functional but far less compelling
4.75-by-4.75-inch booklet.xxiii
Peter Asher, Apple Records executive and famed producer of James Taylor
and Linda Ronstadt, among many others, pointed out that the precision
of the CD table of contents enumerating the start times for each song did
not always serve the artistic goals of the creators. For example, Side 2 of the
Beatles’ Abbey Road contains the famous 16-minute suite of eight short songs
blended together. The idea was Paul McCartney’s, and it was an artistic land-
mark. But Asher said it caused a debate among those involved with the CD
reissue: it was unclear whether and how these individual sections should be
delineated in the ToC on the CD.
Rhodes took great pride in sequencing the tracks on Duran Duran records.
He was conscious of the emotional journey, the tempos, and the keys as one
song led to the next. For vinyl, one had to be conscious of selecting the last
song on side A to entice the fan to get up and flip the record over to hear
more. But one also had to deal with the time limitations and the greater like-
lihood of the needle jumping the track if the final song on the side had too
much bass. Jac Holzman, the founder of Elektra Records, said “Sequencing
was an art form, and I always reserved the right to do it on my records.”xxiv
Peter Asher remarked that, when he sequenced albums during the CD era,
he would come to the middle of the album and think, “We should put some-
thing in here to fire everyone up”—realizing that he would be choosing the
Side 2 leadoff track of a vinyl LP.xxv
Rhodes and Holzman both felt that the random access feature for tracks on
the CD removed a valuable tool from their bag of tricks to create a great ex-
perience for fans. Using it now on Abbey Road randomizes those eight short
songs that McCartney and producer George Martin so carefully assembled.
Albhy Galuten, Grammy-winning producer of the Bee Gees, summed up the
creative community’s attitude toward this feature succinctly: “Shuffle is the
devil.”xxvi
As discussed in the opening to this chapter, how the 74-minute length of
the recording time on the CD became the number is a subject of ongoing
Compact Discs 207
heavy metal had first emerged as a genre in the 1970s, the raw, raucous sound
was well-suited to the CD’s sonic capacity. Headbanging worked better when
the sound was louder, and therefore it is not surprising that two of the best-
selling albums of the decade were AC/DC’s Back in Black and Guns N’ Roses’
Appetite for Destruction.xxvii
One challenge to the growth of CD revenues was that it did not require a
criminal enterprise to make pirated copies. All it took was a CD owner de-
ciding that the album they had purchased no longer deserved a spot in their
collection. Music fans were quite familiar with the pristine sound emanating
from the discs no matter how many times they had been played. They also
were well aware that the CD was much more difficult to damage or to break
than a vinyl record or a cassette. The characteristics that made CDs so valu-
able as a format to purchasers also made them ideal to sell as used items.
There had been a viable business in selling used vinyl albums for years,
as we discussed in Chapter 4, which was enabled by the first sale doctrine
embodied in copyright law. This says that objects such as photographs,
books, and records are “staple articles of commerce,” like clothing or cars or
furniture, and can be “alienated” (resold, loaned, given away, etc.) without
any involvement from the original seller. The US Supreme Court enumerated
this doctrine in 1908, and it was codified in the Copyright Act of 1976 as
Section 109 of the law.
Once a vinyl album or a CD has been sold lawfully or even transferred le-
gitimately for free, the copyright owner’s interest in that physical item is “ex-
hausted.”10 The purchaser can decide to give that purchased item to a friend,
sell it to a retailer, use it as a drink coaster, or simply throw it away. They can’t
make a copy of the work and then sell or give it away, as that would implicate
the reproduction and distribution rights. Owners of CDs—as well as LPs and
other sound recordings—are not permitted to rent them to anyone else. This
last restriction is different from how books or movies are treated. The first
sale doctrine allows rentals of those products, which enabled businesses such
as Blockbuster Video to thrive renting VHS tapes in the 1980s and the orig-
inal Netflix DVD-by-mail business to boom at the turn of the century.
The different policy between music and movies on rentals was established
in the Record Rental Amendment of 1984. This exception to the first sale
Compact Discs 209
doctrine applies to sound recordings that contain only musical works. It does
not apply, for example, to audiobooks or movie commentary tracks. This nar-
rowly crafted section of the US Copyright law was implemented to prevent
anyone from creating a music rental business. The record industry success-
fully lobbied for this change, during a period when home taping on cassettes
was becoming widespread, because it believed that such rentals would en-
courage home copying without paying royalties to artists or labels.11
Independent record stores relied on the first sale doctrine to offer used
CDs. The low prices of $2–$8 made them compelling alternatives to new discs
that were priced about $10 higher for many shoppers. Though the industry
and artists did not earn any royalties from these transactions, the losses were
limited as long as sales were confined to the indie retail community. In 1992,
the Hastings Books and Records chain based in Texas, with over 100 retail
locations, began selling used CDs. Wherehouse Entertainment, with over
300 stores, followed suit.
Although used CDs sales represented less than one percent of the total in-
dustry sales, the record companies decided to take action before other major
record store chains decided to undercut the enormously profitable new CD
business. The distribution arms of Sony Music, Warner Music, Universal,
and EMI, representing four of the six major record labels at the time, all
withdrew advertising and promotional allowances from any retailer that sold
used CDs.xxviii
The more restrictive policy hit the retailers hard, because these allowances
represented a major portion of what chains spent to advertise CDs. The
Wherehouse chain alone faced the loss of $5 million annually. Wherehouse
had conducted test marketing of used CDs and found they had appeal in
all of its markets, from suburban malls to college campuses. Rather than
back down, Wherehouse, along with the Independent Music Retailers
Association, representing small music stores, filed antitrust lawsuits against
the labels in 1993 for acting in lockstep to end the practice of providing this
valuable financial support.
The major labels blinked first and revised their policy. They restored the
advertising and promotional dollars, but they stipulated that the money
could not be used to support the used CD business. The used CD remained a
niche offering, and the antitrust suit never came to fruition. (Of course, used
CDs are still sold, though the market for used vinyl currently dwarfs that of
used CDs.) But this was not the last time that the major labels would attempt
to use their muscle to influence how retailers marketed CDs.
210 Key Changes
or documents would not qualify, while a CD-R that is specifically targeted for
making personal copies of music CDs would qualify.
Makers of devices and media that fit these definitions are required to pay
levies of 2 percent on recording devices and 3 percent on blank DATs to the
US Copyright Office. The law specifies how those funds are allocated and dis-
tributed to rights holders to compensate, in part, for the money that would
have been paid for purchases of original recordings. There was actually no
technical difference between the CD-Rs that were marketed as being suitable
for music and those not so labeled. The music CD-Rs were priced higher,
however, to cover the levy payments under the AHRA. If you searched for
“CD-R Music” on Amazon or Walmart decades after the peak of the CD
format, you could still find these discs.
Even with the law in place, CD-burning capabilities that cropped up in
homes across the country worried the industry. This development raised
questions about how strong a public posture the industry should take on the
legality of such copying. The industry pushback against home taping, which
we discussed in Chapter 5, had not yielded positive results. The bootleg
businesses just described clearly fell outside the line of legality as they
pursued profits at scale. But individuals making single backups of CDs that
they had previously purchased presented a different set of facts. And the fact
pattern was affected by whether the consumer put the copy in a drawer or
gave it to their next-door neighbor as a gift.
The RIAA spelled out its views on permissible behavior with regard to
CD-Rs on its website as follows:
• “It’s okay to copy music onto special Audio CD-R’s, mini-discs, and dig-
ital tapes (because royalties have been paid on them)—but not for com-
mercial purposes.
• Beyond that, there’s no legal “right” to copy the copyrighted music on
a CD onto a CD-R. However, burning a copy of CD [sic] onto a CD-R,
or transferring a copy onto your computer hard drive or your portable
music player, won’t usually raise concerns so long as:
◦ The copy is made from an authorized original CD that you legiti-
mately own.
◦ The copy is just for your personal use. It’s not a personal use—in fact,
it’s illegal—to give away the copy or lend it to others for copying.
◦ The owners of copyrighted music have the right to use protection
technology to allow or prevent copying.
Compact Discs 213
In other words, the RIAA takes the position that what you’re doing if you
copy music onto CD-Rs, or other blank digital media, is technically a vio-
lation of copyright law but admits the practicality that you’re extremely un-
likely to get caught if you’re copying your own music and the copy stays in
your possession (not to mention that such actions may be fair use).
As discussed earlier in this chapter, using the CD as a loss leader to drive
retail traffic concerned the labels because the practice could devalue their
product in the eyes of the average consumer. It certainly made it harder
for smaller record stores to compete when they did not have the luxury of
making up for losses on music with profits on other goods. Yet by law, the
labels could not dictate the retail price of their goods. So instead they created
a workaround called the “Minimum Advertised Price” (MAP) to encourage
retailers to keep prices at what they considered to be reasonable levels. The
labels contributed marketing dollars to help fund those weekly circulars if
the retailer kept their CD prices above the MAP.
As with their earlier efforts to withdraw advertising funds from retailers
that sold used CDs, these tactics landed the major labels in legal hot water.
The Federal Trade Commission launched an inquiry in August 2000 as to
whether marketing agreements such as MAP were intended to raise prices
by putting an end to the price wars being pursued by Target, Best Buy, and
others—in which case they would be unlawful restraint of trade. The FTC
believed that the result of the program was that consumers had paid an extra
$5 per CD, resulting in almost $500 million in overcharges over five years.
Shortly thereafter, Florida and New York led a coalition of 41 states in
filing a price-fixing lawsuit against the record companies. A settlement was
reached in 2002 that included the then five major labels (BMG, EMI, Sony,
Universal, and Warner) and music publishers, as well as retailers including
Tower Records, Trans World Entertainment, and Musicland. The parties
admitted no wrongdoing, though they agreed to pay a fine of almost $70 mil-
lion and distribute more than $75 million in CDs to nonprofit and public
groups.xxxv
The legal issue was not really about the specifics of the MAP program itself.
The retailers remained free to set retail prices at whatever levels they chose.
However, the circumstances were evaluated under different antitrust legal
standards if competitors discussed implementing these policies collectively
214 Key Changes
and then the entire group put them into place in a unified and coordinated
fashion. It was the horizontal agreement across the industry, which the FTC
alleged amounted to restraint of trade, that created the much greater legal
risk that led to the eventual settlement.
Conclusion
The CD was an unparalleled success for the music industry. Getting music
fans to buy new music as well as copies of all their favorite older albums, at
higher prices than vinyl or tape, drove revenues and margins to stratospheric
levels. That money bankrolled big deals for executives and artists alike. (It
also covered the costs of wrecked hotel rooms and a wide variety of illicit
drugs.) The desire for more profits even compelled the labels to pull back on
the release of singles to boost CD sales further. Though this was great for the
bottom line, the business strategy fueled consumer frustration. Fans resented
the need to buy an entire CD to get the one song they cared about.
The unprecedented prosperity and fan frustration from the CD were
like plates under the Earth’s surface rubbing against each other to form
a fault line. The conditions were in place for a major earthquake triggered
by technologies that were coalescing in the 1990s to create yet another
music format. This time, the format enabled college students to vent their
frustrations and to wrest control from the grasp of the industry movers and
shakers who were loath to change their lucrative ways.
8
Downloads
Don’t Download This Song—Weird Al Yankovic
Turning analog audio files into a collection of zeros and ones that could
be embedded on a plastic disc was the fundamental innovation that led to
the flourishing of the music industry in the late twentieth century. Every
Compact Disc (CD) is a digital “master.” With the right tools, which had be-
come standard on every PC or laptop by the end of the last century, every
song file on the CD could be copied, compressed, and then “shared” with
a million “friends” over networks. Unlike the DVD, which was developed
in the mid-1990s, the CD specification of the early 1980s contained no
provisions for encrypting the content, nor were any other mechanisms to
deter copying included.
With the benefit of hindsight, it is easy to criticize this omission, but it
was more than just a simple oversight, because of the state of computer
technologies in the mid-1980s. A typical personal computer at the time of
the CD’s introduction was equipped with a 10 Megabyte hard drive, while a
typical CD contained over 700 Megabytes of data. There wasn’t even enough
space on that drive to hold the data from a single song, let alone an album.
Not to mention that the removable memory slot on that PC was for a 360 kil-
obyte floppy disc.i
CD drives for computers did not become standard until a decade later. The
sheer size of the data files on the CD created what appeared to be, at the time,
an insurmountable security barrier to copying. Thus it is not surprising that
Sony and Phillips chose not to address these issues when creating the CD.
There were some early downloadable music concepts that pointed the way
for what was to come. As early as 1982, Frank Zappa proposed aggregating
a catalog of tracks and distributing them over cable networks to fans who
would record as many as they wished on tape for a flat monthly fee.ii Just over
a decade later, the Internet Underground Music Archive (IUMA) launched
to allow unsigned artists to share their music with fans via FTP sites.1
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0008
216 Key Changes
If you put all those pieces together, you had the technology foundation
for MP3-based music distribution. Music fans could create MP3s from their
own CDs and store them on their machines and play them back later at ade-
quate quality. They could download already compressed song files from on-
line sites, save the bits on their hard drive, and then play them back whenever
they wanted. Although it could take 15 minutes to download a single track
using a modem, that was a lot quicker than driving to the record store. With a
writeable CD drive, those decoded tracks could be burned back to a disc and
be listened to on any CD player whether in the home or on the go.
It didn’t take long for several startups to take these elements and inte-
grate them together into what were, in essence, Minimum Viable Products
(MVPs) for downloadable music e-commerce services. Liquid Audio and
a2b music both developed such services that relied on software security to
deter the copying of the digitally distributed songs. Liquid Audio was formed
in 1996 in Redwood City, CA, by Gerry Kearby, Robert Flynn, and Phil Wiser
to offer an end-to-end solution for secure downloadable music. At the same
time, a2b music was formed as an internal venture at AT&T Labs by Larry
Miller and Howie Singer. It used Advanced Audio Coding (AAC), a percep-
tual audio coding scheme like MP3 with improved sound quality at compa-
rable bit rates. A2b music’s security architecture was based on public/private
key cryptography, which could also run in software on the more powerful
PCs of the later 1990s.
The focus on security was essential to getting the record labels to consider
these new platforms. The music companies had come to realize the threat
posed by the music files on the CD being “in the clear,” and they would not
contemplate distributing music online without a stringent Digital Rights
Management (DRM) solution being built-in. DRM gave rights holders the
ability to set up policies or rules regarding the use of the music files even
after they were stored on an end-user’s computer. Those rules could include
such things as “play this song a limited number of times only” or “play the
song during a limited time window” or “this song may not be copied onto
a CD.” Songs were downloaded with encrypted licenses specifying these
permissions and, before playing the music, the software decrypted and
checked that the license contained the correct approvals for the action
requested.
Even though these entrepreneurs had crafted their services to meet the
needs spelled out by the major labels, the industry didn’t embrace these new
initiatives. They expressed mild interest, but given the gusher of cash that was
Downloads 219
flowing in from CD sales, none of the music companies was ready to jump in
wholeheartedly. At one of the first meetings to discuss a2b music, one major
label executive, after hearing the compressed digital files played through ex-
pensive speakers in his boardroom, said, “No one is going to listen to that
shit.” Other label meetings elicited even more colorful language.
There was more at work here than mere reluctance to abandon the enor-
mously profitable CD. The top label executives were simply ill-equipped to
wrestle with the upheaval of the late 1990s. No one exemplified this more
than Doug Morris, then the CEO of Universal Music. A “record man” in the
truest sense, he even wrote “Sweet Talkin’ Guy,” a 1966 hit song recorded by
the Chiffons. When looking back at this era, Morris basically admitted that
music business leaders were clueless: he said, “There’s no one in the record
company that’s a technologist.” Even that assessment was questionable, as he
acknowledged in the same interview that he “wouldn’t be able to recognize a
good technology person.”ix In fact, the major labels had people who under-
stood the issues clearly, though their advice typically fell on deaf ears.5
At the same time these nascent businesses were attempting to address the
major labels’ security concerns, others were taking advantage of the unpro-
tected CD files and the availability of MP3 software tools to feed the desires of
music fans and indie artists directly. Justin Frankel, a teenage dropout from
the University of Utah, developed the Winamp music management software
for Windows PCs, shown in Figure 8.1, to make it simpler to create and to
playback MP3 files.x Version 1 of the shareware program quickly became the
Internet’s most popular player for MP3-encoded tunes with over 3 million
downloads. The controls to play and to skip files were straightforward, and
there was even a graphic equalizer that allowed users to adjust the sound of
the music. Though the software enabled songs to be imported from CDs and
added to the users’ collections, many people obtained the music files without
authorization or payment via Internet Relay Chat (IRC), an early text-chat
system that could be used to send and receive files.
With the MP3 encoder built in, Winamp and other similar programs were
equipped to take all the digital data that constituted the songs on a CD and
turn them into compressed music files. Unfortunately, there was a critical
set of information that the CD lacked: the name of the disc, the artist’s name,
and the song titles. The CD specification did not consider the possibility
that the audio tracks would someday need to be extracted as data files to be
catalogued, so there was no metadata (data about the data that made up the
songs) on the disc other than the human-readable text. The Compact Disc
220 Key Changes
Figure 8.1 WINAMP user interface with graphic equalizer. Used by permission
of Winamp Inc., All Rights Reserved.
Database (CDDB) was created to avoid relying on every person entering this
detailed song data themselves. To use the CDDB, first an identifying code
was calculated based on characteristics of the CD inserted into the PC’s
drive, including the number and length of tracks. That number was used to
look up the other data about the disc in the database. If crowdsourced data
was available, the title and artists fields would be populated automatically. If
the particular CD was not yet in the database, the user entered all the relevant
metadata, which then became available for all subsequent lookups. CDDB
Downloads 221
Figure 8.2 AT&T a2b music flash memory prototype. Courtesy of AT&T
Archives and History Center.
The Diamond Rio PMP-300 had a small LCD screen that showed the name
of the track being played and buttons for controlling playing, stopping, and
skipping. It came with 32 MB of memory and a SmartMedia flash memory
card slot to increase the storage space available. A single AA battery in the de-
vice powered 8 to 12 hours of playback. To get the music onto the device, you
connected it to a computer’s parallel port and transferred the files over. The
product came with a copy of MusicMatch software, a competitor to Winamp,
to manage the MP3 files on the PC. If your songs were compressed at a bitrate
of 128 kbit/s, the device could hold only 30 minutes of music: not even enough
songs to get through a typical workout without repetition. In addition, the de-
vice had several physical design problems that led to parts frequently breaking
or falling off.xiii Yet Diamond Multimedia faced a bigger issue than a few de-
fective parts: the RIAA sued it for copyright infringement (which we will dis-
cuss in more detail later in the Copyright section of this chapter).
Like the first digital music services, these initial products with their lim-
ited features would certainly be classified as MVPs. The Rio eventually sold
Downloads 223
over 200,000 units, which is significant for a new consumer electronics cate-
gory but far from mass-market success. The Apple iPod was the MP3 player
that broke through to mass-market popularity when it was introduced three
years after the Diamond Rio. Many elements of the product reflected Steve
Jobs’s elegant design sensibility, such as the inclusion of a click wheel to make
scrolling through thousands of songs and selecting the right one intuitively
simple. But the product could not have contained thousands of songs without
a major innovation from Toshiba. All the other MP3 players in the market
used flash memory to store compressed music files. That storage technology
was coming down in price rapidly but not quickly enough to achieve the “a
thousand songs in your pocket” realized by the iPod at $399.xiv
Jobs’s interest in developing a portable music player in 2000 had initially
come to naught because the necessary components to build a product with
the right combination of price and capability simply did not exist. Then Jon
Rubinstein, the head of Hardware Engineering at Apple, visited Toshiba
on a regular supplier visit in February 2001. Toshiba’s engineers had devel-
oped a tiny 1.8-inch hard drive, but they were not sure how they could use
it. Rubinstein knew immediately that the drive was the linchpin for Apple’s
music device. He asked Jobs for a $10 million check to purchase the drives,
saying, “I know how to do it now.”xv
The first iPod, shown in Figure 8.3, launched in October 2001. Users
connected it to their Macs to transfer songs from iTunes, the music man-
agement application that Apple had launched earlier that year. Apple’s ad
campaign famously featured the slogan “Rip. Mix. Burn.” which encouraged
users to encode the music from their CDs and to make copies for others by
burning CDs. This did not sit well with music executives, and adding music
portability with the iPod did not make them any happier.
Eventually, Apple added the sale of downloadable music to its hardware
and software suite. Elegantly integrating software and hardware, Apple gave
music fans an end-to-end solution that made it simple to buy and download
music and transfer it easily to an iPod. It needed to convince the labels, in
spite of the objectionable iMac marketing message, that it would be a good
partner and that licensing their catalogs to Apple for download sales made
sense. Steve Jobs’s powers of persuasion certainly played a role, but the labels
were ready to listen—because of the impact of illegal file-sharing triggered by
the launch of Napster.
Founded by Shawn Fanning and Sean Parker, who were respectively 19
and 20 years old in 1999, Napster used “peer-to-peer” (p2p) technology
224 Key Changes
Figure 8.3 First Generation Apple iPod. Apple Newsroom. Used by permission
of Apple Inc. All rights reserved.
as the underpinnings for their file-sharing service. P2p meant that users’
devices shared files directly among one another rather than going through
a central server such as one operated by CompuServe or America Online
(AOL), the dominant consumer online service providers of the time.
Yet Napster wasn’t a pure p2p model: it had a central index that indicated
which files could be found where. When someone installed Napster, it created
a list of all the music files on that individual’s machine that became part of the
centralized Napster index. When another user wanted a song, they searched
that index and selected the track they wanted. That user was then connected
directly with the PC of the user who had the desired song, and a file transfer
occurred from one “peer” or user directly to another.
Hundreds of college students at Northeastern University, where Fanning
studied, were actively exchanging music just weeks after the beta launch. The
small size of MP3 files and the high-speed networks in dorms were a perfect
Downloads 225
match. Form followed function. The early adopters were willing to live with
a spreadsheet-like interface filled with text and numbers, as shown in Figure
8.4. The experience was geared toward tech-savvy users searching a database
of tracks. There were no graphical elements related to artists or playback.
The music industry had dealt with unauthorized copying before with cas-
sette tapes and CDs, but Internet file-sharing brought heightened challenges.
In addition to the network effect of the service quality improving simply by
adding more users, Napster had other advantages. The costs of the computer
and connectivity were, in essence, fixed costs, and there was no need to buy
blank tapes, nor were there any incremental shipping costs. Cassettes had to
be recorded in real time (or perhaps half real time), but it took virtually no
time to copy a digital file particularly if both parties were on a fast campus
network. Finally, those digital copies sounded remarkably like the originals,
while cassette copies had reduced sound quality.
A variety of Internet protocols, including Usenet, Hotline, and IRC,
were being used to distribute MP3 files before Napster. Finding the content
you wanted required searching through a variety of services and message
Figure 8.4 Napster user interface. Used by permission of Napster Inc. All rights
reserved.
226 Key Changes
threads. It worked, but it was much simpler and far more effective to search
via Napster’s index. Furthermore, the more people who downloaded the soft-
ware, the larger the catalog of music available became.6 The more popular the
song, the more likely that you could find a peer that had the file, which meant
you could easily get your own copy. Napster went viral, spreading rapidly
among universities and then to the broader population. By March 1999, MP3
files were more popular than sex—at least in terms of Internet searches.xvi
Napster’s mainstream appeal was reflected in the media attention accorded
to Shawn Fanning and his cofounder Sean Parker. Fanning appeared as
a presenter at the MTV Music Awards in 2000. MTV News featured the
cofounders in an interview, where Parker said, “We think that when music
transitions to digital distribution, people will pay to receive music to their
cellphones or to their portable devices.”xvii That argument was not persua-
sive to the music establishment, as Napster usage peaked just five months
after launch with 80 million users transferring approximately 2 billion files
per month without paying for the privilege. Imputing a conservative price
for each one of these songs, the estimated value of these copies was on a
pace to exceed the annual revenues of the entire US recorded music busi-
ness. By October 2002, Napster had become so mainstream that Fanning was
pictured on the cover of Time magazine.
Sony and Bertelsmann created Sony BMG Music as a joint venture in
2004. In this case, two heads were not better than one in dealing with the
downloadable music threat. One year later, that new entity managed to pro-
duce the greatest technology fiasco of the downloadable music era: the copy-
protected CD. The CD was the fountainhead for all the music that ultimately
ended up on the file-sharing networks because no copy protection had been
included in the original specification. Labels experimented with technology
to prevent copying the files on CDs, but most abandoned their efforts be-
cause of compatibility issues with the more than 2 billion CD players in the
market. In fact, Philips asserted that such discs should not even display the
CD logo.
Sony BMG decided to add an additional piece of software to every CD
that it released. When one of these CDs was inserted in a computer drive,
a software program called XCP, developed by the company First4Internet,
installed automatically that altered the operating system to prevent CD
copying. This software installed a “rootkit,” a way of accessing low-level func-
tionality in the PC operating system, as part of a layer of DRM technology for
the CD where none had existed before. The rootkit created vulnerabilities to
Downloads 227
malware and was difficult to uninstall. This retroactive attempt to limit the
CD’s utility and deter file-sharing caused a consumer uproar, a class-action
lawsuit, and a government investigation. Sony BMG eventually paid resti-
tution to some affected users and abandoned the entire effort in 2007. The
unprotected digital masters on CDs continued to serve as an essentially lim-
itless supply of free music for anyone with the right software.7
As we’ve seen before, new formats and products test the boundaries of es-
tablished copyright law and make it more difficult to enforce rights. The cre-
ation, copying, and transferring of digital music files triggered a variety of
legal actions that were aimed at stopping these disruptions in their tracks.8
The first legal skirmish of the download era was over the Rio Portable
MP3 Player. As discussed in Chapter 7, Congress passed the Audio Home
Recording Act (AHRA) of 1992, which required that digital audio re-
cording devices incorporate Serial Copy Management Systems (SCMSs), or
technologies designed to ensure that only limited first-generation copies of
digital audio recording could be made, and that royalties on those copies be
paid to songwriters and recording artists. The AHRA was originally intended
to apply to Digital Audio Tape (DAT), a putative successor to the standard
audio cassette that never succeeded as a consumer media format, though it
was used in recording studios and as a backup format for computer data.
The Recording Industry Association of America (RIAA) sued Diamond
Multimedia, claiming that the new product did not employ a SCMS and
therefore violated the AHRA. Digital audio recording devices as defined by
that law were required to make payments to the industry. In essence, the in-
dustry did not want to allow MP3 songs on computers to be transferred to
devices that did not pay compensation of artists and labels as part of their
cost structure.
The lower court denied the RIAA’s request for an injunction to stop the sale
of these devices, and the labels lost the higher court appeal as well.xviii The
court ruled that the Rio did not qualify as an audio recording device under
the AHRA because the copies of songs on the device were transferred from a
computer hard drive and not directly from a transmission of audio files. The
limited functionality of the device and its reliance on the somewhat incon-
venient step of loading the music from the PC had insulated the Rio from the
228 Key Changes
law’s requirements. The court also ruled that computer hard drives were not
digital audio recording devices, because making digital audio recordings was
not the primary purpose for the device. Finally, the court ruled that “space-
shifting”—moving the music from one user device to another—facilitated
fair use as intended by the law. Instead of enabling the music industry to re-
ceive payments for all manner of PCs, laptops, and solid-state music players,
the courts had created fertile space for the consumer electronics business to
thrive in digital music.9
The stakes in the Rio case were certainly significant, but they were dwarfed
by those of the next major lawsuit in the MP3 era. The RIAA sent cease-and-
desist letters to and sued several sites that offered downloads of free MP3 files
from their servers. These sites either shut down immediately or did so as part
of a settlement with the industry.
Napster’s reliance on p2p technology meant that it was not providing the
files itself, and this changed the legal arguments substantially. Nevertheless,
the music industry viewed its “hockey stick” growth as an existential threat.
If every fan could find and download any song for free, who would buy those
$15 CDs that were funding the labels and their executives’ bonuses? On
December 6, 1999, just a couple of months after Napster launched, 18 record
labels, all part of the “big four” major-label groups at the time (EMI, Sony
Music, Universal Music Group (UMG), and Warner Music Group (WMG),
filed a district court case alleging that the service was liable for copyright in-
fringement. They requested a preliminary injunction to halt the exchange of
copyrighted songs immediately.
It wasn’t just the labels that thought Napster was taking money out of their
pockets; many artists shared these concerns. The heavy metal band Metallica
filed its own lawsuit against Napster in April 2000. Not only did it allege that
Napster was liable for copyright infringement, it charged the company with
racketeering under the RICO statute, which was most notably used by the
FBI to take down the bosses of organized crime families. Metallica’s suit even
looped in several universities, including Indiana University, Yale University,
and the University of Southern California, for allowing their networks to be
used to steal their music via Napster.
At a rate of $100,000 per track downloaded illegally, Metallica requested
at least $10 million in damages.10 The band hired the consulting firm NetPD
to gather detailed information about the more than 300,000 Napster users
who were providing copies of their songs to others.xix The list came to 60,000
pages. Metallica’s drummer, Lars Ulrich, delivered that list to Napster’s office
Downloads 229
himself and requested removal of all these accounts from the service. The
backlash from the Napster community was swift and fierce, as Ulrich began
to receive death threats. Dr. Dre, the hip-hop pioneer, adopted a similar ap-
proach and presented his own list of 230,000 misbehaving users to Napster.
Napster banned those accounts from the service, which only inflamed
Napster users further.xx
In July 2000, Judge Marilyn Patel of the Federal District Court in San
Francisco ruled that Napster was in violation of copyright law. Her ruling
stated that Napster “knowingly encourages and assists” users to exchange
copyrighted music and that this activity was not fair use. The earlier Betamax
case, which we saw in Chapter 5, established a precedent that the maker of
a technology that had “substantial non-infringing uses” could avoid copy-
right liability. Napster’s lawyers tried to make the same argument, but the
court didn’t buy it because the infringing activities far outweighed the au-
thorized uses. Furthermore, Napster’s centralized index of the songs that
were available demonstrated that it had knowledge of the infringements,
thereby increasing its liability.11 To stop the ongoing damage to the labels’
business, Patel issued an injunction that compelled Napster to remove every
copyrighted song from the service.xxi
Judge Patel’s injunction, if upheld, would be exceedingly difficult to imple-
ment. Not only was there a vast number of infringing songs available through
Napster, but identifying all of them accurately was a near-impossibility. And
if the service could be cleansed of that material, it would effectively lose the
very thing that was attracting fans. Napster quickly appealed to the Court of
Appeals for the Ninth Circuit, and just two days later, that court issued an
emergency stay pending its hearing of the case.
The stay gave Napster a window of opportunity. It started talking to tech-
nology companies about how to implement the changes that would be
imposed if its appeal failed.12 In addition, Napster tried to reach a settle-
ment with the record companies that would allow it to stay in business in
exchange for payments to enable the exchange of copyrighted songs. Those
discussions led to Bertelsmann announcing in October 2000 that it would
invest in Napster to support the shift to a licensed model. Bertelsmann—a
large, privately-held German media company best known for its activities in
book publishing—owned BMG Entertainment, which was one of the labels
suing Napster. Because of the deal, Bertelsmann dropped out of the ongoing
legal action against Napster. Bertelsmann Chairman Thomas Middlehoff
said, “Napster has pointed the way for a new direction for music distribution,
230 Key Changes
and we believe it will form the basis of new and exciting business models.”
The plan was to turn Napster into a secure, membership-based service with
the participation of the other major labels.xxii
Mike Bebel was part of the Universal Music Group team that met with
Napster executives to discuss transitioning to a legitimate offer. Napster’s in-
itial offer was five cents per downloaded track. UMG threw them out of the
office.xxiii In February 2001, the press reported that Napster had upped their
proposal to the industry to $1 billion to license music for the service and
settle the legal actions.xxiv That “B” word makes it sound like Napster was
making a very large offer, but it was not quite as rich as it sounded. The total
covered five years of fees, equivalent to $150 million per year to each of the
major labels and another $50 million to independent labels and artists. But
the recorded music industry had just closed out a year with more than $14
billion in US revenues. Risking the majority of those revenues in exchange
for an annual check that represented less than 15 percent of that total cer-
tainly represented a poor deal.
The conventional wisdom is that the music industry made a mistake by
turning down Napster’s entreaties. However, people directly involved with
Napster at the time indicate that the company was never really committed
to a legitimate model. Napster offered Ted Cohen the position of CEO, and
when he declined, it hired him as a consultant to help negotiate with the
labels. Cohen was a long-time entertainment executive who had worked at
EMI Music, Warner Brothers Records, and Philips Media. He was (and is)
well known as an evangelist for new technologies, gadgets, and the promise
of digital distribution. He has confirmed that some of the Napster team were
not serious about a settlement: it simply wanted to “slow walk” the discussions
with the labels, because it had been convinced by their legal team, headed by
heavyweight corporate lawyer David Boies, that Napster would ultimately
prevail in court.xxv
On February 11, 2001, the record labels rejected Napster’s settlement offer
because, in their view, it proposed neither an adequate security solution
for music nor a satisfactory business model.xxvi,xxvii The next day, the Ninth
Circuit issued its ruling on the case, which essentially upheld all of Judge’s
Patel’s key findings. The court said that “a preliminary injunction against
Napster’s participation in copyright infringement is not only warranted, but
required.”xxviii
That ruling rejected a variety of defenses that Napster had put forward. One
of these was the AHRA. In exchange for the levies that the AHRA imposed
Downloads 231
Congress defined four DMCA safe harbors according to the types of online
service providers that existed in the 1990s. Napster claimed that it qualified
for two of these (the safe harbors for ISPs and online storage services); the
court said no. Though the DMCA argument was not central to the Napster
case, it would become one of the key factors in enabling future services such
as YouTube to offer music uploaded by its users, as we’ll see in Chapter 10.
Judge Patel’s ruling did require the lower court to make some changes, in-
cluding a requirement that as part of the process to remove works, the labels
had to give massive amounts of data to Napster that would enable it to iden-
tify which of their songs were being infringed. As predicted, keeping all the
copyrighted songs off the service, as required by the reinstituted injunction,
proved impossible for Napster given its scale. In April 2001, the judge stated
that Napster’s efforts up to that point were “disgraceful.”
Napster tried using a new technology called acoustic fingerprinting to find
the infringing files. Acoustic fingerprinting is a way of identifying music files
by their audible characteristics. An acoustic fingerprinting service analyzes
a file of digital audio and outputs a series of numbers—the “fingerprint” of
the file—that represent the file’s musical characteristics as a human would
perceive them. The way acoustic fingerprinting works, all files containing a
specific song—no matter what codec or bitrate it is, and no matter how it is
digitized—should yield the same fingerprint. The service then looks up the
fingerprint in a database of fingerprints of copyrighted songs and sees if it
matches any songs in the database. If so, the file is a copyrighted work and (in
this case) should be removed from Napster. (We will revisit acoustic finger-
printing in Chapter 10, as it was key to YouTube’s success.)
Although acoustic fingerprinting technology was relatively new and
untested at scale at the time, the technology that Napster used achieved a
99.4 percent accuracy rate. Nevertheless, in July 2001, Judge Patel said this
was not good enough and ordered the service shut down until it could keep
all the copyrighted songs off. The case was partially settled in September
when Napster paid $26 million to rights holders for unauthorized uses of
their music—a tiny fraction of the billions in damages for which Napster
could have been held liable under the theory of statutory damages for willful
infringement.
Even after the shutdown, Bertelsmann remained convinced that Napster
could form the basis of a legitimate offering. It pumped $85 million into
the company in an attempt to build a “copyright respecting” version of the
service that used acoustic fingerprinting to keep unlicensed songs off. But
Downloads 233
it was all in vain: Napster filed for c hapter 11 bankruptcy in May of 2002.
Bertelsmann offered to buy the assets out of bankruptcy for $8 million more,
but a Delaware judge blocked the transaction in September after objections
from rights holders. Sean Fanning’s Napster was finally dead, although as
we’ll see, the brand name would be resurrected later for a licensed streaming
service that still operates today and otherwise has nothing in common with
the original file-sharing network.
Finally, in what can only be described as the crowning irony of this sordid
legal tale, an alliance of publishers and music labels sued Bertelsmann and
Napster’s venture capital firm, Hummer Winblad, for the damages caused by
their financial backing of the p2p service, to the tune of $17 billion. That case
was settled for tens of millions of dollars several years later.
Napster was gone, but p2p technology was not; and the catastrophic
impacts on the music business were not finished. In early 2001, attracted
by Napster’s enormous popularity, a whole new wave of p2p-based clients
launched. These new services, including Morpheus, Grokster, and Kazaa,
were considered the second generation of p2p technology. Judge Patel’s
ruling that a centralized index established knowledge of and responsi-
bility for the infringements on Napster’s service was essentially a roadmap
for p2p developers. The protocols underlying these services, such as
Gnutella and FastTrack, were designed without a centralized server. This
meant less efficient searching for files from an engineering perspec-
tive while cleverly sidestepping the legal argument that had proved fatal
to Napster in the US courts. Morpheus’s parent company was based in
Nashville, but Kazaa’s main offices were in Amsterdam, and Grokster’s on
the island of Nevis in the Caribbean, thereby making any copyright litiga-
tion more difficult.
By August of 2001, over 3 billion files had been downloaded through these
Napster “clones.”xxix Regardless of corporate location, the RIAA decided to
sue these next-generation services for copyright infringement.xxx This time,
the record labels were joined by the Motion Picture Association of America
(MPAA), as movies had become just one more “file type” in the catalogue of
these services. Connectivity had advanced to the point where it had become
feasible to download the very large files that constituted an entire compressed
movie—roughly one gigabyte per hour of video. The lawsuit against Kazaa,
Morpheus, and Grokster, filed in October 2001, was destined to make it all
the way to the US Supreme Court—but not before some significant twists
and turns along the way.
234 Key Changes
First, the US district court found in 2003 that file sharing software was
not illegal, according to the “substantial non-infringing uses” standard es-
tablished in the Betamax decision, even though some estimates found
that 90 percent of the files downloaded via Grokster were copyrighted.
Furthermore, the developers of these programs were held not liable for the
copyright infringement committed by their users. Things did not go better
for the RIAA and the MPAA in their appeal to the Ninth Circuit. A year later,
that court ruled yet again that Grokster was not liable for contributory or vi-
carious copyright infringement.xxxi
Given the enormous threat represented by these file-sharing programs,
rights holders were not prepared to give up. Not only did they appeal the
case to the US Supreme Court, they also decided to make the most con-
troversial legal move of the download era: suing individual users.xxxii The
technology advocacy organization Public Knowledge, siding with the p2p
software companies, supported “strategically targeted legal action against in-
dividual infringers,” though they did not believe that large-scale suits would
be effective.xxxiii
Public Knowledge was correct. In fact, the lawsuits turned into a public
relations nightmare: teenagers and grandmothers were caught up in these
cases, becoming liable for tens of thousands of dollars in damages, making it
a “David versus Goliath” story that the press ate up with relish. The fact that
the RIAA did not have access to personal details about the file-sharers before
taking legal action did not assuage the criticism. But at the time the RIAA
implemented its plan, the courts had ruled out meaningful action against the
p2p software developers.
The US Supreme Court ruled on MGM Studios, Inc. v Grokster, Ltd. in June
2005. In a unanimous reversal of the lower court rulings, the Court found for
the plaintiff labels and studios.xxxiv The decision did not turn on the Betamax
test of whether or not p2p software was capable of substantial noninfringing
uses. Instead, the ruling found that Grokster had “induced” their users to
commit copyright infringement. The court pointed to several areas where
Grokster took active steps to encourage its users, including targeting former
Napster users to join Grokster and advertising and instructing how to
infringe.14
The ruling took many in the industry by surprise, given the trajec-
tory of recent copyright rulings, and the reaction was almost immediate.
Grokster shut down and settled with the media industry, agreeing to pay
Downloads 235
an undisclosed amount to the movie studios and $100 million to the four
major labels. The RIAA sent a copy of the decision to the other p2p services,
emphasizing the new “inducement” standard established by the Court. Many
services simply closed their doors. Some attempted to “go legit” by filtering
unlicensed content from the service through the application of acoustic fin-
gerprinting technology to identify copyrighted works. This proved to be as
problematic as it had been for Napster—not so much because of the accuracy
of the fingerprinting technology but more because filtering out copyrighted
songs meant that afterward the remaining catalog was insufficient to retain
many users. Even when some independent labels decided to allow these
second-generation p2p services to offer their songs, it wasn’t enough to build
a user base.
A few second-generation p2p services tried various tactics to strike a
balance between respecting copyright and offering access to major-label
material. Kazaa, for example, tried to persuade their millions of users to
switch to paid downloads by providing 30-second samples in response to
song searches and then directing users to sign up for a fully featured li-
censed service. Two other startups, Weed and Peer Impact, tried to imple-
ment services in which users could sell DRM-encrypted files to each other
over a p2p network, with rights holders getting a commission on each
sale. Yet another startup called Bitmunk tried p2p file-sharing with dig-
ital watermarks embedded in music files that contained user IDs, presum-
ably so that users would only share files with people they knew and trusted.
None of these ideas resonated with either major labels or very many users.
See Chapter 10 for more on this.
As other second-generation p2p networks ceased operating because of
legal pressures or failed attempts to shift to licensed models, LimeWire
continued to thrive and was ready to absorb the millions of users still on
the hunt for a free option. NPD, a market research firm, estimated that
at one point, LimeWire was responsible for 80 percent of all of the illegal
downloads in the US. In August 2006, the RIAA filed a $1.4 billion cop-
yright infringement suit against LimeWire. It took four years to resolve
the case, but, relying on the inducement standard established in Grokster,
the music industry won. LimeWire and the labels reached a settlement,
and Mark Gorton, the founder of LimeWire, agreed to pay $105 million in
damages.xxxv
236 Key Changes
pursued did not have much chance of success, for two main reasons. First,
the major labels divided their efforts—and catalogs—between two incom-
patible services. Second, the labels codified a set of complicated and restric-
tive usage rules that would be enforced via DRM. Getting users to pay for the
“privilege” of having a poor user experience simply could not compete with
the unencumbered and free world of MP3 files.
Universal Music and Sony Music teamed up to create a joint venture
called pressplay, which launched in December 2001. The other three major
labels at the time, EMI, Warner Music, and BMG, did license some of their
music to pressplay, but the catalog was far from complete. In a move that one
could argue was ahead of its time, any song in the catalog could be streamed.
However, the streams were low quality and were capped at 300 listens total
per month. Buyers got 30 song downloads per month, but not every song
in the catalog was available for this feature. In addition, the downloads all
expired at the end of the month and were no longer playable. None of the
streams or downloads could be transferred to any portable music player.
Instead, to enable listening away from the PC, ten songs per month could
be burned to a CD, but those ten songs could include no more than two
tracks from the same artist—effectively preventing duplication of an album.
Microsoft’s Windows Media DRM technology ensured that all these rules
were enforced. The service cost $9.95 per month.
One reason BMG, Warner Music, and EMI held back some of their songs
from pressplay was that they wanted to support their own joint venture with
streaming audio technology pioneer RealNetworks called MusicNet. While
pressplay was a direct-to-consumer retail service, MusicNet was a wholesaler,
a “white label” infrastructure that companies could use to offer retail serv-
ices with their own brand names. AOL Time Warner (the corporate owner
of Warner Music Group at the time) and RealNetworks were the first two
companies to launch services based on MusicNet. Real Networks branded
its service as RealOne, and it had access to over 100,000 tracks. MusicNet
initially used a DRM technology from RealNetworks that ensured that files
could not be burned to CDs or copied to portable MP3 players; it was in-
compatible with pressplay’s Microsoft-provided technology. The user’s music
experience was capped on this service as well: the $9.95 per month price got
you 100 downloads and 100 streams per month.xlii
Although pressplay and MusicNet were different services with different
participants and different business models, there was some overlap between
the two. EMI, the smallest of the major labels, licensed its content to both
Downloads 239
services. The other point of commonality had to do with the way that the
musical compositions embedded in the sound recordings were licensed. At
the time, the National Music Publishers Association (NMPA)—the equiv-
alent of the RIAA for music publishers—owned the Harry Fox Agency
(HFA), which was almost-but-not-quite a “one stop shopping” service for
obtaining the so-called mechanical licenses to musical compositions that
online music services needed. The RIAA made a deal with the NMPA in
which all major label–owned services would license musical compositions
through HFA.
But this first iteration of major label–affiliated services was doomed
from the start. Even if you asked someone who had decided to pay for one
of these services, chances are they could not have remembered all the rules
and restrictions. That was the crux of the problem. There were suspicions at
the time that major-label management set these complex, Byzantine rules to
discourage adoption, to avoid cannibalizing the CD sales that their bonuses
depended on. Meanwhile, the competition was just a “click away” using a p2p
client; one could download any song (and we do mean any) for free, keep it
forever, burn it to CD, and pay the bargain price of zero. In 2006, PC World
rated MusicNet and pressplay as tied for ninth place on their list of the “25
Worst Tech Products of All Time” and described the features as “stunningly
brain-dead.”xliii
The music labels’ joint efforts were compromised in terms of user experi-
ence, but the US Department of Justice (DoJ) became concerned about anti-
trust implications. Given that the labels constituted an oligopoly, it raised the
possibility that they could work together to squeeze out others in the market
even though neither of the two services involved all of the majors. The DoJ
started an investigation, which Napster tried to use as evidence during its
court battle. The argument did not succeed, but Judge Patel did observe that
the labels coming together in that way “looks bad, sounds bad, and smelled
bad.” Pressplay and MusicNet didn’t stand a chance—although as we’ll see
shortly, MusicNet did stick around for a while strictly as a wholesale service
provider.xliv
The first truly successful commercial digital distribution effort didn’t
come until 2003. By then, the iPod was the most popular MP3 player in the
world with over 700,000 units sold. Steve Jobs became convinced that an on-
line store was needed to sell songs to fill up those miniature hard drives (and
trade up to larger models when they were full). That same year, the music in-
dustry was staring at revenues declining 20 percent from the industry’s peak,
240 Key Changes
with no bottom in sight. The labels were ready to agree to terms with Apple
that they would not have even contemplated a few years earlier.
By building commerce right into the music management software which
already had the capability to transfer files to the iPod, Apple had created a so-
lution that could compete with Napster on simplicity. Steve Jobs personally
demonstrated a beta version of the iTunes Music Store to key executives at
the labels to drive that point home and to persuade them to agree to finan-
cial terms that had a chance for success. Paul Vidich led the strategy team at
Warner Music Group, which was the first major label to come to terms with
Apple. He remembers seeing the demo and thinking, “This is so simple. It
works. It’s great.”xlv Any song in the store could be purchased individually
and not necessarily as part of an album. This represented a major shift in
policy, but it was the only way to compete effectively with Napster, which
already had broken up the album as a unified entity for distribution. Vidich
says he suggested the $0.99 per single price, which Jobs accepted readily (and
took credit for later), as it was far lower than the more than $3 price tag being
proposed by Sony.xlvi,xlvii Digital albums were set at $9.99 each.
The labels all agreed to deals with Apple but kept it on a short leash. The
terms were only for one year. They covered the United States only, and the
store was limited to Mac computers. The files were compressed using the
MPEG4-AAC codec, which Apple touted as providing superior quality to
MP3. DRM was still needed to satisfy the labels’ concerns about security.
Apple bought a small startup called Veridisc, which had built a DRM tech-
nology called FairPlay, and Apple incorporated this into iPods’ firmware and
the iTunes software. FairPlay was (initially) much simpler technology than
the Microsoft DRM used in pressplay and MusicNet, and Apple’s implemen-
tation enforced looser rules: purchased tracks could be transferred to up to 5
iPods and burned onto an unlimited number of CDs.
Some people objected to the use of any DRM solution, but, despite that
concern, the simplicity of the end-to-end solution carried the day. The store
sold a million tracks in its first week, in April 2003. By the fall, the total had
grown to 10 million, and the labels agreed to expand to a broader set of coun-
tries and to Microsoft Windows–based PCs. By the end of its first year in
business, sales totaled 50 million tracks, and the industry finally had a suc-
cessful digital distribution alternative to the p2p services.
The iTunes Store was the only place on the Internet to get music from
major labels for the iPod legally. Apple’s steady stream of iPod enhancements
and upgrades, along with its masterful marketing, including the iconic and
Downloads 241
the rise of Napster. Amazon was especially attractive to the labels because of
its massive trove of customer and sales data. But Amazon needed something
extra to be persuaded to sell digital music; it would not be yet another online
store that sold $0.99 downloads that wouldn’t play on iPods.
The solution was simple, even if it was a bit hard for the labels to
swallow: Amazon would need to sell DRM-free files. In mid-2007, Amazon
reached agreements with Universal Music Group and EMI to sell lim-
ited catalogs of DRM-free music. The AmazonMP3 store launched in
September 2007, the other majors followed with licensing deals, and soon
Amazon offered a catalog of DRM-free music from majors and indies that
rivaled Apple’s. Eventually the labels agreed to let Amazon offer free MP3
downloads of albums that users bought on CD; the so-called AutoRip fea-
ture became available for the vast majority of recent titles in Amazon’s
catalog.xlix
Some in the industry saw DRM-free as inevitable during the run-up to
Amazon’s September 2007 launch. One of those was Steve Jobs. In February
2007, he posted an “open letter” called “Thoughts on Music,” in which he
exhorted the industry to give up on DRM.l Many people took this at face
value and praised Jobs for his vision, especially when it became reality a
year later. But “Thoughts on Music” was really just another example of Jobs’s
legendary Reality Distortion Field: he saw the momentum for DRM-free
coming down the tracks and decided to get on that train despite the value of
consumer lock-in.17
In any case, the quest to create a secure digital distribution system that
had begun more than a decade before was essentially over. The music in-
dustry finally conceded that for digital downloads the open MP3 and AAC
file formats had won the day.18
The download era is a simple story when it comes to money. The industry
gave back all the gains it had realized during the years of CD dominance.
By 1991, CDs accounted for more than half of the total US recorded music
revenues, which were $7.8 billion that year. With CDs fueling growth, the
industry peaked at $14.6 billion in 1999. By 2009, revenues had fallen back to
$7.8 billion again. That wasn’t even the bottom of the trough, as revenues fell
below $7 billion in 2014 and 2015.
244 Key Changes
sales. The labels tried various approaches to protect album revenues even if it
meant constraining downloadable single sales.
One example is Shakira’s English-language album Oral Fixation, Vol. 2,
which Epic Records released in 2005. Neither the album nor the first single
“Don’t Bother” lived up to expectations, with sales of 400,000 units in the
crucial Christmas season. Epic decided in early 2006 that it needed a crea-
tive approach to get her career back on an upward trajectory. Shakira and
Wyclef Jean went into the studio and came out with “Hips Don’t Lie,” which
became the biggest hit of Shakira’s career. The album was rereleased to phys-
ical stores with the hit single added, but not to the online retailers. No down-
loadable version of “Hips Don’t Lie” was available for purchase. But CD sales
did not take off despite the inclusion of the most played track on radio that
year. However, when the single was finally released digitally in mid-June, it
shot to no. 1 on the download chart and has sold over 13 million copies since
then, with the album selling a million copies.lii,liii
The labels also tried adapting some of the same marketing and promotion
techniques that had proved successful in the physical world to drive album
sales. The album images on the main page or the Rock page of the iTunes
Store became desirable real estate that helped to improve awareness and to
increase sales. The labels worked their relationships with Apple to get these
spots, just as they had previously lobbied retailers for album slots on the
endcaps of store aisles. Revenue per square foot as a key measure effectively
became revenue per pixel.
Brick-and-mortar stores that sold records wanted special edition albums
that included bonus tracks exclusive to them as a way to drive foot traffic to
their stores. Similarly, the labels created deluxe versions for iTunes that in-
cluded tracks and other digital items that were designated as “album only”
to encourage fans to pick the album over singles.19 For example. Ace of Base
released Hidden Gems in 2015; fans purchasing it on iTunes received two ad-
ditional tracks and a 15-page “digital booklet.”liv
A variation on the bonus track idea that was specifically geared toward
the download environment was quite successful. When a release was made
available for preorder before the official street date, there was the option to
offer fans an “Instant Gratification” track (or tracks). When the user placed
a preorder, they got immediate access to download that track before it was
officially for sale.
Apple went even further and introduced a new album format called
iTunes LP in 2009 to add bonus content like liner notes, lyrics, and videos
246 Key Changes
when he ran Seagram. The newly independent Warner Music went public
in 2005 and then was privatized yet again in 2011 when Access Industries,
owned by Ukranian-born billionaire Len Blavatnik, purchased the com-
pany. For the sake of completeness, WMG went public for the second
time in 2020. Vivendi sold a minority stake in UMG to the Chinese firm
Tencent Entertainment before it listed the music company separately on the
Amsterdam Stock Exchange in 2021.
This litany of corporate machinations meant that by the time Spotify
entered the US market in 2011, the number of major labels had been cut in
half. The Big Six had become the Big Three: Universal, Sony, and Warner.
These consolidations were intended to create efficiencies to help survive the
industry’s decline by exploiting larger catalogues while eliminating duplica-
tive functions and cutting staff.
The labels hoped that despite the revenue pie shrinking, the shift to dig-
ital distribution represented an opportunity for a broader swath of artists to
achieve success. The “Long Tail” theory, put forth by Chris Anderson in his
2006 book of that name, examined the implications of the Internet removing
the constraints facing physical retailers such as Wal-Mart or Best Buy in de-
termining which products should be displayed on their shelves and stocked
in their storerooms. These brick-and-mortar stores could make only a lim-
ited product selection available, so it was no wonder that so few titles—
whether albums, movies, or books—captured the vast majority of revenues.
The theory predicted that the infinite shelf space of online stores would allow
a much greater percentage of sales to go to a wider variety of goods created by
a broader spectrum of artists.
The reality, however, did not live up to these expectations. A study
conducted in 2008 found that 80 percent of the tracks available online sold
exactly zero copies. Furthermore, 80 percent of the revenues garnered by the
one-fifth of the tracks that did sell went to only 52,000 songs.lviii In music,
the best-known artists continued to reap the majority of the rewards.lix
This would not be the last time that smaller, independent artists would see
Internet platforms as overpromising and underdelivering for them.
The artist reaction to the MP3 format and to Napster in particular was
decidedly mixed. Instead of lawsuits, some musicians, with the support of
Downloads 249
their labels, tried to use file-sharing technology itself to push back on unau-
thorized downloading. They placed so-called “spoof files” on the networks.
These files were crafted to look like full-length songs but instead contained
only snippets that repeated over and over again. The hope was that by adding
some friction to what was a very simple process, users would decide to pay
instead.
For example, before Madonna’s album Music was released, unfinished
portions of the title song showed up on Napster. So as the clock ticked down
to the release of her next album, American Life, Warner Brothers Records,
Madonna’s label, flooded the file-sharing networks with spoof files to deter
piracy. These fake files contained no music, just Madonna saying, “What the
fuck do you think you’re doing!”lx The approach did not succeed, as copies
of the new album still turned up online before the release date. In fact, some
people were so upset with the tactics that they hacked Madonna’s website and
placed copies of those leaked tracks for download right there.
Unlike Metallica and Madonna, other musicians jumped onto the MP3
bandwagon despite the lack of compensation. The Beastie Boys made live
tracks and unreleased songs available for free to promote the release of their
album Hello Nasty. They Might Be Giants offered entire albums. Chuck D of
Public Enemy saw MP3 files on the Internet as a path for artists to create direct
relationships with their fans and disintermediate the labels. Free of his label
deal with Def Jam Records, he told Rolling Stone that “digital distribution levels
the playing field” for artists to maintain greater control over their work.lxi
In general, the labels pushed back on these artist-driven efforts even if they
helped to promote new releases to a growing file-sharing community pas-
sionate about music. They felt that this group of fans was less likely to pay and
that artists should not be encouraging greater adoption of the MP3 format
before acceptable e-commerce solutions were in place.
Artists who examined royalty statements saw their income declining as
less lucrative single-song downloads failed to compensate for the payments
from the CDs they used to sell. Several very high-profile artists, including
the Beatles, simply refused to allow their music on the iTunes store. In the
case of the Fab Four, their reluctance related to the business model as well
as outstanding trademark issues.21 Seven years after launch, the parties re-
solved their differences and the Beatles’ music finally became available on
iTunes. Some other artists, like AC/DC and Garth Brooks, informed Apple
they would hold out unless Apple offered their music on an “album-only”
basis; no single tracks allowed. Apple refused to concede.
250 Key Changes
As we’ve seen over the course of recorded music history, even when out-
side forces created a new way to experience music, the industry had the clout
to decide whether or not their songs or artists would be used to support the
new format, and the law would often help them to enforce those decisions
expeditiously. The decisions to experiment with the early secure download
252 Key Changes
start-ups on a very limited basis or to support SDMI were rooted in the belief
that the levers they controlled would lead to positive results.
Yet relying on control in the era of MP3 files and p2p networks was like
chasing a mirage. Licensed services with restrictive DRM rules appeared to
preserve that control, but consumer uptake was limited. The widespread dis-
semination of MP3 encoding software meant that every fan could create a
replica of their entire music collection and make it available to others. The
labels no longer decided which tracks were offered up, the fans did. No need
to complain about the need to buy the entire album to get the song or two you
wanted. And if a user downloaded those songs using Morpheus, they were
free. The labels’ pricing power had evaporated, and leaks of newly released
songs would upend the their label carefully laid marketing plans.25
In a CBS News survey conducted in 2003, 58 percent of Americans who
were aware of file-sharing over the Internet considered it an acceptable prac-
tice, and that percentage went up if only people aged 19–29 were included.lxvi
Even if people viewed the behavior as somewhat sketchy, the argument
quickly turned to “who was being hurt.” There was not much sympathy for
artists and the elaborate lifestyles shown on MTV Cribs, or for the wealthy
record company executives who appeared to be technology Luddites.
In general, fans were far less concerned than artists about the sound quality
of downloaded songs. As with cassettes in the 1970s, convenience and price
trumped audio quality. Users’ view of quality related more to data errors in
the file than to how the song itself sounded when played back. “Poor quality”
files on the p2p networks could contain a burst of noise or a period of si-
lence that was not in the original song because of issues introduced during
encoding by individuals.26 Apple and the other licensed stores put quality
assurance processes in place to avoid these kinds of issues.
Jonathan Berger, a Stanford University Professor of Music, surveyed his
incoming students each year to get their reactions to various music formats
including MP3. In 2009, he noted “not only that MP3s were not thought of
[as] low quality, but over time there was a rise in preference for MP3s.” He
believed that people came to like the sound of MP3 files as they became more
familiar with it.lxvii
If you had asked a college student about the size of their music collection
in the late 1980s, the answer might have been something like, “I have 200
albums,” referring to vinyl records and CDs. After Napster, people started
measuring their music collection in gigabytes rather than albums or tracks.
That made it easier to determine how much of their music could fit on their
Downloads 253
portable MP3 player and if they needed to upgrade to the latest iPod with an
even larger capacity.
The file-sharing experience raised consumers’ expectations. Fans ex-
pected to easily find any song they desired. And any song they desired had
to be available to download when they wanted it. To meet these heightened
requirements, the catalogs of the online stores had to be enormous. Using
these platforms had to be drop-dead simple. The timings of releases as part of
carefully crafted marketing campaigns went out the window. Even when all
those factors were considered, it was impossible for the legitimate options to
match the file-sharing price of zero.
Paying for music had become completely optional. Fans could decide
which music to buy and which to take for free based on their own sense of
connection to the artist and a desire to support their creativity. They could
decide to purchase concert tickets or merchandise as a means of sup-
port rather than paying cash for the content itself. One study conducted
by American Assembly, a public policy forum associated with Columbia
University, showed that file-sharers did purchase more music than non-p2p
users.lxviii The study was interpreted incorrectly as supporting the narrative
that consumers who shared files were creating additional value for the in-
dustry rather than hurting the business.27
Attempting to communicate the economic harm being done to art-
ists was fruitless in the face of studies like that one and the headlines
about suing file-sharers. The language used by both parties exemplified
the gap in perceptions. The record labels referred to downloading music
for free as “piracy,” and strictly speaking, one definition of piracy in most
dictionaries is the act of reproducing or disseminating copyrighted ma-
terial. However, to people trading files online, this was simply “sharing,”
and weren’t we all taught in kindergarten that sharing was good? When
you share something you typically have less of the item afterward. In p2p
file-sharing, you never gave up possession of any of the music that you
started with.
To some degree, the language and attitudes were simply a rationalization
for behavior that seemed less problematic when conducted online. People
who would never dream of going into a record store and walking out with
a CD they had not paid for had no issue with copying the entire catalog of
Led Zeppelin using Napster or Kazaa. By the iPod’s third generation in 2004,
there were models that could hold 10,000 songs; using p2p networks to fill
that hard drive at zero cost was a completely rational economic decision.
254 Key Changes
The openness of the MP3 format and the availability of the next file-sharing
platform, even when your favorite software was found to be infringing
copyrights, gave music fans the ability to decide, song by song and artist by
artist, when to play by the industry’s rules. As Michael Robertson, CEO of
MP3.com said: “The consumer will have the last word. The real wild card is
the Internet because it empowers consumers.”lix
Conclusion
Downloadable music brought an abrupt end to the enormous sales and exor-
bitant profits of the Compact Disc. Though the number of music transactions
climbed higher than ever before, the revenues from those sales marched
steadily downward for the entire first decade of the twenty-first century.
Reversing that decline was going to require a fundamental change in the
music experience. The “celestial jukebox” had been predicted for years and
it was poised to become a reality at last because of advances in technology.
9
Streaming
Islands in the Stream—Dolly Parton and Kenny Rogers
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0009
256 Key Changes
year over four years. By 2022, China had continued its ascent to become the
fifth largest music market in the world. During the second quarter of 2020,
Tencent’s revenues alone approached $1 billion as paying users grew more
than 50 percent year on year.iv,v That growth—and the enormous valuation
for Tencent Music—were driven by the shift to music streaming. Instead of
buying music as a product, hundreds of millions of fans around the world
now access “Music as a Service”1 in exchange for a monthly subscription fee,
and billions more stream for free in exchange for listening to ads.
Of course, in terms of global growth, the leader was not Tencent but
Spotify. Its revenues in the final quarter of 2022 exceeded $3.4 billion on
a base of 489 million active users and 205 million premium subscribers,
dwarfing those of Tencent Music.vi That Swedish company established a suc-
cessful beachhead in the Nordics and used that experience to negotiate the
licenses needed to enter the US market in 2011. The successful public listing
of its stock in April 2018 gave the company a valuation of almost $30 billion
and no doubt encouraged Tencent to follow in their footsteps.vii
Spotify, Tencent, and a host of other streaming music services, including
those from some of the most valuable companies in the world—namely
Apple Music, Google Play/YouTube Music, and Amazon Music—serving
music to more than a half a billion paying subscribers have led the music in-
dustry to six years of sustained revenue growth for the first time since 1999—
the year Napster launched (see Figure 9.1). In 2022, the on-demand music
streams in the United States alone exceeded 1 trillion.viii
Though the industry resurgence of the past several years has been driven
by streaming adoption, the ability to offer a “Celestial Jukebox” was already
part of the conversation around the turn of this century. In the December
2000 issue of the now defunct INSIDE magazine, Charles Mann wrote an
article entitled “The Hot New Bad Idea.” His thesis was summarized as
follows: “What an amazing notion. Every song ever recorded, delivered to
you digitally via any device, paid for with a simple subscription fee. Everyone
from AOL to Vivendi is touting the ‘Celestial Jukebox’ as nirvana for an in-
dustry under siege from Napster. One small problem: It’s not going to work.”ix
Fortunately for the music labels, the reality has proved to be closer to nirvana
than this prediction.
A proof of concept already existed for that Celestial Jukebox. In April
1998, at roughly the same time that the early music download platforms were
vying for customers and content, the Rock & Roll Hall of Fame and Museum
in Cleveland opened an updated Hall of Fame section with signatures of
30
0.5
25
2.4
0.6
0.7 0.5 1.1
0.8 0.5
20 0.9 2.3
0.4 0.9 0.5
1.0 1.0
2.6 1.2
0.1 1.2
2.0 0.4 2.7
0.2 1.3 0.4 1.5
2.7 0.3 0.3 0.3 2.4
15 1.3 0.3 0.4 1.7
0.2 0.3 2.3
3.4 1.4 1.5 1.6 1.8 2.0 2.6 16.9
3.8 1.9
24.1 0.3 3.2
22.6 23.3 3.9 4.3 13.6
21.7 0.4 4.4 3.8
10 20.0 19.4 4.4 4.0 11.4
18.1 0.4 9.3
16.5 0.6 6.6
1.0 4.7
14.2 1.4 2.8
1.9
12.0
5 10.5
9.1 8.3 7.6 6.8
6.0 5.8 5.6 5.2 5.0
4.7 4.5 4.3
0
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Total revenue
24.1 22.6 24.0 22.4 20.8 20.8 20.2 19.6 18.4 17.1 15.9 15.1 15.0 15.0 14.7 14.2 14.7 16.1 17.3 18.9 20.4 21.9 25.9
US$ billions
Total Physical Total Streaming Downloads & Other Digital Performance Rights Synchronisation
Figure 9.1 Global Recorded Music Industry Revenues 1999–2021 (US$ Billions). IFPI Global Music Report
2022. Used by permission of IFPI.
258 Key Changes
consumers on the value of this new model required a large marketing budget,
and Rhapsody did not have the resources. Adding a mobile experience even-
tually enabled streaming to meet the expectations of millions of music fans
and to achieve mass market adoption.
The smartphone in general, and Apple’s iPhone in particular, brought
a wide range of innovative capabilities right into the hands of billions of
consumers. These powerful devices encompassed everything needed to
make streaming services a great experiences, thus providing Spotify and its
ilk with the runway needed to take off. App stores gave music fans an easy
way to obtain and to pay for a streaming service as well as the means to up-
date the service easily to offer new features. The touchscreen improved the
user interface, which allowed subscribers to scroll through and select songs,
albums, artists, and playlists much more easily than the iPod’s scroll wheel.
Every single day, 100,000 new tracks (worldwide) are added to the already
enormous catalogs of tens of millions of songs available for streaming on
these services.xii And unlike the complexity of the download era, a subscriber
can access any of those songs by touching the screen of the device in their
hand a few times.
The widespread availability of affordable high- speed Internet band-
width was critical to the adoption of smartphones for communications and
entertainment. In the Chapter 8, we observed that the explosive growth
in broadband connectivity to the home in the 2000s let millions of music
fans download entire albums in seconds rather than waiting more than 10
minutes for a single song to arrive on their PCs. The 3G cellular data standard
was introduced in 2001; it offered download speeds of 5–10 megabits per
second (Mbps). The 3G penetration rate was limited because of the costs
of the associated data plans. In many cases, corporations covered the costs
of their employees’ data plans to enable e-mail connectivity on those then-
ubiquitous Blackberry devices.xiii,xiv But average consumers chose to keep
their wallets in their pockets or purses rather than making the move to mo-
bile broadband. That is, until the iPhone swept in (followed closely by similar
Android-based smartphones) and dominated the market for mobile devices.
Apple’s creation of yet another “must-have” device unleashed the desire
among millions of users for broadband data on a mobile phone.
One might think that the uptake in mobile data by 2010 would have been
sufficient to fuel the growth of streaming services from a technology per-
spective. The Celestial Jukebox, with a catalog numbering in the tens of
millions of songs, sat in the cloud waiting for music fans to select a track
Streaming 261
to listen to and then have the bits stream directly to their smartphones for
playback. Though 3G networks could support mobile Internet access that
was fast enough for music streaming, performance was spotty and dropouts
were frequent. Data plans often included caps on the amount of data that
could be used each month, with hefty overage charges. Wi-Fi hotspots that
are ubiquitous now were not yet broadly deployed. Having a slight delay in
loading your emails or loading a webpage was annoying, but it did not com-
pletely undermine the value to the consumer. But if you were listening to
your streamed music on the streets of NYC, heading down into the subway
would simply break the connection. How could one charge $10 per month
for an online streaming service promising access to any song at any time if
the connection was incapable of delivering on that promise?
From an engineering perspective, the solution to this problem was ob-
vious: make copies of lots of songs and store them in the device when con-
nectivity was available and play them back when it wasn’t. Storing copies of
compressed music files had already been solved for the download era. The
desire to broaden the distribution of purchased files beyond Apple had
contributed to the decision to remove DRM, which reduced friction in the
user experience. The labels had decided that allowing unfettered copying of
songs that were already paid for was not without risks, but was, on balance,
a good business decision. Yet that same decision did not make sense in the
world of streaming. If subscribers were permitted to have copies of thousands
of songs stored on their device, the service needed to ensure that access to
those tracks disappeared when a user stopped paying for the service.
Fortunately, mobile phone designers, including Apple and Google, de-
veloped the capability in their platforms that enabled Spotify and other
streaming services to cut through this Gordian knot. Though far less ob-
vious to the user than a high-quality display or responsive touchscreen,
the mobile platforms contained state-of-the-art encryption capabilities.
A Trusted Execution Environment (TEE), a secure area on the main proc-
essor, was designed into most smartphones starting in the early 2010s.
This hardware-based solution provided better security for passwords,
cryptographic keys, and other credentials. They allowed the streaming
companies to store music files securely and to meet the requirement that
any such files would become unplayable should a subscriber cancel their
service as well.
The tracks stored on the device were encrypted using a private key. That
private key was stored in the TEE, and it was used to decrypt the music if and
262 Key Changes
only if the device had a valid license from the service. Each month, when the
user’s mobile phone communicated with the service, it would check that the
user had paid their monthly fee. If so, a new license was issued and stored
securely in the device, thus renewing access for another month. If the sub-
scriber stopped paying, the license eventually expired, and the key to decrypt
the stored tracks was no longer valid.
Unlike the DRM originally used on downloaded tracks, users accepted
this approach. The experience was consistent with their expectations. If
they stopped paying for their wireless service, they expected their mobile
phone to stop working, and it was only logical for the same to be true for
their music. Given the negative reactions to DRM in the download era,
it is not surprising that streaming services avoid mentioning those let-
ters. But as David Hughes, the former CTO of the RIAA, said in 2008,
“Any form of subscription service . . . still requires DRM. So DRM is not
dead.”xv
This solution was the final technology piece needed for a great streaming
music experience on mobile phones. Fast wireless connections meant that
music could be delivered in real time to mobile phones with streaming music
apps ready to be accessed at the touch of a finger. The tunes stored on the
device could fill in the gaps, enabling a seamless experience even when the
network was unavailable or glitchy. And the labels’ concerns about people
continuing to have access to thousands of tracks without paying were
alleviated.
Before we discuss the broad implications of this shift to streaming, it is
important to recognize that this change is fundamentally different from
the many format shifts that preceded it. Whether it was vinyl records or
cassettes or CDs, each of these physical items contained limited amount of
music. Fans acquired that music by paying a fixed one-time fee to purchase
the physical carrier that held the songs from a record store. The physical
embodiment of the music disappeared in the digital download era, but fans
still purchased individual tracks or albums for a set price paid to retailers
through a web storefront. Getting real estate on the iTunes home page was
not all that different in practice from getting space on an endcap at Best
Buy or Tower Records when it came to selling records. To be successful,
artists had to create music that was good enough for fans to pay for. And
music labels had to market that music to drive awareness and to convince
consumers that they had to have a copy—whether on plastic or as a dig-
ital file.
Streaming 263
The value of music streaming to the industry can be told through a few key
statistics. In 2022, US recorded music revenues approached $16 billion dollars,
and 84 percent of that total was generated by fans streaming songs. The total
annual number of on-demand music streams of audio and video is now in the
trillions. The seven-year run of increasing recorded music revenues fueled by
streaming drove the value of song catalogs to new heights, with iconic artists
from Bruce Springsteen to Neil Diamond to John Legend selling their rights
for hundreds of millions of dollars. Universal Music Group, the largest major
label, went public in September 2021 with a valuation exceeding $50 billion.
Though still tiny compared to the tech giants, this represents an enormous in-
crease in enterprise value over the past decade.xvi,xvii
It is impossible to understand the financial impacts of streaming without
getting into the implications of service economics and how it differs from
product business models. Fans in most developed countries pay the equivalent
of $10 per month for unlimited access to all the tracks and albums in the catalog.
For a streaming service, that means a subscriber who logs in can listen to any one
of the many millions of available tracks. But unlike a product purchase, playing
a particular song does not generate a direct payment to the rightsholders.
The following formula is a simplified version of how streaming services deter-
mine the amount of money that goes to the labels and artists for the music played:
Net Revenue means the fees collected from paying subscribers each month,
which are aggregated into a single pool of funds. Share Percentage is the por-
tion of the net revenues allocated to the record labels and music publishers
based on contractual agreements. The remainder of the net revenues is
retained by the service to operate its business. The labels’ share percentage
is approximately 52 percent, and the publishers receive roughly another 10–
12 percent. That resulting amount is then divided among all the labels based
on the Pro Rata share of plays of tracks on each label recorded in the service
that month. A play is tallied only if the track is listened to for more than 30
seconds. For example, if 25 percent of the total plays for the month were songs
by Sony Music artists, then that company receives 25 percent of that label
264 Key Changes
share along with a detailed data feed that includes the number of times each
song was played, which Sony uses to calculate the ultimate artist payments.
The actual calculations are more complicated. Each streaming music
service has multiple service offerings with different prices to consumers, such
as student discounts and family plans. Prices also vary by country, and rates
differ according to local economic conditions. For example, Apple Music is
available in almost 170 countries; a music fan in Bhutan pays less than one
in Boston. In addition, there is a minimum per-user fee, which ensures that
there is a floor to the amount of money going to the revenue pool for labels.
Adjustments for billing fees, sales taxes, annual subscription discounts, and
free or reduced-price trials are factored in as well.
The approach just described lays out, in a general way, how the on-demand
music streaming services calculate royalty payments for paying subscribers.
However, Spotify is somewhat different, because unlike its competitors, it
really consists of two closely coupled service tiers: the premium tier where
users pay $10 per month (or different fees in other countries) and the free tier
where users pay with their attention by listening to advertisements.
When one accounts for the revenues from both paid subscribers and free
users, one can complete the calculations we’ve laid out and then derive an ef-
fective “per play” rate. That rate is a result of the methodology described and
is not an input parameter to the royalty methodology. The effective per-play
rate differs from month to month and from service to service. Furthermore,
if users engage more with the service and listen to more songs, the effective
per-play rate goes down. Recent estimates peg the Spotify per-play rate at
$.0035 and the Apple Music rate at $.00675.xviii Comparing these rates is not
particularly informative. Tidal may have a higher imputed per-play rate than
either Apple or Spotify. But its smaller subscriber base means that they end
up with far smaller total payments to labels and artists.
The lack of a fixed per-play rate and the declining effective per-play rate
as the audience for streaming grows contribute to the frequent complaints
about inadequate compensation from streaming. In a 2013 interview, Thom
Yorke, the lead singer of Radiohead, colorfully described Spotify as “the last
fart of a dying corpse.”xix Taylor Swift took more direct action and removed
her catalog from Spotify in 2014 because she felt the service was inadequately
compensating artists, though she did return her music to Spotify in 2017.
Her concerns were not limited to Spotify as she chose not to release her 2015
album on Apple Music for the same reason. When questioned at a confer-
ence in 2019, Jim Anderson, a former Spotify executive, described artists as
Streaming 265
“entitled” because the service was not created to make artists money. He went
on to say, “I think Taylor Swift doesn’t need .00001 more a stream.” Needless
to say, this sort of public posture has not exactly ameliorated performers’
concerns.xx
The calculation of the pro rata share based on the aggregated plays of
millions of users is perceived as unfair because the money paid by a sub-
scriber who listens to music from a single indie artist alone does not directly
benefit that artist. As an alternative, Soundcloud and Deezer are adopting
a “fan-centric” model where the net fees paid by an individual are paid out
only to those artists that they listen to. In July 2022, Warner Music Group
became the first major label to agree to use the Soundcloud methodology
to calculate royalties for its artists.xxi In early 2023, Universal Music CEO
Sir Lucian Grainge stated that a new “artist-centric” scheme was needed for
streaming payments and, just a few weeks later, his company announced
they were examining alternatives with Tidal and Deezer.xxii Spotify has
stated that they would consider changes but that they “require[] broad in-
dustry alignment.”xxiii Without that, we can expect the issue of inadequate
streaming royalties to continue to spark debate.3 And Spotify’s decision in
March 2022 to spend an estimated $300 million on a sponsorship deal with
the Barcelona Football Club does little to assuage artists’ concerns about un-
fair compensation.xxiv
and difficult. Like any other relationship between a “wholesale” supplier and
the “retailer” that offers the goods to a consumer, there is a tension between
the retailer wanting to reduce the amount it pays to its supplier for its “goods”
and the supplier wanting to increase that same number.
The value of a streaming service is built on the premise that the monthly
charge offers access to every song. And even though these catalogs now en-
compass over 100 million tracks, individual songs are not substitutable. If
a fan wants to listen to “As it Was” by Harry Styles, one of the most popular
tracks of 2022, there really is no other song that fulfills that desire. And, if a
service lacks that song, any “Styler” (as his fans are known) is unlikely to feel
their payment is worth it. Amazon took the approach of licensing a limited
catalog with lower costs for its launch of Amazon Prime Music. The service
was viewed as an added benefit to Prime subscribers, so a comprehensive
music selection was less critical. However, when it decided to offer a sep-
arately priced premium service, Amazon Music Unlimited, it needed to li-
cense the full catalog in the same manner as the other streaming services,
with similar economics.
But not all listening on these services comes from the most popular tracks
on the Billboard charts. People often are satisfied with mood music for
exercising or chilling out. There have been several reports that Spotify paid
producers to create music for its popular Ambient Chill or Sleep playlists
on a flat-rate basis. The artists for some of these tracks were obscure bands
such as Deep Watch and Enno Aare. These musicians were so obscure that
they had virtually no presence on any services other than Spotify. This
strengthened the case that these so-called fake artists were being placed on
popular playlists to attract listens that were financially more advantageous
to Spotify than Taylor Swift songs or other major label content. Spotify has
strenuously denied that it participated in such schemes, but concerns remain
that this sort of behavior remains a tempting path to reduce royalties and
enhance profitability for streaming providers.xxv And as we will discuss in
Chapter 11, Artificial Intelligence will make it even simpler to generate con-
tent with a lower royalty structure.
Another way to reduce costs would be for a streaming service to inte-
grate “vertically” and become a record label through ownership of a mean-
ingful music catalog. It could then distribute that music with more favorable
pricing. Though there have been recurring rumors of just such a major pur-
chase, none of the streaming players have pulled the trigger.4
Streaming 267
terms for content. In 2018, Spotify tried again by announcing that it would
license content from certain up-and-coming artists and legacy artists
who had control of their rights.xxviii Artists could upload music to the service
directly, bypassing labels and other distributors. These moves raised sig-
nificant concerns at the labels. Daniel Ek, Spotify’s chief executive, tried to
assuage those fears during the July 2018 earnings call by saying, “Licensing
content does not make us a label, nor do we have any interest in becoming
a label.”xxix That argument was less than persuasive, and Spotify abandoned
its direct licensing efforts in July 2019 after what must have been very con-
tentious discussions with their main content suppliers. The major labels had
complained vociferously about the service competing with them directly.xxx
We’ve already mentioned that Amazon Music was launched to enhance the
value of Prime membership for its most valuable customers. The margins for
the streaming service on a standalone basis are a rounding error to Amazon
in the larger scheme of driving profitability of their e-commerce business.
Even its full-catalog Amazon Music Unlimited offer, which is priced at the
same $9.99 per month as its competitors, costs only $7.99 if you’re already an
Amazon Prime member, or only $4.99 if you want the “single device” plan for
your Echo smart speaker. Similarly, Apple made the vast bulk of its revenue
from hardware products and has historically considered services such as
Apple Music more as drivers of hardware purchases than as revenue sources
in and of themselves. And Google, analogously, has viewed content services
primarily as drivers of revenue from ads and user data.
Thus Spotify, unlike its largest competitors, was a “pure play” music offer
for years after launch, with no other lines of business to supplement those
revenues. To rectify this situation, Spotify added podcast playback function-
ality to its apps in late 2018; the vast majority of podcasts are available freely
from podcast publishers to any app that wants to list them. By supplying
both types of audio content through a single interface, Spotify embodies a
simpler solution than Apple’s two separate applications (Apple Music and
Apple Podcasts). But they didn’t stop there: Spotify invested upwards of a
half a billion dollars acquiring several podcast studios and publishing tools
as well as rights to individual high-profile podcasts. These acquisitions in-
cluded Gimlet Media, Anchor, Parcast, and Bill Simmons’ sports-related
platform The Ringer. In May of 2020, Spotify licensed rights to the hugely
popular Joe Rogan Experience podcast in a deal estimated to be worth over
$200 million.xxxi,xxxii
Streaming 269
demographic profile than Spotify: over half of Spotify users are aged 34 or
younger, while that statistic sits at only 40 percent for Apple Music.
Google and YouTube can similarly leverage Android and the free
YouTube service to promote its paid subscription option. Recently, Google
invested $4.5 billion into Jio Platforms, the technology arm of Reliance,
the Indian conglomerate.xxxiv The companies will collaborate on affordable
smartphones, but Reliance is also the parent company of JioSaavn, a leading
music subscription service in India. It would not be surprising if those new
devices help to advance the growth of the content offering in markets where
few can afford iPhones or high-end Android devices.xxxv
Though much of the press tends to focus on the competition between
Apple and Google, Amazon, through its Prime Music offer, is not very far
behind in terms of share of paid subscribers. With over 150 million Prime
subscribers worldwide, it has a sizable addressable market for all its content
offerings.xxxvi In November 2022, Amazon made every of the 100 million
tracks in its music catalog available to Prime members as an additional perk,
although the songs can only be accessed in “shuffle” mode rather than fully
on-demand.xxxvii And it certainly helps when Amazon programs its market-
leading smart speakers to ask users if they’d like to upgrade to their Amazon
Music Unlimited services if they request a song that is not included as part
of the basic Amazon Prime bundle. The power of voice interfaces to enable
a superior music experience and Amazon’s dominant share will be discussed
in more depth in Chapter 11.
Marketing tie-ins with companies outside of the music space help to at-
tract additional streaming users with discounts or complimentary offers as
well. Such offers were quite popular in Europe and Latin America for years.
The United States was somewhat late to the party, but this practice is quite
common now. Verizon Wireless includes a free Apple Music subscrip-
tion for its Unlimited data plan customers. Sprint, which is an investor in
Tidal, bundles that service with its top plan as well. AT&T offers Spotify to
its Unlimited mobile package customers as an “entertainment perk.” Not
all these offers are related to mobile services: for example, Spotify recently
offered a $9.99 bundle that gave subscribers access to its service as well as the
ad-supported Hulu video service.xxxviii
However, Spotify’s most effective tool for driving paid subscriber growth is
its ad-supported free tier. The users viewing ads are far less lucrative than paid
subscribers. But the hurdle to get a music fan to try a completely free service
is commensurately lower as well. From the time of its launch, Spotify argued
Streaming 271
that the free tier was the best way to attract users from piracy. It asserted that
offering access with a better experience than unauthorized services would
convince fans to give Spotify a try. And once those fans were in Spotify’s mar-
keting “funnel,” they would eventually see the value in the service and decide
to pay for the premium ad-free tier.xxxix
To say that the music labels, who had been damaged so dramatically by
illegitimate services giving away their music, were skeptical about Spotify’s
free tier would be an understatement. In fact, Spotify’s launch in 2008 was re-
stricted to its home market of the Nordics because the labels were unwilling
to risk broader licenses that could jeopardize their fairly robust download
business. Spotify’s success in its home country of Sweden, both in terms of
attracting people to the free tier and in converting a significant percentage of
those users to become paid subscribers, persuaded the labels to allow Spotify
to enter other markets.xl Repeating those results in country after country
gave Spotify the data to prove that its “freemium” approach did indeed re-
duce piracy and foster paid subscriptions.xli
Spotify’s greatest hurdle was to agree on the terms that allowed them to
enter the US market. According to some sources, Steve Jobs tried to block
their entry into America; and the negotiations were difficult as the labels
wanted to ensure that the free tier was restricted in ways that maximized the
number of paying subscribers.xlii The data showed that a subset of users who
were on the free tier for more than a year or two would still decide, after all
that time, to take the leap and to start paying. Spotify’s conversion rate, north
of 40 percent in most markets, has shattered the expectations that the in-
dustry had for it in the early days.
America is the largest music market in the world, and the adoption of
streaming in the United States contributes mightily to American serv-
ices leading the world in market share (see Figure 9.2). In 2021, global
music subscriptions climbed more than 20 percent year over year to almost
500 million, with Spotify representing more than 30 percent of all these
subscriptions. Apple Music was second with a little over 15 percent of all
subscriptions, and Amazon and Tencent Music tied for third place.
Yet as indicated by Tencent Music’s position, focusing solely on the United
States would be myopic when discussing the distribution channels in the age
of streaming. The biggest players in the United States have global reach. For
example, Spotify is available in 92 countriesxliii while Apple Music services
fans in 167.xliv The list of territories served includes countries such as Russia
272 Key Changes
Other
Yandex 10%
2%
Spotify
Deezer 31%
2%
Net Ease
6%
Figure 9.2 Global streaming music subscription share. © MiDiA Research. All
rights reserved.
that have traditionally generated little to no revenues for artists and labels.5
Deezer, which is based in Paris, beats them both by being active in over 180
countries. Deezer offers a free ad-supported tier similar to Spotify; it has over
20 million active users, with one-third of that total paying for the subscrip-
tion service. All the global players are facing homegrown competitors in vir-
tually every part of the world.xlv
This chapter began by looking at the IPO of Tencent Music based on
its success in the Chinese market. China has over 700 million active users
on its various music services. And China is not the only region with a
service that is scaling up quickly to serve tens or hundreds of millions of
streaming users. Over 300 million Indians enjoy streaming Bollywood
music and international content on local services such as Gaana and
JioSaavn.xlvi Anghami has 70 million users in the Middle East and Africa
and was the first Arab tech startup to go public. In South Korea, Melon
has over 28 million subscription users. Boomplay in Africa has 75 mil-
lion subscribers.xlvii The average revenue per user may be lower in these
markets, but as mobile connectivity improves and cellular data becomes
more affordable, these markets have become significant sources of
growth for the industry. The proposition of accessing the world’s songs
Streaming 273
selection of music on YouTube. Having any music video in the world playable
on-demand for free, albeit with ads, meant that for many people there was
simply no need to pay for streaming music. During the late 2000s, as we’ll see
in Chapter 10, YouTube was rapidly compiling—thanks to user uploads—a
“music catalog” of tens of millions of tracks while the labels were lagging be-
hind in feeding their catalogs even to download services such as iTunes.
Today you can pay for an enormous variety of services and goods by
signing up for subscription services. Affordability may be an issue, given
the panoply of services available, but it seems compelling for millions of
consumers to subscribe to regular deliveries of razor blades, meal kits, or
even clothing items. (Not to mention that every major media brand, whether
it is HBO or NBC, seems to be launching their own new streaming video
platform.) But when streaming music first arrived, the only consumption
models for recorded music were physical products or radio. Persuading fans
that they did not need to hold the music in their hands but could be satis-
fied paying for more ephemeral access was no small feat. The early streaming
services, such as Rhapsody, simply did not have the marketing dollars or
resources necessary to change consumer behavior. Spotify’s free tier, along
with the enormous unpaid media coverage they received were certainly key
factors in establishing credibility for paying for music as a service.
Free trials and bundles with other desirable services could get people
across the starting line and get them to give streaming music a try. But what
ultimately persuades fans to pull out their wallets and hand over their hard-
earned cash to subscribe? And what keeps them paying month after month?
Fans clearly value the ability to select any song that they can think of and have
it start playing immediately. But they also appreciate elements of the service
or applications that go beyond that core proposition—elements that have
become more and more important in differentiating one streaming service
from another.
Music fans frequently prefer more of a “lean-back” experience, where the
service selects the music, harkening back to the days of radio. But rather than
being constrained by the limitations of a single broadcast channel (or even a
fixed set of channels), streaming services have the ability to satisfy the tastes
of virtually any fan. Subscribers can request music by any artist, genre, or
decade. They can find music that is suitable for a long drive or a party. They
may want music for working out or suitable for chilling at home.
The services can build personalized recommendations that their users re-
ally enjoy by relying on the data generated from each individual’s interactions
Streaming 275
over time and marrying it with similar info from millions of others with sim-
ilar tastes. Spotify’s Discover Weekly playlist feature is perhaps the ultimate
expression of this recommendation technique. The product of an internal
hackathon at Spotify, Discover Weekly uses sophisticated data analysis to
find new songs that a specific subscriber is more likely to enjoy. And if the
subscriber skips one of those newly recommended tracks, the algorithms can
use that information to make smarter suggestions the next week. Or the anal-
ysis can identify songs the user liked in the past and remind them to give
that music another listen. The algorithms that generate these personalized
playlists have become increasingly sophisticated over time and represent one
of the first applications of artificial intelligence/machine learning techniques
to the music industry; we will examine these technologies in more detail in
Chapter 11. Based on the results, one would have to judge Discover Weekly
to be a very successful use of these techniques: between July 2015 and June
2020, Spotify users spent more than 2.3 billion hours listening to Discover
Weekly playlists.liii
One can get a sense of how music consumers value playlists by examining
Spotify’s statistics in even more detail. According to the company, its platform
contains over 4 billion playlists in total.liv That number seems large, but if one
considers it in the context of hundreds of millions of active users, it is some-
what less daunting. It is even more instructive to focus on the playlists that
have the greatest number of followers. Twenty-four of the 25 most-followed
playlists are “editorial” playlists that are created and managed by Spotify it-
self. Today’s Top Hits has over 28 million followers, while Rap Caviar, the
influential hip-hop playlist, has almost 17 million.lv,lvi
It is not surprising that having so many individuals following these pop-
ular playlists leads to greater listening time. Spotify-curated playlists con-
stitute about one-third of the total time spent listening, with a little more
than half of that dedicated to playlists that Spotify has tailored to each lis-
tener based on their past behavior. Just over another third of listening time
is consumed by user-generated playlists. Which means that users are actively
selecting individual tracks to enjoy for less than one-third of their time spent
on Spotify.
Apple was late to the game in shifting from downloads to streaming, in
part because of Steve Jobs’s long-standing view that “The subscription model
of buying music is bankrupt.”lvii When it finally launched Apple Music in
June 2015 in over 100 countries, it emphasized the hand-crafted nature of
their service rather than the algorithmic aspects, a holdover from the Beats
276 Key Changes
Music service that Apple acquired (along with the Beats headphone busi-
ness) and used to form the basis of Apple Music. The most hyped element of
the new service was Beats One, an online radio station available around the
clock. Apple poached DJ Zane Lowe from the BBC as the leader of a group of
radio personalities who created programming for the station. Major music
stars such as Elton John and Drake were also recruited to program other
channels as part of the launch. Though they were called “stations,” they were
actually curated playlists of tracks to attract new customers.
As much as consumers listen to the playlists published by the services,
many users want to play a more active role and create their own playlists for
any occasion. To entice those users, the streaming providers make it easy to
add tracks to personal playlists and to share those with friends or broader
sets of listeners.6 And once fans invest the time and energy to craft numerous
playlists to enhance their experience, they are far less likely to jump ship to
another service and then need to repeat all that work just to access those
same playlists.7
The hurdle to convert users of the Spotify free tier represents an even
greater challenge. If music fans enjoy the service and can access music in-
definitely for free and listen to the popular podcasts published by the service,
what would motivate them to pay? The answer is that the free tier is not iden-
tical to the paid service. Spotify degrades the consumer experience in several
ways to make it less satisfying. The gaps between the free and paid tiers in-
centivize users to upgrade to the premium tier, and Spotify tinkers with those
gaps constantly to improve its conversion rate even further.lviii For example,
at this time of writing, free-tier users can’t save tracks for offline listening and
can only listen to certain playlists in order (others can be shuffled). Specific
track selection is limited, as are skips per hour. These restrictions apply to
mobile devices only; the full feature set (plus ads) is available to free-tier
users on computers.
The same song catalog is available to the free and paid audiences, but with
one key distinction. Although the major labels were adamantly opposed to
exclusives, as we described earlier, they do have the right under the more re-
cent contracts with Spotify to “window” new releases. That is, the labels can
select certain new songs or albums to be playable by paid subscribers but not
by free users for up to two weeks. The feature limitations and ads may not af-
fect how much casual music fans enjoy the free version of Spotify. But taken
collectively, they remind frequent users over and over that there is a better
experience just on the other side of the paywall. Those nagging imperfections
Streaming 277
in the service help fuel Spotify’s ongoing success in converting free listeners
to subscribers.
Given the low revenues per user for those on the ad-supported tier, there
is a temptation to clamp down on the free experience further to try to en-
courage greater premium adoption. Spotify has been able to demonstrate
that such tactics can backfire. In 2011, for example, Spotify restricted ad-free
listening in five European countries, including Spain, to satisfy the demands
of certain rights holders. Individual tracks could only be listened to five
times, and the total listening hours was capped at 10 hours per month for free
users. Conversion rates dropped significantly, and the restrictions were soon
eliminated.lix Instead of implementing drastic changes, Spotify’s formula for
success improves conversion rates via small, incremental shifts backed up by
careful analysis.
On the other hand, windowing content yields positive results even in
markets where paying for online music is a relatively new behavior. Tony Yip,
Tencent Music’s Chief Strategy Officer, explained that “approximately 10%
of the streaming volume on our platforms” was of music windowed for pre-
mium users only. And he expected that this statistic would grow over time.lx
We discussed the long tail theory in Chapter 8; many viewed this as
a law of the Internet rather than simply a hypothesis. But even with the
increased selection in download stores, fan interest did not shift mark-
edly away from the superstar artists who had dominated every format
that had come before. The move to streaming, with its virtually limitless
catalog available to any subscriber at the touch of a button, seemed like
the perfect setup to drive a change in demand. However, the data showed
just the opposite: 80 percent of the streams were coming from just 3 per-
cent of the songs, while millions of tracks were never listened to at all.lxi
The week after Drake released his Honestly Nevermind album in June
2022, the Canadian rapper accounted for one out of every 60 streams in
the United States.lxii Despite the theories, the hits are still the hits, and
artists still want to climb up to the rarefied air and join Drake at the top
of the charts.
Moreover, recent data indicates that the hits fans are listening to are older
than they used to be. According to MRC/Nielsen, album consumption of
catalog music rose from just over 60 percent in 2018 to over 66 percent
in mid-2021.lxiii The official definition of “catalog” includes any mate-
rial released more than eighteen months ago. That includes music from
one-name icons such as Elvis, Madonna, and Prince, but it also includes
278 Key Changes
will play a hint of the chorus in the first five to 10 seconds so that the hook is
in your ear, hoping that you’ll stick around till about 30 seconds in when the
full chorus eventually comes in.”lxvi According to Midia Research, the top 10
hits have an average intro of 7.4 seconds today versus just over 13 seconds
20 years ago.lxvii
Creators are modifying more than just the introductory sections of tracks
for optimal performance on streaming. As discussed in Chapter 3, radio sta-
tions preferred shorter versions of songs to maintain interest. For streaming,
every track that is listened to for more than 30 seconds counts as a play, but
making it all the way through the song is a factor in future recommendations
as well. As the Grammy-winning producer and performer Mark Ronson
said in an interview in The Guardian, “All your songs have to be under three
minutes and 15 seconds because if people don’t listen to them all the way
to the end they go into this ratio of ‘non-complete heard’, which sends your
Spotify rating down.”lxviii
Blogger Michael Tauberg examined the data in detail and observed that
the average song length for the top hits has been declining since the turn of
the twenty-first century.lxix From over four minutes at the turn of the century,
the top tracks now are now more than a half minute shorter. He observed that
the number of words in song titles has been decreasing as well. Examining
the songs occupying the no. 1 chart position reveals an even more dramatic
result: nearly two-thirds of the songs that achieved that lofty spot in 2021
(through August) were under three minutes long.lxx Ironically, these limited
duration tracks would have fit comfortably on the early cylinders and pho-
nograph records, whose length limitations (as we discussed in Chapter 2)
constituted a major impediment to the artistic ambitions of artists of that
time. The streaming payment methodology provides the incentive for these
abbreviated songs: more short songs means more plays squeezed into any
listening session with less likelihood of fans abandoning tracks before they
complete.
Ironically, this shorter song approach has been complemented by albums
that are getting longer in terms of the total number of tracks. In Feb 2018, the
hip-hop band Migos had the no. 1 album in America. Culture II contained
24 tracks and lasted one hour and 45 minutes, which was almost double the
length of their previous Grammy-nominated release. Many fans listen to the
entire album from their favorite acts, at least the first time through, and the
more songs on that album, the greater the income that is generated. New York
magazine music critic Craig Jenkins said that Migos was “trying to game the
Streaming 281
250
240
Duration (seconds)
230
220
system.”lxxi And Migos was not alone. Chris Brown’s Heartbreak on a Full
Moon in 2017 had 45 songs. Drake followed his 20-track 2016 album Views
with More Life in 2017, which contains 22 tracks that Drake himself referred
to as a “playlist.”
In 2022, the British indie band Pocket of God took these trends to a new
extreme when they released their album 1000 x 30: Nobody Makes Money
Anymore. The title said it all: the band was protesting inadequate compen-
sation by offering an album comprising one thousand tracks of just over 30
seconds in length. The first song is titled “0.002,” referencing the rate paid
per song that results from the royalty calculations. The tracks on the album
have been streamed over 600,000 times and, as a consequence, Spotify has
modified its rules to allow short songs to be pitched for inclusion on their
playlists.lxxii
And if you think of an album as just a different incarnation of a playlist,
it is not surprising that the order of songs on the album will be designed to
achieve more plays as well. Instead of sequencing the tracks to tell a partic-
ular story or to move through different moods and tempos, the likely hit
single is getting placed first. Ezekiel Lewis, Executive VP of A&R at Epic
Records, said, “people are definitely putting the focus tracks towards the
282 Key Changes
front of the album to make sure those get the proper shot.”lxxiii Some major
hip-hop artists, including Kendrick Lamar and Drake, are taking things a
step further and not even designating a “lead single” for their album releases.
All the tracks are available at once on streaming, and the fans’ listening be-
havior establishes the priorities.lxxiv
Combining these techniques with engaging music can result in superior
streaming performance for tracks and albums. Taylor Swift’s 2022 album
Midnights occupied every single one of the top 10 slots of the Hot 100 chart
shortly after its release.lxxv This record-breaking feat is certainly attributable
to Taylor’s musical talents and ability to engage with her passionate fanbase;
but in addition, some believe that the construction of her songs is particu-
larly well-suited to attract listeners on streaming.lxxvi For some albums, the
streaming results have been so good that they created a backlash in terms of
official chart measurements. Ed Sheeran’s March 2017 album ÷ (pronounced
“Divide”) had 16 tracks in its UK release; fans listened so much to the en-
tire album that the Official Chart Company reported that all those songs
had made it into the Top 20. The industry was concerned that this outcome
blocked a wider variety of music and artists from reaping the benefits of
occupying the top chart positions. By June of that same year, the “Ed Sheeran
effect” caused the UK industry to change the rules for positions on its charts.
To prevent the most popular artists from dominating the singles chart with
all the tracks from a single album, it was decided that an individual artist, re-
gardless of the actual streaming numbers, could only have up to three of their
most popular tracks in the top 100 at any point in time.lxxvii
Musicians who choose to create longer albums consisting of many shorter
tracks, whether motivated by artistic or commercial considerations, have
impacts beyond that individual project. The algorithms and curators respon-
sible for improving playlists will notice the shorter tracks that are success-
fully crafted to be catchier from the virtual needle drop. The algorithms and
people responsible will then move these tracks onto more popular playlists
and place them in more lucrative higher-up positions. Given the dominance
of playlist listening, this helps to generate even greater pro rata shares for
these songs. The financial benefits will reinforce the behavior of artists and
labels, and the cycle will continue.
Artists want to have songs that get maximum industry notoriety, whether
measured by sales published in Billboard or by holding down a slot on
Spotify’s Today’s Top Hits. The publicity is always a plus, but there is a lot
of money at stake too. Complaints about inadequate compensation from
Streaming 283
were going to find other ways to skirt the rules besides infringing copyrights.
If more listens and better playlist positioning mean a greater slice of the rev-
enue pie, then that’s where you will find the next crop of disreputable serv-
ices. Just as Willie Sutton robbed banks because “that’s where the money is,”
then creating fake streaming activity (that looks real) is where the money is.
In advance of Spotify’s direct stock listing in April 2018, the company felt
it was necessary to highlight this threat as one of the risks in its F-1 filing.
It stated, “We are at risk of artificial manipulation of stream counts and
failure to effectively manage and remediate such fraudulent streams could
have an adverse impact on our business.” There are several techniques that
can be used to create popularity or to generate cash for music that has not
engendered meaningful fan interest on its own. Spotify even offered some
useful tips to potential scammers in its S-1:
industry over whether streaming services have obtained all the appropriate
licenses has been an even larger issue.
It all goes back to those mechanical licenses created in the Copyright Act
of 1909 to pay royalties to songwriters for piano rolls. The compulsory me-
chanical license allows a streaming service (in this case) to play any musical
composition without asking for permission in advance, but until recently
the service had to notify the owner(s) of that composition and pay them
the royalties specified in the statute. If it couldn’t locate the songwriters and
music publishers who hold the composition rights, then it was allowed to
file a form called a Notice of Intent (NOI) with the Copyright Office instead.
The service had to do this individually for every track it ingested from record
labels (or other sources, such as indie distributors).
But there was a problem with that: the service needed to know the
identities of the compositions and the rights holders. Record labels don’t nec-
essarily include that information when they send their music to streaming
services, so the services were on the hook to figure that out for themselves.
This turned out to be a major headache for the streaming services.
When the major labels started pressplay and MusicNet, they worked
with the Harry Fox Agency (HFA), then a subsidiary of the National Music
Publishers Association (NMPA), to obtain mechanical licenses for the tracks
they were offering. This arrangement resulted from the deal that the RIAA
had made with the NMPA that we mentioned in Chapter 8. HFA had been
the go-to source for mechanical rights clearance for decades, through the
eras of physical recorded media; by the time interactive streaming services
were ready to launch, HFA had amassed a database of information about the
compositions being performed on millions of recorded music tracks and the
songwriters and music publishers that owned the rights to them. And when
Rhapsody launched with major-label licensing in 2002, it made its own ar-
rangement with HFA to match recordings to musical compositions and clear
the mechanical rights.
Rhapsody set a pattern that Spotify and most of the other Celestial
Jukebox–type streaming services would follow in later years: instead of
relying on labels themselves to clear mechanical rights for the music that
labels sent to them, streaming services would engage HFA to match the
recordings to compositions and clear mechanical rights. Another company
called Music Reports Inc. (MRI), which traditionally cleared music rights
for television, stepped in to compete with HFA in what was clearly going
to be a highly lucrative market for these services. So did a startup company
Streaming 289
called RightsFlow, which Google ended up acquiring for exclusive use with
its own music services, such as YouTube. By the mid-2010s, these rights-
clearance services were processing tens of thousands of tracks per day for
each of their music service customers—the vast majority of which were re-
dundant (the same sets of tracks for each service). Yet none of these serv-
ices had databases of rights owner information that were entirely complete,
accurate, and up to date.
This situation continued without much complaint through the early 2010s
while interactive streaming was only a small part of the market. But once
streaming began to grow in importance, the rights ownership data problems
rose in importance too.
It’s important to bear in mind that matching recordings to compositions
is often not straightforward. A classic singer-songwriter track like James
Taylor’s “Fire and Rain” is simple: James Taylor is both the recording artist
and songwriter. You know who the songwriters are when you hear the Stones’
“Jumpin’ Jack Flash.” But most of today’s hip-hop, pop, and country songs
have many different songwriters working with multiple music publishers;
and occasionally a song will be released before they all agree on the “splits”—
the ownership share percentages. There are old or obscure recordings where
the songwriter credits are lost in the mists of time. And there are many other
cases where composition rights holder information is missing or inaccurate
due to various sorts of data errors that can creep in along the way.
Such errors and omissions are inevitable when that much data gets passed
down the line from through multiple links in the music supply chain. For
physical music formats, music publishers worked with record labels to figure
out mechanicals.12 But streaming services have had no direct relationships
with music publishers, so they have had to rely on intermediaries such as
HFA to figure out the ownership data after the fact.
So as streaming began to become the dominant source of revenue in
music in the mid-2010s, more songwriters and music publishers began to
notice that their mechanicals weren’t being paid properly or at all, and the
royalties involved began to matter enough to motivate them to do something
about it. The situation came to a head in the mid-2010s, when songwriters
and music publishers—some individually, some facilitated by the NMPA—
filed lawsuits against Spotify, Apple Music, and other streaming services for
unpaid mechanicals. Spotify settled the NMPA-facilitated lawsuit against
it in 2016 for $30 million;xcvii the settlement covered the many publishers
that chose to opt into its terms. But several publishers did not opt in and
290 Key Changes
instead filed their own lawsuits. Within the next couple of years, representa-
tives of dozens of songwriters of compositions recorded by Neil Young, Tom
Petty, the Doors, Stevie Nicks, the Four Seasons, Thomas Dolby, Camper
Van Beethoven, and many others brought their own lawsuits. The aggregate
claimed damages went well up into the billions of dollars.xcviii
The industry had, of course, known about this problem since the days of
MusicNet and pressplay in the early 2000s. The trade associations involved—
the RIAA for labels, NMPA for publishers, and DiMA (Digital Media
Association) for streaming services—made various attempts to solve it by
updating the copyright laws. The fact that “phonorecords” in the streaming
era are simply collections of bits transmitted via various internet and wireless
protocols from cloud-based servers to smartphone applications makes the
case that an update of the law was probably necessary.
One such attempt at legislative reform was the Section 115 Reform Act
(known as SIRA) of 2006; Section 115 of the Copyright Act defines the com-
pulsory licenses for making and distributing phonorecords. SIRA would
have set up an infrastructure for mechanical licensing modeled on the one
for performance licensing through the performance rights organizations
(PROs): each streaming service could take a blanket license for mechanicals
instead of having to license each composition individually, and an entity
known as a “designated agent” would handle collective licensing and ensure
that payments would be made to the proper rights holders. But the various
factions involved couldn’t come to agreement on various details, and the leg-
islation failed; but the sense of urgency wasn’t there yet in 2006.
That sense of urgency arrived with the rash of lawsuits filed—and the as-
sociated money at stake—in the mid-2010s, and the industry began casting
about anew for solutions. The NMPA and DiMA ultimately arrived at a so-
lution, achieved buy-in from the various industry factions, and presented it
to Congress as the Music Modernization Act (MMA). Perhaps even more
remarkable than the broad support across the industry for this bill was that
both houses of Congress passed it unanimously in such a politically polarized
era. The MMA became law in 2018.xcix
Like SIRA, the MMA created a blanket license for mechanicals that
replaced the onerous track- by-
track licensing requirement under the
old scheme. This was a huge savings in administrative overhead for the
streaming services. The MMA also implemented collective licensing for
streaming mechanicals, but in a different way than SIRA contemplated. It
created the Mechanical Licensing Collective (MLC), a nonprofit entity that
Streaming 291
Conclusion
The shift from paying for products to paying for access has not been simple.
This fundamental change required new strategies and approaches to be
successful, but streaming represents a bridge to sustained growth after
two decades of troubled water for the music industry. Meanwhile, the
technologies that enabled this shift can also transport video along with
sounds; and newer technologies enabled ordinary users to create not just
copies of recorded music but music embedded into their own videos. In one
sense, the resulting combination of user-created videos with pre-existing re-
corded music was an incremental extension of streaming music; but in other
senses, it was the next major source of disruption for the music industry.
10
Streaming Video
Throw Away Your Television—Red Hot Chili Peppers
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0010
294 Key Changes
even (in many cases) wanted: videos contributed by users. As we’ll see, user-
contributed music videos solved two problems at once: they were generated
in extremely high volume at no cost to labels or artists; and they had a low-
cost, DIY feel that—at least at first—suppressed fans’ expectations of high
production values.
The third obstacle was copyright law regarding user-contributed con-
tent. Napster may have shut down in 2001, but it led to many efforts to
create content-sharing services that tweaked the Napster model in hopes of
withstanding legal challenges. As we saw earlier, some—like LimeWire and
Grooveshark—didn’t succeed. But others did, and by the time entrepreneurs
figured out how to thread the legal needle, the aforementioned technological
limitations had been overcome, and so the focus of content sharing shifted
from text and audio to video.
The resulting environment around 2005–2006, therefore, was one in
which an Internet content-sharing service featuring user-contributed videos
could flourish and grow to become a leading distribution channel for music.
And that’s exactly what happened. The race for domination was on, and there
was one resounding winner: YouTube.
YouTube is approaching 20 years old today, but it is still the biggest re-
corded music platform in the world, video or otherwise. Yet YouTube didn’t
particularly invent new technologies. Streaming video had yet to reach the
mainstream in 2005, but the codecs and other fundamental technologies had
been in place since the late 1990s. It also wasn’t the first video-sharing web-
site. Of the ones that launched earlier, those that are still around today are
Metacafe (launched July 2003), Vimeo (November 2004), and DailyMotion
(March 2005).
Steve Chen, Chad Hurley, and Jawed Karim, all employees of the pio-
neering online payments service PayPal, actually started YouTube in 2004
as an online dating service, one that enticed women to upload videos of
themselves. Finding too few women willing to make such videos (despite the
promise of $100 rewards), they pivoted to general videos around May 2005
when the site went into public beta. The site’s official launch was in December
2005, after the Silicon Valley VC firm Sequoia Capital made a $3.5 million
investment in the startup.
Streaming Video 295
Figure 10.1 A Flip Video digital video camera plugged into a laptop computer.
Alan Levine, CC 0 (public domain) via Wikimedia Commons.
made its site more desirable by striving to get as much content uploaded to its
site by as many people as possible as quickly as possible. It made videos easier
to distribute by enabling users to embed videos in web pages, email messages,
blog posts, or anything else that used HTML, by providing embeddable links.
And it worked with Apple to produce a simplified mobile app—the first of its
kind—for the iPhone at the device’s launch in 2007.
YouTube also showed videos at relatively low resolution compared to
other early video-hosting services, to increase ubiquity by trading off pic-
ture quality. Lower resolution meant faster uploads as well as smoother
streaming at lower Internet access speeds. YouTube chose the Flash video
format from Adobe, which was desirable at the time for its adaptability across
many different devices.2 It also chose relatively low 320 × 240 pixel resolu-
tion, a quarter of the size of the VGA standard of 640 × 480 that just about all
computers of that time supported. New monitors in the mid-2000s typically
displayed up to 1280 × 1024 resolution, 16 times that of YouTube videos.
(YouTube increased its picture quality continuously over time; today it
supports up to 7680 × 4320, known as 8K resolution, the same as in cameras
widely used to shoot network television.)
The third success factor was external to YouTube: the steep growth of
broadband Internet access at that time, giving users enough speed to upload
digital video in reasonable timeframes as well as stream video with decent
Streaming Video 297
quality. In February 2004, less than a quarter of US adults had broadband ac-
cess through cable TV or phone lines (DSL). But by March 2005, this figure
had risen to 30 percent, then to 42 percent by March 2006.i Meanwhile mo-
bile broadband connectivity at speeds high enough to support streaming
video didn’t come until the early 2010s with the major mobile operators’
buildouts of their 4G networks.3 Without broadband access, residential
Internet users had to contend with 56 kbps (or slower) dialup lines, which
meant a few minutes of digital video took several hours to upload. Typical
consumer broadband connections of the day reduced that time to a few
minutes per minute of video uploaded.
All of these factors put YouTube on a jaw-droppingly rapid growth
path, the likes of which the industry had never seen previously. Google ac-
quired the company in November 2006; by that time—less than a year after
its launch—YouTube had an estimated 50 million users worldwide and
accounted for almost half of all video traffic on the Internet. (The price tag of
$1.65 billion also caused jaws to drop, but no one has ever accused Google of
overpaying after the fact.)
Of course, YouTube was not designed to be a music service; and as we’ll
discuss later in this chapter, it has had various phases of growing pains on its
way to becoming the largest licensed music service in the world. Yet music
was integral to YouTube from the beginning. Probably the most important
video to appear on YouTube in its earliest days was “Lazy Sunday,” a hip-hop
parody music video that originally aired on NBC’s Saturday Night Live (SNL)
in December 2005. Several users posted unauthorized copies of the video to
YouTube. It garnered over seven million views—despite the fact that the site
had only officially launched two days before the clip aired on SNL.
By the end of 2006, the top 10 most-watched videos on YouTube included
three music videos: two young comedians lip-syncing to the Pokemon theme
song; a teenage Korean guitarist shredding his way through a rock version of
Baroque composer Johann Pachelbel’s Canon in D Major; and a DIY-style
video of “Here It Goes Again” by the pop-punk band OK Go. The slickly
produced music videos would come in time; but otherwise, these three low-
budget clips encapsulated YouTube’s appeal as a medium for music. Ironically,
seven million views weren’t quite enough to put “Lazy Sunday” in the top ten
most-watched videos of 2006, though it came close. (The Pokemon theme
song video garnered 18 million views.)
However, at the same time, many users started uploading videos
containing commercially released music. We’ll see shortly how YouTube
298 Key Changes
addressed the inevitable copyright challenges, but for now, suffice it to say
that the solution depended on another important technology: acoustic
fingerprinting.
We first saw acoustic fingerprinting in Chapter 8, when the technology
was used in attempts to block unlicensed content from being uploaded to
peer-to-peer (p2p) networks. Several technology vendors had introduced
acoustic fingerprinting solutions, and by the mid-2000s, Audible Magic had
emerged as a de facto standard for copyright filtering applications. By 2005,
the major labels began to accept its use in conjunction with licensed com-
mercial music services.
One example of this was a p2p network based in Israel called iMesh. Like
several other p2p networks in the wake of Napster, as we saw in Chapter 8,
iMesh was sued by the major labels and lost. But unlike the others, iMesh
worked out a deal with the labels to remain in operation while it transformed
itself into a licensed service. The service that iMesh eventually launched
was a file-sharing network that blocked uploads of copyrighted material
identified by Audible Magic and allowed the uploads if Audible Magic didn’t
find a match. If a user searched for a track that wasn’t in the fingerprint da-
tabase, they could download it. But if a user wanted a track that was in the
fingerprint database and paid a monthly subscription fee, they would get a
DRM-protected file that only played on their PC. iMesh announced its first
major-label license—with Sony BMG—in July 2005,ii though the new service
didn’t launch until September 2006.iii
YouTube emerged in the same timeframe. Bear in mind that in 2006,
revenues from CD sales were plummeting from their peak in 2000, but CDs
were still the dominant source of revenue by an order of magnitude over
downloads. And interactive music streaming, available at that time through
services such as Rhapsody and MusicNet, was a much smaller niche than
downloads. YouTube represented another form of interactive streaming. So
while YouTube was gaining enormous amounts of user traffic every month,
it represented a music usage model that hadn’t really caught on with the
public yet.
This—along with a sense that file-sharing may have been more of a missed
opportunity than the labels admitted publicly—may have led the labels to
want to negotiate with YouTube, or at least to wait and see how YouTube
could develop as a music outlet, instead of merely trying to sue it out of exist-
ence. At the same time, the latter strategy had its own challenges, as we’ll see
shortly.
Streaming Video 299
Warner Music Group was the first to make a deal with YouTube; it did so
in September 2006, a mere nine months after launch. In the deal, YouTube
agreed to implement Audible Magic’s acoustic fingerprinting system and to
populate its fingerprint database with Warner Music’s catalog.4 Whenever a
user tried to upload a video containing music from that catalog, YouTube
would allow the video to go up on the site and would share a portion of the
revenue from ads shown alongside the video with the label. (Other video-
sharing sites such as DailyMotion still use fingerprinting simply to block
uploads that contain music in their fingerprint databases.)
As for the other majors, Sony BMG Music and Universal Music reached
deals with YouTube shortly after Warner Music did, but they were limited
to catalogs of videos that the labels owned and did not license YouTube for
user uploads of the labels’ recordings.iv EMI cut a similar deal—known as a
“whitelist” deal—in May 2007.v Warner, Sony BMG, and Universal all took
equity stakes in YouTube around October 2006, even as some of them were
filing lawsuits against other video-sharing services.vi The equity investments
lent legitimacy to YouTube that helped it secure its $1.65 billion price when
Google acquired it two months later.
As we’ll see, Warner Music’s deal became the archetype not only for other
labels’ deals with YouTube but also for countless other content owners’ deals
with user-contributed content services to come.
Meanwhile, music continued to be a major draw on YouTube. A 2008
study found that almost 20 percent of the 78 million videos—more than
15 million—on YouTube by then were categorized as Music (one of 13
categories at that time).vii To put that figure in perspective, the iTunes Music
Store in 2008 had 6 million songs in its catalog.5
YouTube’s domination of online music video held steady for a decade,
even as the proportion of videos on the site in the music category declined.
It kept original competitors such as Vimeo and DailyMotion trailing far be-
hind and fended off numerous would-be startup competitors. Unlike those
independent companies, YouTube benefited from several aspects of Google’s
technology infrastructure, such as its AdSense contextual ad technology,
content delivery network, and immense cloud storage capabilities.
The first serious challenge to YouTube’s music video supremacy6 came in
2016 with the launch of A.me, a video sharing app from a Chinese startup
called ByteDance. The app changed its name to Douyin and became incred-
ibly popular incredibly quickly in China, amassing 100 million users in one
year. In September 2017, it launched globally as TikTok.
300 Key Changes
live streaming of musical performances was a small market and split into
two distinct segments: big, expensive, professional productions featuring
major stars; and small DIY facilities for everyone else. YouTube started
experimenting with live streaming with a concert by U2 in 2009 using a
third-party technical infrastructure. It began limited rollout of its own live
streaming infrastructure in 2013 and eventually made live streaming avail-
able through its mobile apps.
A few small outfits provided live streaming functionality geared toward
musical performances in the same timeframe. A prominent example was
Livestream (formerly Mogulus), which hosted Internet-only concerts from
the likes of the Foo Fighters starting in 2009. The company was acquired
by Vimeo and renamed Vimeo Livestream in 2017. Facebook added live
streaming to its namesake platform as well as Instagram in 2015 and 2016,
respectively.
The advent of the COVID-19 pandemic in March 2020 led quickly
to a parade of startup launches focusing on music- oriented live
streaming: Mandolin, Veeps, Sessions, Bulldog DM, Dreamstage, and others.
Online gaming platforms such as Twitch and Fortnite have also gotten into
the act. Paid live streams are now an accepted part of the music industry.
Although we aren’t focusing on it any further in this book, live streaming is
now a significant force in the industry and will continue to play a role now
that the pandemic has faded into the rearview mirror.
YouTube met with roughly the same legal challenges that Napster faced,
but it did so at a much later time. The major labels didn’t sue YouTube as it be-
came a huge draw for unlicensed content. But Hollywood did.
As we saw in Chapter 8, the Digital Millennium Copyright Act (DMCA)
of 1998 established safe harbors for online services that limit their liability
from their users’ copyright infringements if they play by certain rules. One of
these rules is known as “notice and takedown”: online services must process
requests (notices) sent by copyright owners to have them remove their con-
tent if it is there without permission. Another rule is that service providers
must terminate the accounts of users who are repeat infringers.
By the time YouTube got started, processes for handling takedown notices
had become established among online service providers, as were “three
302 Key Changes
strikes, you’re out” policies for terminating repeat infringers.8 YouTube has
a mechanism for processing takedown notices and a “copyright strike” re-
peat infringer termination policy. However, other aspects of the DMCA were
less clear-cut regarding online service providers’ responsibilities. The statute
suggested that online services would not be required to affirmatively monitor
their services for copyright infringements to qualify for DMCA safe harbors
but that they did have some culpability through knowledge of infringements
taking place on their services, known in legalese as “red flag knowledge.” But
as online services evolved, the boundaries among red flag knowledge, its le-
galese cousin “willful blindness,” and affirmative monitoring became sources
of legal uncertainty.
As we’ve seen, YouTube uses fingerprinting technology to help it deter-
mine when a user uploads a work that a copyright owner has claimed as its
material. YouTube’s legal position was (and remains) that it isn’t legally re-
quired to use fingerprinting, because that would be tantamount to affirmative
monitoring for copyright compliance. Some content owners were happy with
neither this legal positioning nor YouTube’s technological solutions after the
Warner Music deal in 2006. Users found ways to circumvent Audible Magic’s
fingerprinting technology when they uploaded content to YouTube, such as
by inserting a period of silence before the start of a song or shifting the song’s
pitch a little bit. And Audible Magic didn’t work at all on video content (un-
less it happened to contain music that was in Audible Magic’s audio finger-
print database).
Hollywood studios and TV networks in particular were not amused, even
though at that time YouTube was limiting the length of uploaded videos to
ten minutes. In 2007, the giant entertainment company Viacom—parent of
MTV, as well as Paramount Pictures and Cartoon Network—sued YouTube,
claiming a billion dollars in damages. Viacom’s intent was to get a court to
firm up online service providers’ responsibility to police infringements on
their networks.
But Viacom’s gambit failed. The Second Circuit appeals court affirmed
that YouTube had no obligation to monitor its network for copyright issues
and sent the case back to the lower court. The case settled for an undisclosed
amount and didn’t go to trial. Around the same time, Universal Music Group
(UMG) sued Veoh, another video-sharing site that launched about a year
after YouTube. The results were similar; in addition, the judge in that case
stated that the plaintiff had no right to demand—as UMG had tried to—that
the online service use a particular fingerprinting technology.
Streaming Video 303
While these lawsuits were being filed, Google replaced Audible Magic with
its own content recognition technology, which it called Content ID. Content
ID recognizes video as well as music; it is also more tightly integrated with
Google’s advertising technology. When a user uploads a video to YouTube,
the Content ID system takes its fingerprint and looks it up in a fingerprint da-
tabase, just as Audible Magic or any other fingerprinting technology would.
If it finds a match, it gives copyright holders a number of options for actions
that YouTube can take. Originally there were three choices: in addition to
the ad revenue share option a la the Warner Music deal, the options were
to block the upload or to track the usage statistics of the video for the copy-
right holder. Yet the ad revenue share option has become the overwhelming
choice of record labels and other copyright owners. YouTube subsequently
added a fourth option: to allow the upload but not allow users who qualify
for monetizing their videos to share in the ad revenue.9
In other words, the Content ID revenue share scheme had become—as
Google may have foreseen—a de facto licensing arrangement for music on
YouTube before some of the major labels made actual licensing deals with the
service. As we’ll see, this arrangement has changed the fundamental financial
dynamics of the music industry.
Yet although the labels mostly took the ad revenue share option, they
weren’t happy with certain loopholes in the Content ID system. Labels
wanted more control over which user-uploaded videos could go up and
which couldn’t; for example, they didn’t want user- contributed videos
that cast artists in a disparaging light. It is possible to use the notice-and-
takedown mechanism to have a video removed from YouTube, but as we’ll
discuss later in this section, users can repost videos immediately after-
ward. Users also have the option to issue counternotices and have the videos
replaced. Users can easily claim that their videos are fair use,10 and YouTube
will leave the video up for two weeks before taking further action. Labels
consider this to be unreasonable.
Like Audible Magic, Content ID originally recognized copyrighted sound
recordings in user-created videos that contained them. But that process
doesn’t work when users upload their own cover versions of someone else’s
musical compositions. In that case, royalties are owed for the compositions;
YouTube has to identify the compositions and figure out whom to pay
royalties.
Automatic detection of cover versions of compositions—melodies, lyrics,
and so on—is a much harder technical problem than automatic detection
304 Key Changes
of sound recordings. And the solutions that exist are more recent. While
acoustic fingerprinting technologies for sound recordings were first invented
in the late 1990s and came to market in the early 2000s, cover detection
techniques weren’t even in the R&D stage until the mid-2000s. Starting in
2006, researchers have competed on song-cover detection in an annual con-
ference called Music Information Retrieval Evaluation eXchange (MIREX).
In the 2020 competition, the winners—a team from Peking University and
Tencent in China—only got it right 84 percent of the time when asked to
find cover versions of a given song in the test data.xiii Acoustic fingerprinting
technologies generally achieve much better accuracy on sound recordings.
Since at least 2012, Google has been using Content ID to detect covers on
YouTube.xiv Content ID’s accuracy on cover detection is uneven but has likely
improved over time.11 Music publishers usually choose the Content ID op-
tion to allow uploads of cover videos but not allow user monetization. (This
is appropriate for musicians who cover well-known songs to get exposure or
share with their friends and family.)
YouTube may not be the only content-sharing service that attempts to de-
tect covers of musical compositions, though it’s unclear. SoundCloud, for ex-
ample, does not publicize the ability to detect covers.
YouTube’s origins as a user-generated content service make it a sort of
catchall for music. The Content ID system gives YouTube an architecture for
copyright compliance that could be described as “opt-out”: copyright owners
have to submit their content to Content ID for it to be flagged when someone
tries to upload it. If someone uploads content that isn’t in the Content ID
database, then YouTube makes it available without restrictions (and doesn’t
share ad revenue with any labels). In contrast, services like Spotify and
Apple Music are “opt-in”: these services won’t offer content unless the cop-
yright owner (or appointed representative, such as a digital distributor like
TuneCore or Symphonic) provides it. That’s why YouTube is often a “music
service of last resort” for fans of old or obscure music: there’s no artist or label
to either provide it to “opt-in” services or flag it on YouTube.12 Conversely,
there are virtually no artists anymore who distribute their music online but
keep it off YouTube—despite the royalty disparity that we’ll discuss shortly.
Another important implication of YouTube’s “opt-out” licensing model is
that it favors globalization. Licensing is inherently national in scope; artists
are signed to contracts under national laws with labels that distribute music
on a country-by-country basis. But a video uploaded to YouTube has global
reach by default, and not one country’s but many countries’ licensing entities
Streaming Video 305
must choose whether or not to opt out. Later in this chapter we’ll see how this
has led to a class of global megastars that wouldn’t be as possible—or as pos-
sible as quickly—without YouTube’s opt-out effect.
TikTok also started out as an opt-out service, but it has changed to more of
an opt-in model. At first, TikTok enabled users to add whatever music they
wanted to their videos. But once record labels started expressing copyright
concerns, it made users choose from a library of music tracks, although it
has been possible for users to add any music they want anyway through var-
ious hacks.
The obvious resolution to this situation was for record labels to do license
deals with ByteDance so that TikTok users could legally select the labels’ re-
corded music as backgrounds to their videos directly within the app. Sure
enough, between November 2020 and February 2021 the three majors each
signed licensing deals with TikTok. TikTok has also launched its own licensed
subscription streaming music service, called Resso, in India, Indonesia, and
Brazil; it hopes to expand the service to other countries but has been getting
pushback from the major labels.xv
As we write this, the opt-out model is under increasing legal pressure, es-
pecially in Europe. The European Union (EU) recently enacted a law, Article
17 of the 2019 EU Copyright Directive,xvi that imposes responsibilities on
services like YouTube to take licenses to copyrighted material before making
it available to users. Just how much responsibility—and what sorts of tech-
nological schemes will satisfy rights holders under the law—remains to be
seen, especially since the law must be implemented in each of 27 EU member
states according to their national copyright laws. But most commentators be-
lieve that it will result in services like YouTube being required to take more
aggressive steps to block copyrighted content from being uploaded by third
parties without permission.xvii
And in the United States, Congress is considering reforming the DMCA,
in part according to recommendations from the Copyright Office after a
multiyear study, which concluded that the DMCA, over twenty years of
precedents set by litigations, has become tilted too far in favor of online
service providers.xviii
One of the main provisions that copyright holders have been asking for in
DMCA reform legislation is something they call “notice and staydown.” The
problem that copyright owners have with “notice and takedown” is that once
you send a DMCA notice to have content removed, the same content can be
reposted shortly afterward—a phenomenon that has come to be known as
306 Key Changes
publishers see synch as their last available point of control over licensing: they
can determine whether each song gets licensed at all, to what services, for
what video content, and at what rates.
Music publishers have long complained about having to play second
fiddle17 to record labels in digital licensing because they have been relegated
to distant relationships with service providers. They will undoubtedly rely on
synch licensing as a chance to improve their leverage, and this could affect
the label-versus-publisher dynamics of the industry. However, because most
of these newer video services are platforms for user-uploaded content, they
are likely to be in for years of fighting with publishers over questions of cop-
yright liability for service providers that are analogous to those that YouTube
and Veoh faced.
After Warner Music made its deal with YouTube in September 2006, sev-
eral years elapsed before YouTube garnered full-catalog licenses from all of
the major labels. As we mentioned, the other majors made “whitelist” deals
with YouTube later during the ensuing year that only involved their own
music videos and did not cover user-uploaded content. Yet once YouTube
had implemented Content ID, most record labels began registering their
content in the system and received ad revenue shares. In other words, the
labels may have had what lawyers call implied licenses even if they didn’t
have actual license agreements with YouTube.
YouTube eventually achieved formal license agreements with all of
the major labels (as well as thousands of indie labels), which covered user
uploads as well as terms other than royalty payments. The terms of these
agreements are confidential, and in most cases it’s not publicly known when
those agreements were signed. But it’s clear that by the time Spotify launched
in the United States in July 2011, YouTube also had licenses covering virtu-
ally full catalogs from all the major labels.18
As YouTube continued to grow, some of the major labels got together in
a joint venture to create a service called Vevo, which they called “Hulu for
music videos” (referring to the video streaming service Hulu and its joint
ownership by multiple TV networks). In reaction to their experience with
MTV, the labels wanted to build a video service featuring their own con-
tent that they could control. The project started at UMG; UMG brought
308 Key Changes
Even so, YouTube’s fit in the music ecosystem has been uncomfortable. As
we’ll explain later in this chapter, YouTube has had trouble building a com-
pelling user experience for music fans. And many label executives have never
been thrilled with the idea of an all-free, all-ad-supported music service,
even leaving aside the question of royalty payments (which we’ll also discuss
later).
In addition, the YouTube user experience does not inherently funnel users
toward big releases as that of Spotify and its ilk. For example, YouTube has
no equivalent of Spotify’s editorial playlists such as Rap Caviar; nor does
Spotify’s concept of influencer users with massively popular playlists have
a real equivalent on YouTube. Far more people use YouTube for music lis-
tening than they use Spotify or Pandora, but the play counts of many hit
songs on YouTube are proportionally lower than those on the other services.
For example, Billie Eilish hits such as “bad guy” and “when the party’s over”
have gotten about two-thirds the number of plays on YouTube as they’ve had
on Spotify; Drake’s, Ariana Grande’s, and Taylor Swift’s numbers are similar.
YouTube has also had a challenging relationship with other outposts of
the Google empire (though this is not at all unusual for large technology
companies that have grown by acquisition). This has particularly been true
with respect to the Android division. Google introduced the Android oper-
ating system for mobile devices in November 2007, a year after it acquired
YouTube and just a few months after Apple launched the first iPhone. The
first Android devices came out a year later. The idea of Android was essen-
tially to do for smartphones and tablets what Microsoft did for PCs with
Windows: create an operating system that would run on devices from a wide
variety of manufacturers, as opposed to Apple’s combined proprietary device
and operating system offerings.
To compete with the Apple iPhone (and eventually the iPad), Google
launched a set of apps and services for Android devices that were analogous
to those available on Apple devices, such as iTunes, which by 2008 was of-
fering downloaded movies and TV shows as well as music, and iBooks (now
Apple Books), an e-book service launched with the iPad in 2010. Google
launched a series of analogous apps under the banner of Google Play, its
Android app store: Google Play Music, Google Play Movies and TV, and
Google Play Books.
Google launched Google Play Music in 2011 as an iTunes-style paid
download plus Internet radio service. In 2013 it launched Google Play Music
All Access, an interactive streaming service to compete with Spotify. (It
310 Key Changes
would be another two years until Apple launched Apple Music, its interactive
streaming service.)
Google Play Music All Access was a much better-organized music lis-
tening experience than YouTube, but it didn’t make much impact on the
market. Google treated Google Play Music as more of a “checklist item” for its
Android ecosystem than as a marquee offering, as Apple did with iTunes or
Spotify did with its service; it devoted far fewer resources to design and mar-
keting. Its listenership in 2015 was less than half that of Spotify and a smaller
percentage of that of Pandora and iHeartRadio.xxv
YouTube and Google Play Music operated in separate silos of a large and
fast-growing company. Despite YouTube’s drawbacks as a music experience—
and undoubtedly because it’s free—YouTube’s popularity as a music service
always vastly eclipsed that of Google Play Music. At the same time, the major
labels had been putting pressure on Google to take steps to steer users toward
the paid subscription service, just as they have done with Spotify.
Google began a series of attempts at integrating the two services in
2014. This resulted in a series of confusing combinations of branding and
technical features over a period of six years. The first try was YouTube
Music Key, which launched in late 2014. For a $10 per month subscrip-
tion, YouTube users got ad-free music videos and the ability to play the
app in the background and save videos for offline playing. Google Play
Music still operated independently, but its subscribers eventually got ac-
cess to YouTube Music Key as well. Then in 2015, YouTube replaced Music
Key with YouTube Red, which provided ad-free access to all videos, not
just music, for the same $10 per month price. None of this was very suc-
cessful: not only was the branding confusing but some labels refused to
license music into the service.
Google tried again in 2018. By that point, interactive streaming had be-
come the largest part of the music industry by revenue, and Amazon had
emerged as a fast-rising competitor with Amazon Music. Google launched
YouTube Music as a successor to Music Key. (It also launched YouTube
Premium as a successor to YouTube Red at a higher subscription price.
More branding confusion.) YouTube Music started out essentially as a paid-
subscription, ad-free version of YouTube that attempted to impose some
music-oriented order on the vast YouTube chaos, as we’ll discuss shortly.
Google took the final step in late 2020—nine years after the launch of Google
Play Music—by transitioning its subscribers to YouTube Music and phasing
Google Play Music out entirely.
Streaming Video 311
licensing professionals and engage with record labels and music publishers.
Yet it wasn’t obvious how Facebook—by that time a sprawling behemoth—
should enter the world of licensed music services.
Video was where the opportunities for growth and differentiation were by
the mid-2010s, so most of Facebook’s music-licensing activity has been re-
lated to video content. In recent years, it has added a number of music video
features, most of which are reactive to features already on YouTube or TikTok.
Facebook also launched its own content identification technology, Facebook
Rights Manager, in 2016 to supplement its Audible Magic implementation.
Facebook has launched a series of video features since then. It launched
a streaming video service called Facebook Watch in August 2017. Then
starting later that year with Universal Music Group, Facebook signed a series
of deals with major and indie labels and music publishers to license the use
of music in “social experiences” (user-uploaded content including video) on
all of its platforms including Instagram and Oculus in addition to Facebook
itself. In June 2018, it introduced Lip Synch Live, an obvious competitor to
TikTok and Musical.ly. In August 2020 it released a feature that automatically
adds music videos to artists’ pages (unless they opt out)—in other words, a
feature for turning Facebook’s vast agglomeration of artists’ pages into com-
petition for YouTube. It also added a dedicated music section on Facebook
Watch. And in September 2021, it launched Reels on Facebook, an offshoot
of the similarly named service on Instagram.
their own videos and upload them quickly. And that weighs against the orig-
inal advantage of YouTube over MTV that we discussed at the beginning of
this chapter: that labels can achieve scale economies in video production by
offloading it to users.
For example, the biggest category of user-uploaded music videos in the
early days of YouTube was lyric videos. These typically consisted of still
images or plain backgrounds with the text of song lyrics on them. These
were easy to produce and provided a service to users, so they became quite
popular, especially when—as was often the case—they were among the first
videos of songs to be posted.20 A cottage industry of fast (and presumably
automated) lyric video producers emerged. The labels eventually responded
with official lyric videos, which tend to be more lavishly produced than user-
contributed lyric videos, especially for big-name artists; yet these are still the
exception rather than the rule.
The major labels’ frustrations about the disparity in royalty payouts be-
tween YouTube and other services actually dates back much further than the
IFPI’s “value gap” argument: it dates back to 2008, when Warner Music Group
was negotiating a renewal of its original license agreement with YouTube.
The parties failed to reach terms, so in December 2008, WMG pulled its cat-
alog from YouTube. Although the main digital music royalty benchmark
at that time was iTunes downloads, WMG found that its ad revenue shares
from YouTube even fell short of other ad-based online music services of the
time, such as AOL Music and MySpace Music.xxviii The other major labels
did not join in the boycott, however, and WMG negotiated a new deal with
YouTube in September 2009.
Although music services’ agreements with labels are confidential, various
exposés over the years, based on artists voluntarily disclosing their royalty
payments, have shown much lower royalty rates from YouTube than from
opt-in streaming services. For example, a story published in the Guardian in
Britain in 2015 indicated that an artist would have to get almost four times
the number of plays on YouTube as on Spotify or Deezer to earn the same
royalties, and almost ten times the number of plays on Beats Music (the pre-
cursor to Apple Music).xxix IFPI’s value gap analysis in 2017 estimated that
the worldwide music industry earns revenue per user from YouTube that is
less than a twentieth of revenue per user from Spotify.21
In addition to the royalty rates themselves, YouTube’s ad-based royalty
structure makes it more difficult for labels to negotiate the kinds of rate
Streaming Video 315
structures they would like. The labels have evolved Byzantine rate structures
with the opt-in services that depend on locations of users, minimum per-
user fees, fan engagement metrics, the size of the label, stature of the artist,
time of year, and other factors. If the revenue to be shared with a label starts
with ad revenue earned from what could be several videos containing the
same song submitted by users from anywhere in the world, in addition to of-
ficial videos, labels have less control over the rate structure. In a paradoxical
way, this is analogous to the labels’ frustration with iTunes during the down-
load era, when Steve Jobs insisted on fixing prices at 99 cents per track and
wouldn’t allow for variable pricing, as we saw in Chapter 8. Yet it’s important
to note that the foregoing only applies to “regular” YouTube and that royalty
payouts for paying YouTube Music subscribers are in the same ballpark as
Spotify.xxx
The music industry’s approach to closing the value gap has been fo-
cused on changing laws so that services like YouTube are obligated to keep
copyrighted material off their networks unless they have licenses for it, as
we discussed earlier in this chapter. That’s the intent behind the Article 17
copyright reform in Europe, which the media industry pushed for precisely
because of the value gap. (Note that Europe’s biggest native video-sharing
service, DailyMotion of France, has long observed this approach.) It’s also
one of the media industry’s main objectives in reforming the DMCA in the
United States.
The value gap discussion pertains to all user-contributed content
services, including audio- oriented services like SoundCloud. The
negotiations that led up to Article 17 led it to include several exemptions
for content-sharing services that weren’t in the media industry’s cross-
hairs, such as personal cloud storage services and open-s ource software
repositories. But the spotlight has always been on YouTube because of
its enormous audience; so YouTube should serve as the yardstick for
the music industry’s efforts to address the value gap for the foresee-
able future—a lthough TikTok is also starting to raise concerns over
royalty payments as it continues its meteoric growth.xxxi As is the case
with the labels’ deals with other audio streaming services, TikTok’s roy-
alty rates are confidential and vary according to several factors. The
majors are even more dissatisfied with their payouts than they are with
YouTube and have been pushing for higher royalties in contract renewal
negotiations.xxxii
316 Key Changes
Consumers in the early 2010s were finally beginning to understand the value of
interactive streaming, even though, as we’ve seen, it had been available with major
label licensed music since 2002. Two distinct models of interactive streaming be-
came popular. One is the paid-subscription, opt-in licensing model that provides
a highly curated experience for music fans, as provided by Spotify, Apple Music,
Amazon Music Unlimited, Deezer, and so on. The other is the YouTube model,
based on user-uploaded content, free access, ad revenue, and opt-out licensing.
These two models lead to different consumer experiences. YouTube’s
barriers to consumer adoption are virtually nonexistent: you don’t even need to
register to use it, and free unregistered users have access to the full catalog with
no restrictions. Music is available on YouTube that can’t be found on the opt-in
services, including old and rare music as well as music from all over the world.
On the other hand, it’s fundamentally impossible—or at least extremely
difficult—to create a music experience on YouTube that’s as well curated, rich
with information, and easy to navigate as in services like Spotify and Apple Music.
The most obvious problem is that YouTube isn’t just a music service; music exists
on YouTube in a vast ocean of content, most of which has nothing to do with
music. But even if YouTube were somehow limited to music videos only (what-
ever that means), it would still have an enormous problem with one ingredient
that is absolutely essential in offering a curated music experience: metadata.
We’ve discussed metadata elsewhere in this book, but it’s especially crucial
to YouTube’s challenges in building a successful music service. The simplest
way to define metadata is as “data about data.” In the music context, metadata
includes basic factual information about recorded tracks such as artist, title,
label, length, and release date, as well as more detailed information such as
musician and producer credits. It also includes descriptive information such
as genre, tempo, mood, and so on. A single music track can have hundreds of
items of metadata associated with it.
Curation, search, navigation, and discovery are simply impossible for a
music service without metadata that is sufficiently detailed, consistent, ac-
curate, and up to date. Labels supply reasonable-quality metadata to opt-in
services, and those services augment that with carefully maintained meta-
data that they create themselves or get from commercial third parties. But
with user-uploaded content, the vast majority of metadata is nonexistent,
minimal, inconsistent, or just plain wrong.
Streaming Video 317
smoother path to paid subscriptions and keep them within the Google eco-
system rather than losing them to Spotify, Apple, Amazon, or Deezer.
As we mentioned, the result of this strategy was YouTube Music, which
launched in 2018. Google’s approach to creating YouTube Music was typically
“Googley”: it decided to rely on advanced technology rather than humans.
Over time, YouTube was able to fuse the oceans of user-contributed content
with the data that labels feed to Google Play Music to produce an experience
that was music-first rather than video-first and offered a fair approximation
to the highly curated experiences of Spotify and other services. It did so by
using artificial intelligence (AI) and machine learning (ML) techniques—
which we’ll discuss further in Chapter 11—to normalize, clean, and wrangle
the metadata. The user experience was a bit rough at first, but after a couple of
years, it got closer to that of Spotify or Apple Music in terms of the accuracy
and richness of the metadata.24
Interestingly, YouTube Music does not offer user-generated playlists or
any other social features from the main YouTube app, though it does offer
videos. It is an uncluttered, almost “purist” music experience, aesthetically
more in keeping with Google’s search engine than with YouTube itself.
While Google labors to create a true music app out of YouTube, its new
competition is going, to some extent, in the opposite direction. TikTok, as
we discussed, isn’t really a music app—it is a social networking app that uses
a lot of music. It uses music in two distinct ways: users take videos of them-
selves (often dancing) that use music; or users who are musical artists upload
short clips of their own music. An increasing number of established artists
have their own TikTok feeds. And while more and more of those artists are
posting short clips of their own as promotion for full-length tracks (whether
on TikTok itself or elsewhere),xxxiii the bulk of activity for major artists on
TikTok currently still consists of users posting clips containing their music.
And those users aren’t necessarily posting clips with current music: TikTok
has proven to be a potent way of revitalizing back catalog without involve-
ment from artists or labels. One example of this was the astonishing resur-
rection in September 2020 of Fleetwood Mac’s 1977 hit “Dreams” in a TikTok
skateboarding video by user Nathan Apodaca (a.k.a. Dogg Face). The clip
went viral, which led to millions of streams of the song, thousands of paid
downloads, and the song reentering the Billboard Hot 100. Two members of
Fleetwood Mac even recreated the video themselves.xxxiv
But the result of such phenomena on TikTok is exposure and fame for the
user at least as much as for the musical artist.25 Some users try to convert
Streaming Video 319
With one exception that we’ll talk about shortly, YouTube’s influence on
musical creative output has not been proportional to its user base. That’s most
likely because, as we’ve seen, Spotify and its ilk have user experiences that
funnel listeners more toward big-label releases than YouTube does, resulting
in proportionally lower play counts for those artists on YouTube. That, to-
gether with YouTube’s significantly lower royalty payouts, means that artists
and labels may be generally less interested in creating music “for YouTube”
than for the opt-in music services.
Early YouTube successes such as OK Go and the SNL “Lazy Sunday” video
fed the story that YouTube, like p2p and MP3s before it (see Chapter 8), was
going to disintermediate labels and make it possible for anyone to be a star.
As with digital downloads, the occasional exception does not prove the rule.
Instead, and indeed as usual with these technologies, early exceptions have
given way to instances where either the same old intermediaries figure out
how to make the system work for their own purposes or new intermediaries
are established.
YouTube has proven most useful to artist managers and label A&R
people as a place to discover talent, though that talent still relies on those
intermediaries to become successful. There has hardly ever been such a thing
as a label signing a new artist and then the artist goes viral on YouTube; it’s the
other way around. There is no “formula” for making a music video that goes
320 Key Changes
viral on YouTube, but as we’ll see, labels and managers do various things to
increase the odds.
Perhaps the archetypal example of this was the Canadian megastar Justin
Bieber, who was “discovered” on YouTube in 2008 by powerhouse artist man-
ager Scooter Braun. Braun got Bieber signed to Island Records later that year;
the rest is history. Both Bieber and Braun will figure into the stories of subse-
quent YouTube sensations, as we’ll also see.
Another example was a song sung in 2011 by a 13-year-old girl from
Southern California. Her parents paid $4,000 to a small label in LA to write
a song for her, make a recording of her singing it (with a heaping helping of
Auto-Tune), and make a video of the recording for YouTube. The video went
viral, thanks largely to activity on Twitter; and most of the reaction was neg-
ative (“worst song of all time”; “worst video ever made”). That did not stop
Rebecca Black’s song “Friday” from garnering 167 million views on YouTube
during its first month, leading to an RIAA Gold certification and appearances
on MTV, a Katy Perry video, and elsewhere. Black was exceptional in that she
did not sign to a major label, but this was primarily because she was tied to
a contract with the small label and could not get out of the contract despite
years of legal wrangling.
But then those examples gave way to artists who were “discovered” on
YouTube but were actually already stars elsewhere. This has been particu-
larly true for artists from one country getting exposed in other countries with
extreme rapidity, due to YouTube’s global and opt-out architecture, as we
discussed earlier in this chapter.
The first major instance of this came in 2012. As much as “Friday” was
the pop-culture sensation of 2011, it was nothing compared to the video of a
short, chubby South Korean rapper that was released the following year. Psy’s
“Gangnam Style” peaked at no. 2 on the Billboard Hot 100 and topped the charts
in more than 30 other countries. By the end of 2012, it had become the first
video to exceed a billion views on YouTube, a record it held for an incredible four
years and seven months until it was surpassed by “See You Again” by rapper Wiz
Kalifa. Even today it is the eleventh most viewed video of all time on YouTube.
To American fans, “Gangnam Style” seemed to come out of nowhere.
Researchers analyzed its explosive growth on YouTube and found that much
of its early spread was due, once again, to Twitter activity. But the reality was
something different. “Gangnam Style” was hardly a fluke viral hit by an un-
known independent artist from an “exotic” land. Instead, Psy was merely the
first K-pop artist to break in the United States.
Streaming Video 321
the song (counting both versions) topped the charts in a total of 47 countries.
As evidence of YouTube’s vital role in creating that global mega-stardom: un-
like the stars mentioned previously whose play counts on Spotify are greater
than on YouTube, “Despacito” has, at this time of writing, gotten 1.6 bil-
lion plays on Spotify but over 8 billion views on YouTube. It was the most-
viewed video on YouTube of all time for three years until it was unseated
in November 2020 by the “Baby Shark” video—another product of South
Korea, though not one associated with K-pop.
The appearances of Bieber on Jepsen’s video and on Fonsi’s remix were also
examples of the most important way that Google and YouTube changed pop
music creatively: artist collaborations, or collabs for short. While there are
many artistically valid reasons for artists to collaborate with each other (or
for managers or labels to have them collaborate), there are two reasons that
have little to do with creativity. One is search engine optimization. A record
with multiple artists is more likely to show up on more search results. This
trend started in connection with Google searches for records in the early
2000s, but it increased once Google acquired YouTube in 2006 and applied
its search-engine prowess to the service; music discovery on YouTube is
based much more on search than it is on other services. The other reason
is playlists: as we saw in the previous chapters, collabs tend to increase the
number of playlists that will include a given track.
Many collabs involve rappers making guest appearances on pop records
and being credited in the song title as “featuring” (or “feat.”), so that both
artists are more likely to show up in search results. This has not only fostered
the crossover of audiences between pop and hip-hop but also undoubtedly
helped usher in the present era where the two overlap almost entirely—that
is, hip-hop is mainstream pop.
The first instance of credited featured rappers was in the new jack swing
style that developed in the late 1980s as a fusion of hip-hop and R&B. “She
Ain’t Worth It,” a 1990 hit by the Hawaiian star Glenn Medeiros, featured a
rap from Bobby Brown. But other uncredited examples date back even fur-
ther: the first was Grandmaster Melle Mel’s opening rap to Chaka Kahn’s
Prince-penned hit “I Feel for You” in 1984.xxxvi Collabs and featured rappers
on pop hits have become extremely common since then. As a measure of
just how common, between a quarter and a half of the songs on the Billboard
Hot 100 around this time of writing are pop songs with “featured” drop-ins
(mostly rappers), multiartist collabs, or both.
Streaming Video 323
Today most of the top music videos on YouTube are by major-label art-
ists. At this writing, almost all of the top 10 music videos on the US YouTube
chart tend to be on labels owned by one of the Big Three. As mentioned
earlier in this chapter, labels and artists usually put up “official” videos for
their releases; the vast majority of label releases today have official videos.
YouTube Channels enable artists to create more personalized presences
there than on services like Spotify. Channels have evolved over time and
currently include Twitter-like social media posts, merch stores, “about the
artist” pages, recommended channels, and, of course, videos and playlists.
Many artists post videos on their channels that aren’t just music videos.
For example, Cardi B’s channel currently features videos of interviews, TV
appearances, and so on; and her store sells T-shirts and jewelry. Lady Gaga’s
channel currently has content touting her makeup line and her charitable
foundation, while her store sells vinyl, CDs, and cassettes.
As for TikTok, it is influencing artists’ creative processes in numerous
ways as its user base continues to grow. TikTok treats audio clips somewhat
like hashtags:26 it lumps videos together if they use the same audio, and it
highlights trending audio clips for users. You can select an option to create a
video with the same audio as the clip you’re watching. Users can thus boost
the popularity of their videos by choosing trending audio to go with them.
This leads to a virtuous cycle that can give certain audio clips outsized popu-
larity very quickly.
This has led artists to consider ways to create TikTok-friendly content.
Some artists, as we mentioned earlier in this chapter, are crafting brief pro-
motional videos of their songs to become trending audio clips on TikTok—
sometimes because labels are requiring them to do so.xxxvii Short clips of
songs can go viral on TikTok and lead the full-length songs to become hits.
For example, a short clip containing the hook from the song “Bad Habit” by
R&B singer/guitarist Steve Lacy27 appeared in almost half a million TikTok
videos on its way to reaching no. 1 on the Billboard Hot 100; the audience
at a Lacy concert in October 2022 demonstrated that they didn’t know the
words to the song beyond those in the viral clip when they stopped singing
along at the end of the lyrics that appeared in the clip. Another trend in this
vein has been to create sped-up remixes of songs made for TikTok—which
are also helping to push the songs up the charts.xxxviii And, consistent with
the notion that TikTok is at heart a social network, pop/rock artist Gayle
wrote the song “abcdefu,” nominated for a Song of the Year Grammy in 2023,
324 Key Changes
in response to a video that she posted on TikTok in which she asked for song
ideas to be posted in the comments.
The next logical step is for artists to create musical short-form content
specifically for TikTok (and/or YouTube Shorts, and/or Reels on Facebook
or Instagram). Given that most TikTok videos are shorter than the 20–
25-second average length of a pop song’s chorus,xxxix that could lead to
changes in pop song structures in the near future: choruses could shorten,
or songwriters could craft songs that contain complete hooks within their
opening seconds. Meanwhile, one sign of TikTok’s movement away from the
record label system is that its biggest current homegrown music stars (based
on earnings), such as Bella Poarch, Dixie D’Amelio, and Loren Gray, have yet
to make a dent in the Billboard charts.
Conclusion
Netflix, Hulu, and Disney+by the tens of millions. But the window of op-
portunity to charge mass-market consumers for music videos may be closed
forever.
At the same time, the newer music video services are engaging hundreds
of millions of users in ways that are drifting further and further away from
the music itself. YouTube started out with user-contributed videos, but the
much smaller number of artists’ official videos now get the vast majority of
views. But now TikTok is primarily about users’ dances and lip-syncs; artists
and labels have yet to find ways to establish the same primacy on these serv-
ices as they did on YouTube.
In other words, from the artists’ and labels’ perspectives, audio services
still rule. And a new breed of technologies and services is putting the focus
back on audio, as we’ll see in the next chapter.
11
Artificial Intelligence and Voice Interfaces
You Took the Words Right Out of my Mouth—Meat Loaf
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0011
Artificial Intelligence and Voice Interfaces 327
Soon after the invention of the first computers, it became apparent that
they could be programmed to exhibit “intelligence” of the type normally
328 Key Changes
small sample of the results and their assessment of what was actually said was
fed back into the system to improve precision.v
Google, meanwhile, introduced the Google Home smart speaker in
late 2016, a full two years after Amazon’s launch of the Echo. It marketed
and priced the device aggressively to catch up with Amazon. Google had
introduced voice search a few years earlier, but it was only capable of passing
commands to Google’s search engine; the Google Assistant technology that
powered Google Home (now Google Nest) devices was a two-way voice-
response system like Alexa and Siri. Together, Amazon and Google effec-
tively operate a duopoly in the smart speaker market; both companies license
their voice-response platforms to third-party smart speaker makers such as
Bose and Sonos, but not to each other or to Apple.4
Since the launch of the smart speaker, technology innovations have
improved the performance of voice assistants on smartphones as well.
Robust mobile data networks now connect to the cloud reliably when on the
go, which enables access to expanded computing and storage. Headphones
and earbuds equipped with active noise cancellation help to distinguish
spoken commands from background sounds. In addition, it is easy to hook
up your phone to take advantage of the microphones, speakers, and display
when listening in your car.
The ability to deliver the right song in response to a voice query is inher-
ently more difficult than providing responses to a text search. When you
search for an album, song, or artist on the Spotify app, the service’s response
is a list of candidates akin to the pages of links in Google search results.
In most cases, that list contains the item being searched for, and the user’s
click confirms that (and generates data to improve future searches as well).
Providing a list of choices means that there are more opportunities to include
the correct response even if the user misspells the artist’s name or song title.
With voice search, there is simply less room for error, as the use case requires
the selection of a single track to play from the millions of options. And in
many cases, those options can be tricky to pronounce (such as Ke$ha) or am-
biguous given frequent changes to artist names (such as Sean Combs a.k.a.
Puff Daddy a.k.a. P. Diddy a.k.a. Diddy).
The challenge of recognizing voice commands is even more daunting
when you consider how the metadata related to music can be interpreted in a
variety of ways. When using the word “Chicago,” it could refer to a rock band
of the 1970s or to a query about the weather in the midwestern city. And the
word “jazz” could be a genre of music or the NBA team located in Utah.
Artificial Intelligence and Voice Interfaces 333
generates data on a scale that dwarfs even these numbers. The vast majority
of the most watched YouTube videos are music videos.viii The number of
songs streamed across Apple, YouTube, Amazon and the other major serv-
ices in the United States alone blew past the one trillion mark in 2019.ix That
total is five hundred times larger than the largest number of units sold ever
tallied in a year—without even including other streaming platforms that li-
cense music, such as TikTok and Peloton.
Other useful song popularity data comes from non-royalty bearing events
on other platforms. Shazam, the music identification application now owned
by Apple, is used by more than 225 million people per month. A Pandora
user giving a “thumbs up” or a Spotify user deciding to follow a specific artist
or an Apple Music listener skipping a track at the 20-second mark are just a
few examples of useful signals that illuminate consumer preferences better.
And that’s not all. Music artists hold five of the top ten Twitter accounts
at this time of writing, with over a half a billion followers in total.x Ariana
Grande, Selena Gomez, and Beyonce collectively have more than a billion
followers on Instagram.xi And we haven’t even considered individual fans
interacting with artist content on these social platforms by liking a par-
ticular Instagram photo or a Facebook post. Companies focusing on live
performances, such as Bandsintown and Ticketmaster, have millions more
entries in their databases covering interest in shows and purchasing tickets.
The need to gather and analyze these enormous data sets covering fan en-
gagement on streaming services and social networks has created the op-
portunity for a variety of companies to offer business intelligence to the
industry.xii
This profusion of data about how fans interact with songs and artists is cru-
cial in building models that are predictive of mass-market success. Creating
music requires detailed information about the songs and performances
themselves. There are many examples of companies investing in song anal-
ysis, although the initial applications are not aimed at generating new music.
Shazam creates an audio fingerprint that uniquely identifies a track based on
a wide range of audio characteristics. It works similarly to technologies for
managing copyrights on user-uploaded content platforms, such as YouTube’s
Content ID system, which we saw in Chapter 10; but Shazam is optimized for
identifying songs via smartphones’ microphones in noisy settings.
Pandora used the Music Genome as the key to its recommendations for
songs to be included on their personalized radio channels. As we saw in
Chapter 3, the company hired musicologists by the dozens to listen to every
Artificial Intelligence and Voice Interfaces 335
track and detail a range of song characteristics based on their expertise. This
yields excellent data, but the approach is simply not scalable to the tens of
thousands of new songs added every day; so Pandora resorted to augmenting
its team of musicologists with automated analysis techniques to make its
music catalog competitive in size to other music services. The Echo Nest, a
spinout from the MIT Media Lab, developed automated tools to analyze the
audio and text associated with songs, including websites, blogs, reviews, and
so on. In 2014, Spotify bought the Echo Nest to add to its engineering exper-
tise in these areas. Now the technology examines every one of the 100 million
available tracks in Spotify’s catalog to determine audio attributes including
beats per minute, length, key, “energy,” and “danceability.” These character-
istics drive the Discover Weekly playlist mentioned earlier, as well as many
other song recommendations on Spotify; they are accessible to anyone via an
application programming interface (API) on the Spotify Developer platform.
The other important input to AI algorithms in voice-response systems is
data about the music itself, a.k.a. metadata. Labels really want to make sure
that when the user asks a smart speaker to play “some crunk,” “that new
Olivia Rodrigo song,” “some 70s rock,” or “some jazz with Wayne Shorter on
saxophone,” it selects the right tracks. In addition to information about the
inherent characteristics of the music, the metadata needs to include informa-
tion about genre, instrumentation, featured artists, release date, focus track,5
and so on. We saw in Chapter 10 that YouTube used advanced ML techniques
to wrangle discographic metadata from the vast ocean of user uploads to
create a decent user experience in YouTube Music. But quality metadata that
describes the music itself can’t be determined through automated analysis;
someone has to assign correct values for all that metadata so that the smart
speaker can do its job well. That’s a huge job.
Luckily, smart speaker makers agree that good metadata is important
for producing the best results for users. So the technology companies have
been collaborating with labels on a standard for metadata about music called
Media Enrichment and Description (MEAD), which is part of a suite of
standards6 for communicating information about music among labels, music
services, and other participants in the music supply chain. Such is the impor-
tance of the single-search-result smart speaker that it has labels working ac-
tively on metadata after decades of treating it as an afterthought.
It is not only the data itself that is being made more broadly available.
Google and Amazon already enable many startups and developers to scale
their businesses cost-effectively by accessing inexpensive processing and
336 Key Changes
checking the weather, or setting a timer. But the number one reason cited
by consumers for purchasing a smart speaker and the most frequent use is
listening to streaming music.xvii Roughly 40 percent of users in the United
States take advantage of that functionality every day.xviii So it is no surprise
that almost half of smart speaker owners increase the amount of music they
listen to. Moreover, almost 40 percent of smart speaker owners decide to
pay for a music subscription service, which provides a significant boost to
revenues. Amazon in particular has benefited from the popularity of smart
speakers for music streaming, as we’ll see shortly.
The current state of AI technology for music creation makes it more suit-
able for reducing costs than for generating additional revenues. However, the
trajectory is clear and, as the technology matures, it will be capable of creating
derivative works based on existing music or entirely new music that resonates
with fans and earns substantial revenues. Today, artists can avail themselves
of AI-powered tools to generate ideas for lyrics or instrumentation to assist
in the creative process. The feedback from these collaborations will fuel the
improvement of the ML tools to become increasingly self-sufficient.
The popularity of the ambient music genre on the streaming services
means that distributors could use AI-generated music to reduce their costs
of content and to improve their margins. It has already been reported that
numerous “fake artists” have tracks on Spotify that have been prominently
featured on these ambient music playlists and are garnering significant ac-
tivity. These reports assert that the service has a special arrangement with
the creators of these tracks to pay reduced royalty rates.xix Clearly, using
AI models to generate content would be even more advantageous in terms
of reduced royalty payments. Just the prospect of such activity reducing
payments for “real” artists could lower operating costs and generate lev-
erage for the streaming services in their future royalty negotiations with
record labels. In November 2022, Tencent Music Entertainment announced
that it had used AI technology to generate over 1,000 songs with replicated
human voices and that one of those tracks had been streamed over 100 mil-
lion times.xx
Though they are not yet ready to write a hit pop song, AI models can create
“functional” music that is suitable for a YouTube video or a low-budget film.
This foreshadows a major shift in the economics of production music. Drew
Silverstein’s background as a composer led him to found Amper Music in
2014. The company has developed an online portal, Amper Score, that an-
yone can use to create a piece of music designed to match a particular piece
Artificial Intelligence and Voice Interfaces 339
of video content. One can select the genre, mood, instruments, tempo, and
the key moments in the video requiring emphasis. In a few seconds, the tool
creates a new piece of music using Amper’s proprietary library of millions of
music samples covering an enormous variety of musical instruments. The
musical selection can be modified easily until it meets the user’s needs ex-
actly. When the user is satisfied with the result, they can download a copy and
obtain a perpetual license to use the newly created track for under $100.xxi In
2020, Shutterstock purchased Amper Music to complement its online image
licensing business.
Amazon’s Prime service launched in 2005 with expedited shipping for all
purchases; soon after its launch, customers who subscribed were spending
even more with the e-commerce giant. Amazon has continued to add items
to Prime since then, making the offer more valuable for subscribers while
raising the annual price too. Just a few months before the introduction of
the Echo smart speaker, Amazon unveiled its Prime Music offer. This was
an advertising-free streaming service that gave all Prime subscribers ac-
cess to two million songs at no additional charge. Though that sounds like a
large music catalog, it was far smaller than what Spotify was offering, which
enabled Amazon to negotiate lower-cost licenses with the major labels. In
addition, the number of songs made accurate voice recognition a somewhat
less daunting task for the Echo speaker. Amazon certainly had the Echo
speaker in mind when implementing Prime Music, though the two were
not developed together.xxii Moreover, the millions of users who were soon
interacting via voice created extensive data to improve the AI recognition
models in the cloud. That helped to prepare for the 2016 launch of Amazon
Music Unlimited, which offered access to a full catalog of tens of millions of
tracks, matching the scale of the other major services. The Unlimited tier re-
quired an additional payment, but it was available to anyone, even those who
did not subscribe to Amazon Prime.
This tight coupling of devices and connected service capabilities enabled
distribution that has proved as valuable to Amazon as the combination of
the iPod and the iTunes Store was to Apple a dozen years earlier. It helped
to accelerate Amazon’s growth in streaming music and position it among
the leaders in streaming alongside Spotify and Apple. Amazon announced
340 Key Changes
technology will creep ever closer to becoming a market reality in the near fu-
ture; Korean researchers have already created an AI model to create a piano-
only cover version of any popular song. AI models will lower the expense
of creating more extensive variations of the same track, including shifting
instrumentation, genre, tempo and even the language of the singer.xxvi
Apple has already acquired a startup called AI Music that has announced
plans to use AI to shift the style of existing music.xxvii Once the quality of
such transformations is good enough and the costs are low enough, we can
expect to see a proliferation of new versions of songs. These alternate takes
will then be positioned on various distribution platforms and playlists that
cater to different audiences or languages. Some will fail to catch on with fans
while others will prove successful. The additional listening and popularity
will then yield a greater share of royalties. The data resulting from both the
successes and the failures will then be fed back into the AI models to improve
the ability to craft ever more popular variations.
Taking this idea a step further, AI models will create versions of tracks that
are even more finely tuned to a user’s tastes beyond shifting to another genre
or giving the song a more upbeat tempo. The services have the historical data
on every listener’s preferences and can determine what engages them the
most. With lower costs and greater computing power, they will be able to
stream truly personalized playlists that feature versions of songs matched to
those preferences. If you want “more cowbell,” that’s what you’ll get.
Radio program directors have long curated the list of songs to play to
keep listeners tuning in. For streaming services, algorithms built using AI
use your listening history and those of people with tastes similar to yours via
collaborative filtering to personalize the songs recommended to each user.
Collaborative filtering is a type of algorithm that compares your behavior and
preferences with those of other users so that it can generate recommendations
for you based on other users’ behavior.7 The better the recommendations, the
more likely fans will keep listening and subscribing because they enjoy the
service. Listener reactions to song recommendations, whether a “thumbs up”
ranking or skipping a track quickly, generate additional signals to improve
those recommendations. Spotify’s popular Discover Weekly playlist is one
such example where ML techniques automatically incorporate user reactions
to improve the results over time.xxviii
AI models can improve not only the quality of each song selected for
personalized lists but can also enhance the overall experience in ways that
go beyond selection of music. Often, streaming services have mismatched
342 Key Changes
volume levels and pauses between their songs and advertisements. Radio
stations use people to solve these issues and to provide information and
branding between the songs (which are often centrally programmed, given
the corporate conglomerates that own many stations). The best DJs relate
interesting facts about the band or the song while the intro plays and then
“hit the post”—finish up just as the lyrics kick in. The streaming services
had no cost-effective way to offer similar capabilities. Super Hi-Fi, a Los
Angeles-based startup, uses AI tools to provide a comparable experience at
a low cost.xxix
Shifting tastes in music point toward another way AI-generated music will
play a role. Over the course of 2020, many people listened to more streamed
entertainment to deal with the stress of the COVID-19 pandemic. Ambient
music, as a genre, grew in popularity because as a background listening expe-
rience it provides a sense of calm or serenity for many people.xxx Ambient is a
type of instrumental music that eschews traditional musical rhythms, struc-
ture, or melodies in favor of tone and atmosphere. Spotify playlists featuring
ambient music, such as Deep Sleep, Peaceful Piano, and Deep Focus, collec-
tively have over 12 million followers.
Compositional elements make the ambient genre particularly amenable
to the current capabilities of AI models. Daniel Jeffries has used Google’s
Project Magenta tools to create ambient music; he explains as follows: “The
hazy softness of ambient music gave us a little more leeway to screw it up just
a bit but still have something that sounds wonderful.”xxxi As another example,
Mubert is a new service that uses AI to produce generative ambient music
that streams royalty-free and is available via API.xxxii
Many artists change their names as their career evolves or opt for spellings
that help them stand out from the crowd of artists (e.g., A$AP Rocky). While
these choices complicate the tasks of recognizing and pronouncing names
for voice interfaces, creators are making decisions about song titles that
simplify things. The data from Spotify shows two distinct trends that both
have ramifications for voice interfaces. First, song titles are getting shorter in
terms of the number of words along with the songs themselves being shorter
(as discussed in Chapter 9).xxxiii The process of recognizing each individual
spoken word has an associated error rate, so fewer words means that there is
Artificial Intelligence and Voice Interfaces 343
time, more artists will rely on AI-based assistants as part of their creative
process.
One of the more problematic AI applications for society is creating
deepfakes, which are synthetic images or voices that are designed to look
or sound like a known person without the person’s permission. In some
deepfakes, the content shows an individual saying or doing things they never
actually did. For example, the BBC posted a video showing Queen Elizabeth
delivering an “alternative” 2020 holiday message to emphasize the risks of
the technology as a tool to spread misinformation.xxxviii Kendrick Lamar
deployed deepfake technology to morph his own image into a series of other
famous people, from Kanye West to O. J. Simpson, in his music video for
his song “The Track, Part 5.”xxxix As the technology improves, it will become
harder and harder to determine whether a particular statement or image is
real or fake.
Given the vast amount of video of musical artists speaking, there is more
than enough training material to create deep fakes showing them making
controversial statements about politics or race that could complicate their ca-
reer or damage relationships with many fans. Google’s AI team has created a
tool called Tacotron 2 which renders any text in a convincing human voice.xl
There is no lack of training material of rap performances either, so it is a
short step to move from faked spoken words to a “counterfeit” musical selec-
tion. Jay-Z’s voice and his vocal style are both distinctive and were used to
train an AI model to imitate him. Though he has never rapped the “To be, or
not to be” soliloquy from Hamlet or Billy Joel’s “We Didn’t Start the Fire,” you
can find YouTube videos of both faked performances.xli Given how close that
song is to rap performance, these fakes are easier to create than a fully me-
lodic performance. However, that does not mean that people are not trying
to do exactly that. Using the OpenAI platform to generate a Frank Sinatra-
like rendition of Britney Spears’s “Toxic” is a bit more of a stretch in terms
of the quality of the output.xlii Jay-Z and Frank Sinatra’s estate will not view
these appropriations of their musical personas positively, particularly when
the new works were created with neither permission nor remuneration.8
But why stop at having a known artist “artificially cover” an existing song
when an AI model can create a completely new track building on an extensive
data set of previous performances and compositions? In March 2019, Google
created a “Doodle” for its home page that honored Johann Sebastian Bach on
the composer’s birthday.xliii The user inputs two measures of notes, and using
rules defined by over 300 Bach compositions, the AI algorithm crafted the
Artificial Intelligence and Voice Interfaces 345
appropriate harmonies in just seconds and played back the results. OpenAI
has used its Jukebox tool to create country music in the style of Alan Jackson
and pop in the style of Katy Perry by training its algorithms on other songs
by those artists. The quality of the music produced by these tools is not great,
but is better than it was a year ago and will continue to improve.
There is one settled case that offers some narrow guidance on these
questions, though it relates neither to music nor to AI-enabled tools. In 2005,
a monkey in Indonesia took a series of selfies using a nature photographer’s
camera. The photographer published the photos in a book. The animal rights
organization People for the Ethical Treatment of Animals (PETA) filed
suit claiming that this action infringed on the monkey’s rights under the
Copyright Act. The court found that only humans can hold a copyright.xlv
On a related question, the US Patent and Trademark Office ruled in 2020
that an AI may not be listed as an inventor on a patent.xlvi In February 2022,
the US Copyright Office extended that same logic by ruling that an AI al-
gorithm cannot be granted a copyright for its artistic creation.xlvii And if we
accept that conclusion, is a new work created autonomously by an AI model
even copyrightable? The Copyright Office has already published guidance
stating that the answer to that question is “no,” that works by generative AI
technologies cannot qualify for copyright registration without substantial
human input (of which a simple text prompt is not sufficient).xlviii
These are the sorts of questions that IP lawyers love to explore, and they
have already been the subject of a great deal of legal scholarship and debate.
The IP law community has taken an active role in convening experts to dis-
cuss the questions and in soliciting opinions from both the technology and
creative communities. On February 5, 2020, the Copyright Office and the
World Intellectual Property Organization (WIPO) held a symposium on the
issues of AI and creativity. The panelists addressed such issues as the level
of human interaction required to make an AI-generated work eligible for
copyright protection and the rights issues associated with using copyrighted
works to train AI models, as well as how these AI-related policies relate to the
goals of copyright itself. If the law’s protections are intended to incentivize
the creation of new works, do AI models even require such incentives?
Also recognizing the importance of AI, the US Patent and Trademark
Office solicited input on thirteen questions related to copyrights, trademarks,
and other IP rights in 2019.xlix Not surprisingly, the responses to the questions
exemplified the split between the creative and technology industries, with
the former arguing for stronger protections of their intellectual property in-
cluding the need to obtain licenses for training material. These processes will
continue well into the future; the Copyright Office has been holding a public
consultations on AI and copyright during 2023.
In the absence of definitive law, some AI companies are explicitly stating
their treatment of copyrights as part of the terms of service to use their
Artificial Intelligence and Voice Interfaces 347
platform. They take the position that the entity creating and offering the
AI software deserves a stake in the work product produced. For example,
Amper Music’s Terms of Service specify that the user assigns and transfers
any and all rights to the music created to the company. The user receives a
royalty-free license to use the music created in perpetuity, but the copyright
belongs to Amper Music.l
We’ve already cited several examples of AI models creating music in the
style of a particular artist or composer. That artist’s catalog of music makes
up the data set used to train these models. As expected, rights holders be-
lieve that such training requires a license, which conceivably could include
approval of any commercial uses for the AI-generated content along with
payments. In July 2022, all the UK bodies representing artists complained to
the government about data mining10 of music to create AI models without
remuneration.li Certainly, any uses of the newly created content that relies on
the name and image of the original artist without obtaining the appropriate
permissions would probably be blocked.
Jay-Z’s legal team chose to file DMCA takedown notices (see Chapter 10)
to remove the YouTube videos containing AI-generated deepfake vocals rap-
ping the Hamlet soliloquy and “We Didn’t Start the Fire.” Their claim was
that “this content unlawfully uses an AI to impersonate our client’s voice.”lii
Filing such a notice is simple, but in this case, YouTube rejected the claim and
allowed the videos to remain available. Their decision may have relied on the
logic that imitating someone’s vocal mannerisms is not copyright infringe-
ment. Creative Commons offered the following view: “It is ill-advised to
force the application of the copyright system—an antiquated system that has
yet to adapt to the digital environment—on to AI.”11 The unidentified owners
of the channel called Voice Synthesis that posted these videos to YouTube
asserted that their videos are parodies, which are protected under fair use.
Jay-Z’s legal team could decide to push back on these theories and go to court
to have their claims adjudicated, but, so far, they have not elected to do so.
Although this example used Jay-Z’s voice only, we can expect to see deepfake
vocals performed by characters that are the product of computer algorithms
as well. The K-pop group Eternity has already cracked the Spotify charts in
Korea with eleven AI-generated members.liii
There are other possibly stronger legal arguments that artists can make with
regard to these deepfake performances. The first involves impersonations of
a musical artist without their permission and has nothing to do with whether
AI was used to create the imitation. In the 1980s, Ford Motor Co. put together
348 Key Changes
an ad campaign for its Lincoln brand to appeal to Yuppies that relied on nos-
talgia for several songs of the 1970s. The company approached Bette Midler’s
manager to request a recording of “Do You Wanna Dance?” for the ads and
were refused. Ford liked the concept so much that it decided to re-record
the song and hired a singer to imitate the Divine Miss M’s style. The impres-
sion was so good that many people thought it was Midler herself singing. She
thought that Ford should have taken her “no” for an answer and filed suit.
The US Court of Appeals for the Ninth Circuit overturned an initial lower
court in favor of Ford. The higher court found that even though Ford did
not refer to Midler by name, nor use her likeness in any way in the ads, it
did misappropriate her voice, and it ruled that it is unlawful to imitate a fa-
mous person’s voice for commercial purposes without their express consent.
Artists will definitely use that argument when unauthorized deepfakes of
their performances are exploited.liv
This rationale regarding the use of an artist’s voice without permission
may be viewed as a more expansive application of the right of publicity (an-
other branch of law, separate from copyright). There is no federal statute
protecting an individual’s right of publicity, although there have been some
discussions that such protections should be considered. Instead, many
states offer legal protection of an individual’s identity (Including name, like-
ness, and persona) from being exploited for commercial purposes without
permission. Some states provide this protection for celebrities only, while
others do the same for all individuals. Regardless, we can expect to see mu-
sical artists using the right of publicity to protect their rights in the age of AI
deepfakes.
The legal framework for using copyrighted material as training data has
not yet been clearly delineated. The technology companies do not see the
need to obtain a license to train an AI model on a catalog of copyrighted
songs and performances. They will argue that having an algorithm
identifying characteristics and attributes contained in the originals is no dif-
ferent from a person listening to all of Taylor Swift’s songs and then writing
a new one in her style from the “lessons” they absorbed. If a portion of one
of her songs was copied and reused in the new work, then the traditional
remedies associated with copyright infringement are available. However,
if an algorithm simply “listens” to her catalog and extracts features mathe-
matically, in a manner akin to audio fingerprinting technologies, then tech
companies will argue that the new song is not a derivative work requiring
permission.12 Instead, they will take the position that the law should treat
Artificial Intelligence and Voice Interfaces 349
the AI company and its model in the same fashion as any other Taylor Swift-
inspired “wannabe” and therefore that the newly created work should be
covered under fair use.
Interestingly, the first lawsuit filed alleging improper use of data for
training AI models has nothing to do with what we typically deem the cre-
ative arts, such as music or photographs. Copilot is an AI-based tool that
generates code to help programmers while they are writing it. It was trained
using the extensive database of programs on the Microsoft-owned platform
GitHub, and it offers to insert chunks of code taken from GitHub repositories.
Although much of the code on GitHub is made available to others under
open-source licenses, the parties filing suit claim that using it as training data
violates GitHub’s terms of service and that Microsoft has failed to provide the
proper attribution to the creators.13,lv
In early 2023, two separate lawsuits were filed by visual artists and Getty
Images (one of the largest image licensing services) against AI-powered text-
to-image generators. The parties claimed that these platforms were using
copyrighted images to train their algorithms without permission or compen-
sation for the creators. The music industry will be monitoring the arguments
and decisions in these cases carefully, as they could establish precedents that
are likely to apply to AI-generated music.lvi As Michael Nash, Chief Digital
Officer of Universal Music Group, stated publicly, “Unless creators are
respected and fairly compensated when their works are exploited to train AI,
the world’s creators will suffer widespread and lasting harm.”lvii
There are other scenarios that could make the licensing questions even
murkier. What if the technology company does not disclose publicly which
artist’s catalog of music was used to train their system or simply applies the
same methodology to the entire Billboard 200 to extract the key elements
from all these songs? The nature of ML is such that the precise steps used
to refine the models are unknown, and the resulting model is equivalent
to a “black box.” That is one reason that unintended biases can become
embedded in AI models used for tasks such as approving mortgages or
selecting job applicants. Therefore, reverse engineering a particular piece of
AI-generated music to determine which songs were used to determine the
parameters of the resulting model is very difficult, if not impossible. Given
all these uncertainties, as commercial uses of AI music expand, we expect
that questions of licensing music for the training of AI models and relevant
ownership of the resulting works are likely to be debated in the courts for
many years.
350 Key Changes
Conclusion
Unlike its portrayal in so many Hollywood movies, AI does not possess any
malicious intent. It is a tool that can be used for good or for ill. The ocean of
consumption-related data will continue to make voice recognition models
easier to use and more accurate when determining the speaker’s intentions.
The availability of AI-powered models for generating music will unleash an
enormous amount of creativity. It will empower individuals with abilities
that they did not previously possess to make interesting and engaging music.
Much uncertainty remains. But, as we’ve seen time and again, the uses of AI
technology will have repercussions across the rest of the business.
12
Coda
Time After Time—Cyndi Lauper
For more than 140 years, technology and the music industry has been a story
of succession. New formats enabled by innovation come to the fore, and the
repercussions propagate throughout the business. Stepping through the
formats in chronological order has allowed us to examine these impacts on
each of them in depth through the lens of the 6C Framework. Now we’d like
to alter our perspective and examine each of the 6Cs across the span of time.
As you’ll see, some different and useful insights emerge. As Mark Twain is
often quoted, “History doesn’t repeat itself, but it does rhyme.” For example,
“The music you want, wherever you want it” sounds like the marketing pitch
for any number of today’s streaming music services. In fact, this phrase was
included in an advertisement for an early phonograph from a century ago.i
Though the circumstances are never exactly the same, common themes
that echo through the decades merit examination. Moreover, these themes
provide relevant lessons for other industries in dealing with their own
disruptions caused by technology. We offer one such theme for each of our
6Cs here.
Edison’s phonograph used cylinders for recording music. Berliner chose flat
discs to fulfill that function for his gramophone. These decisions created a
conflict in the market that took years to resolve. Such rivalries of incompat-
ible technologies are referred to as “format wars”; they recurred throughout
Key Changes. Howie Singer and Bill Rosenblatt, Oxford University Press. © Howie Singer and Bill Rosenblatt 2023.
DOI: 10.1093/oso/9780197656891.003.0012
354 Key Changes
the history of recorded music, including 45s versus LPs and 8-track tapes
versus cassettes.
In each of these battles for dominance, one of the formats vying to be the
winner held a clear technical advantage in what would seem to be a crucial
characteristic: audio quality. However, even when the product provided su-
perior sound, it did not guarantee victory in the market. Instead, consumers
repeatedly chose convenience over quality as an even more critical success
factor.
In general, cylinders provided more accurate reproduction of the original
music than discs. That quality was uniform from the beginning to the end of
each recording. For discs, the constant rotation speed meant that the stylus
traveled a shorter distance over a given period of time as it made its way to-
ward the center of the record, and with each diminution of distance came a
corresponding reduction in sound quality. However, people found it far more
convenient to play and to change recordings on discs. The Gramophone was
designed to deliver entertainment only and didn’t include any recording ca-
pability; this resulted in a less complex product that was simpler to use. The
form factor of the disc also made it easier to identify recordings by reading
the center label rather than trying to discern the far less legible text etched
on the ends of cylinders. And storing discs in their sleeves took up much
less space than stacking all those cylinders in their cardboard cases. All of
that added up to a friendlier experience that proved crucial in winning the
market-share battle. Even Edison eventually conceded and, despite more
than a decade of championing the superior sound of cylinders, began to re-
lease music on discs.
The 12-inch LP provided better sound quality and longer playing time
than the 78 record. With more than 20 minutes of playback per side, the LP
could begin to address the requirements of lengthier classical repertoire with
far fewer discs. RCA initially positioned its competing 45 format as suitable
for the classical market as well, in part because it offered even better sound
quality than the LP spinning at 33 rpm. But the 7-inch discs could only hold
the same four to five minutes of music per side as 78s. To address this issue,
RCA created an automatic record changer. But the convenience of a single
disc providing over 40 minutes of entertainment with a single interruption
to flip to the B-side carried the day—not to mention that classical concertos
and most symphony movements could be heard without interruptions in the
middle of the music.
Coda 355
Of course, the advent of rock & roll in the mid-1950s ensured the lon-
gevity of the 45, as its capacity was a perfect match for the music that millions
of teenagers were hearing on transistor radios. Still, the revenues garnered
from LPs have always dwarfed the dollars earned by 45s.
The technology story for the tape format represents the best example
of a convenient experience winning out over audio quality. Reel-to-reel
machines provided sound that was superior to 8-track tapes. But the sim-
plicity of inserting a cartridge into a player and pushing a single button won
over handling and threading tape. The introduction of the Compact Cassette
brought a further degradation in quality that would eventually be offset by
several innovations, including Dolby noise reduction and chromium dioxide
tape. Once again, improving the user experience proved to be more impor-
tant than the decline in audio performance: cassettes were half the size of 8-
tracks and offered random access through fast-forward and rewind.
As we noted in Chapter 5, cassettes’ convenience motivated the
innovations that improved their sound quality. The same innovations made
their way into 8-tracks: Dolby-equipped component 8-track recording decks
and blank chrome 8-track tapes were both available. But even though both
tape formats enabled recording in the home, consumers didn’t embrace
8-track recording despite its potential for better sound quality because of
the format’s relative lack of convenience. Cassettes became the medium of
choice for home recording—including recording music from the radio and
copying LPs onto cassettes. So although record labels preferred 8-track as a
commercial release format and tried to hamper the growth of cassettes by
withholding their catalogs from the format in its early days, their conven-
ient features proved irresistible, and the labels had to accept that reality. The
cassette became the dominant tape format, surpassing even vinyl for a time
in the 1980s and 1990s, despite the fact that many commercially recorded
cassettes were duplicated cheaply and didn’t sound very good.
As technology evolved to enable the distribution of music files across
the Internet, the same themes recurred. The labels authorized limited
experiments with startups, but they chose to work with companies like
Liquid Audio and a2b music that implemented Digital Rights Management
(DRM). The record companies viewed securing downloaded files as essential
to deterring unauthorized distribution. Their partners had to deploy DRM
that constrained how consumers could enjoy and share the music. Moreover,
MP3 compression was included in software that made it simple to add any
356 Key Changes
song to your library by “ripping” files from CDs without any restrictive usage
rules. The AAC compression scheme used by a2b music provided better
audio quality than an MP3 file of comparable size. But there was no way for
users to create AAC files from their CD collections at that time. The virtu-
ally limitless catalog of MP3s unencumbered by restrictive rules provided
the frictionless experience that trumped audio quality.
that. In the mid-1990s they developed a machine that made it easy to brew
single cups of the coffee of your choice without having to measure, grind,
or clean anything—or commit to a full pot that would taste stale or burnt
by mid-morning. You’d simply insert a pod in a slot in a brewing machine,
place your mug underneath the spout, press a button, and get your coffee
in less than a minute. The next person to come along would open the slot,
throw your used pod in the trash, and start again. Sylvan and Dragone chose
a Danish word meaning “neat” or “proper” as the name for their system, and
in 1998—after six years of trial and error—Keurig delivered its first machines
and K-Cup coffee pods to offices.
The Boston-based company produced its first models for home use in 2004
as the third wave was gaining steam.1 Coffee for the K-Cups initially came
from second-wave Green Mountain Coffee Roasters in nearby Vermont;
Green Mountain would eventually buy the Keurig company. As sales began
to take off, large food concerns such as Salton, Sara Lee, and Procter &
Gamble began to introduce single-cup brewing systems, but Keurig became
the dominant player.iv
As Keurig’s popularity soared, other coffee producers began to obtain
licenses to produce K-Cups of their coffees. When some of its patents expired
in 2012, Keurig introduced proprietary ink-based technology to ensure that
only those who took licenses from it would be able to produce K-Cups that
worked in the brewing machines—a sort of DRM for coffee pods. (Like many
DRMs, Keurig’s was eventually hacked.)v
Keurig achieved an overwhelming leadership position in the single-serve
coffee-maker market over other systems such as Nespresso (from Nestlé of
Switzerland) and Tassimo (from JDE Peet’s of the Netherlands). The Green
Mountain Keurig company ultimately merged with Dr. Pepper Snapple
Group in 2018 to form Keurig Dr Pepper, a publicly traded behemoth with a
market capitalization not far off that of Starbucks. The company took in over
$4.2 billion from Keurig machines and K-Cups in 2019.
However, Keurig had two “dirty” secrets. One was that K-Cups were
wasteful and bad for the environment. The other was that the coffee just
didn’t taste very good. Any serious coffee fan will tell you that the most im-
portant route to a good cup is to start with freshly roasted beans and grind
them just before brewing. K-Cups, filled with preground beans that had been
sitting on warehouse and store shelves for who knows how long, could never
compete. Wirecutter, the New York Times’s consumer product recommen-
dation service, tested Keurig machines and found that they “brewed watery,
358 Key Changes
flavorless coffee that paled against every other kind of coffee we’ve made at
home. At its best, Keurig coffee tastes like diner coffee. At its worst, it tastes
like hot brown water.” One of Wirecutter’s testers found that it “tastes like an
ashtray.”vi
Yet adoption of single-cup coffee machines rose steadily in tandem with
consumption of gourmet coffee in the home. Coffee industry market research
has shown that gourmet coffee consumption increased from its 25 percent
share in 2010 to 61 percent by 2019. And throughout that period, ownership
of single-cup machines—mostly Keurig units—increased to 42 percent in
2019 while drip coffee maker ownership slowly declined (though it remains
the most widely owned type).vii
Single-serving coffee has begun to wane in popularity as other coffee
trends, such as cold brew, take over. But many people still drink it, whether at
the office or at home. And although most third-wave roasters do not sell their
coffee in K-Cups, large producers of gourmet coffee like Starbucks and Peet’s
do. (And to address the environmental harm, Keurig now offers refillable
K-Cups, and a few coffee producers sell compostable cups.) In other words,
many consumers buy “quality” coffee despite the fact that the Keurig system
does not do it justice—just as many music fans have been perfectly happy to
listen to pristinely produced music tracks as easily downloadable, mediocre-
sounding MP3s. Once again, convenience beats quality.
As new formats emerged and dominated the record business, they paved
the way for strategic decisions that maximized revenues and profits. The
labels offered fewer singles to encourage increased purchases of more profit-
able CDs. Dividing streaming music revenues based on the pro rata share of
plays led to huge deals to acquire the catalogs of iconic legacy artists and to
signing new artists who already have millions of engaged followers on social
platforms.
The most consistent theme, however, has been the industry’s willingness
to place significant bets on new products or business models only when its
Coda 359
the vast majority of albums made available for download, but every song on
those albums was available as a single for the modest price of 99 cents.
The download business enabled by Apple’s iTunes store and the iPod be-
came a significant contributor to the industry’s bottom line. But it never
generated enough cash to replace the revenues from CDs as that product
continued its decline. Once again, financial pressures provided the impetus
for the industry to take a fundamentally different approach when Spotify
launched in the United States in 2011. A free tier that provided unlimited
access to the entire music catalog and served as a marketing funnel for paid
subscriptions was no longer a nonstarter in licensing negotiations. More than
a decade later, the streaming business, powered by a variety of global players,
has brought top-line growth back to the music business. The valuations of
the major record labels have reached heights that few, if any, would have
predicted at the turn of the twenty-first century.
Jim Griffin, the technologist who persuaded Geffen Records to make an
Aerosmith track available via a digital download in 1994, characterizes this
as “Tarzan Economics.”2 Businesses need to let go of their current mech-
anism for making money, that is, the “old vine,” at the right time and grab
a “new vine.” When a business relies on returns from its current model and
the revenues from the new approach are insignificant with an unproven po-
tential for profits, that shift requires a leap of faith that many conservative
executives are loath to make (particularly when their bonuses depend on
current quarter results). In the case of new music formats and the business
models that accompany them, that old vine needs to be fraying visibly to
make the case to let it go.
even rent used books by simply entering your school and course number.
Unauthorized copies became available online in PDF format if you knew
where to look. A growing movement of academics started creating Open
Educational Resources (OER)–digital text materials that were free and
published under Creative Commons licenses3 so that no one could restrict
or profit from them. And more and more professors began to adopt free or
cheap nontextbook curriculum materials such as websites and trade books
(books that sell at retail bookstores).
At first, publishers tried to stave off these developments without altering
the basic model. They tried getting textbook authors to release new editions
of their textbooks more frequently so that the used older editions wouldn’t be
as desirable. The major publishers formed a joint venture called CourseSmart
to publish e-book versions of their texts that students could “rent” for less
money than buying print textbooks, but the products were mostly “shovel-
ware”4 that not many students wanted to use. Publishers attempted, in vain,
to get copyright law changed to restrict reselling books. Mainly they just
kept raising prices. As one example, the classic textbook Economics by Paul
Samuelson dates back to 1948. It sold for $10 in 1969. Today, the book—now
in its 19th edition—sells for more than $200. Even when adjusted for infla-
tion, that’s a threefold price increase.
As the effectiveness of these incremental strategies waned, the industry
contracted through consolidation. Then in 2013, Cengage, one of the major
textbook publishers, declared bankruptcy. That was the industry’s “Tarzan
moment.” Cengage emerged from chapter 11 with a new model called
Cengage Unlimited that is analogous to studio-run streaming video services
such as Disney +and NBCUniversal’s Peacock: digital materials at an all-
you-can-read semester or annual subscription price. This was a risky move
that the publisher wouldn’t have attempted during the good (or even just-
okay) times. It required major investment in technology and processes to
convert print-oriented texts to “digital-first” materials that work well online.
It also meant having to contend with authors who hadn’t given Cengage per-
mission to publish their materials in this way; a few of them filed lawsuits,
which settled.
Early indications are that Cengage Unlimited is successful, attracting
millions of students who save money while helping Cengage’s bottom line;
and other publishers—such as Pearson, the largest educational publisher—
have launched similar plans. And the move to these digital-first models has
another advantage: it’s much easier for authors and publishers to keep online
362 Key Changes
Sergeant Pepper’s Lonely Hearts Club Band by the Beatles a dozen years later.
Though more music could be placed on each side of the LP, there were still
technical limitations that affected the creative choices. A particularly loud
song included at the end of the side increased the likelihood that the needle
would jump out of the groove. And the final track on the “A” side needed
to be sufficiently engaging to convince the listener that getting out of their
chair to flip the record was a worthwhile endeavor. These constraints made
the sequencing of the songs on the LP a critical part of the creative process.
Napster disaggregated the album into its constituent tracks for file-sharers,
and Apple’s iTunes store did the same for music purchasers. The one big hit
song was no longer the driver to purchase the entire album, much to the det-
riment of the music labels’ bottom lines. But it did mean that artists like Flo
Rida could dominate the singles charts with far greater frequency than the
album charts. And as streaming became the main source of revenues, artists
began to craft releases with many more tracks of shorter lengths, each with
the hook upfront to minimize song skipping, and to take advantage of calcu-
lating royalties using a pro-rata share of total plays. Throughout the history of
recorded music, artists have altered their output to conform to each format’s
constraints and show off their creativity in the best way while maximizing
potential revenues.
The press probably wrote more stories explaining how the Internet caused
the music industry’s precipitous fall than about any other business. However,
the writers of those stories were part of another industry in the throes of a
crisis caused by disruptive technologies.
The newspaper business was built on advertising. The revenues from sel-
ling papers defrayed some of the costs of running a newsroom, but they were
far from sufficient. The daily editions that summarized the happenings in pol-
itics, the arts, and sports aggregated audiences of readers that were valuable
to advertisers. The papers that served an individual metropolitan area relied
on the ad revenues from that area’s department stores, electronics retailers,
or record shops, promoting their products and latest sale prices, as well as
on classified ads for jobs, apartments, and a myriad of other goods and serv-
ices. A few newspapers, such as the New York Times, Washington Post, and
364 Key Changes
USA Today attracted readers across the country and were able to supplement
those local advertisers with a broader selection of national brands.
As the online audience grew, the newspaper companies built web-based
versions of their newspapers, but they treated them as complements to their
primary print business. Like the music companies, they viewed their digital
businesses as sidelines rather than making them their primary focus. Like
the textbook publishers we mentioned earlier in this chapter, they simply
served up shovelware: HTML versions of articles on the same daily pub-
lishing schedule that had served them well for decades. They gave the news
away for free in the hope of attracting enough eyeballs to generate revenues
from the digital ads they displayed.
But the technology of the web provided capabilities that enabled alter-
native ways of presenting and monetizing the content that had previously
been confined to print. Craigslist, begun as an email list in San Francisco
in the mid-1990s, leveraged its initial success by replicating sites for other
cities with minimal costs. Its text format was so simple, and the interface so
easy to use, that their advertisers could post ads themselves, which kept op-
erating costs extremely low. By capitalizing on these web capabilities and
giving its service away for free to advertisers,5 Craigslist became a behemoth
that served almost 50 million unique visitors with more than 80 million ads
each month across the United States alone, decimating newspapers’ classified
revenues.
Facebook’s social network and Google’s search interface, both supported
by highly efficient online advertising, provided ways to connect with enor-
mous audiences with appetites for information. That enabled sites that
catered to specific niche audiences, such as inside-the-Beltway politics
filtered through a specific ideological viewpoint, or celebrity fashion, or
aggregated movie reviews, to draw enough geographically dispersed users
and sufficient investment to fund their growth (if not profits).
These new players in the news business were not wedded to tradition;
they changed the way they created content for the web. They abandoned the
multiple reviews and editing passes required by newspapers as well as the
entire concept of a daily schedule for setting type and printing presses. The
story went up on the website as soon as it was ready, and then it was updated
as new facts emerged or errors needing correction were found. Multiple
headlines were published on the same story, and whichever one produced
the most traffic was quickly adopted going forward. The most provocative
headlines were often the most effective in attracting traffic even when they
Coda 365
did not accurately represent the content of the story, giving rise to the term
“clickbait.”
News sites would emphasize pictures or videos as the main way to convey
information instead of relying primarily on lengthy text stories with a few
photos. Instead of stories built around a narrative, a collection of pithy
sentences illustrated with photos related to a common theme drew many
more viewers than wonky pieces about tax code changes. These so-called
“listicles,” perfected by sites such as Buzzfeed, featured headlines with num-
bers like “12 Extremely Disappointing Facts about Popular Music.”ix All of
these techniques were part of the search engine optimization (SEO) toolkit to
help boost the stories to the top of Google’s search.
Though creating content to attract clicks was a priority for sites like
Buzzfeed and Huffington Post, it is worth noting that both of them
supplemented these efforts with quality journalism; both have received
Pulitzer Prizes for investigative reporting.
While many local newspapers cut back significantly on their staff of
journalists or closed up entirely, the existing national papers used their larger
addressable markets to adopt story-telling and money-making approaches
more suitable to the Internet era. Rather than simply giving the news away
and relying on advertising revenues to monetize that traffic, they have
implemented paywalls that require paid subscriptions, typically after a set
number of articles are read.
The New York Times’s implementation of a paywall led to significant
growth in paid subscribers, reaching more than 9 million by April 2022 with
overall revenues returning to growth after years of declines. It has also been
able to make more varied offers that would have been impossible in the days
of print. By creating separate subscriptions for crosswords and recipes, the
Times has been able to monetize niche audiences by allowing its readers to
purchase more limited access at a lower price than for the entire paper.
The new technology not only enables news organizations to provide access
to their traditional content online, but it has also allowed them to expand
the types of stories they tell. Podcasts, such as The Daily from the New York
Times, provide journalists the ability to go deeper on the stories they present
in print and online. In 2017, its first year of operation, The Daily grew to al-
most 4 million listeners each day; and it has consistently been ranked among
the most popular podcasts.x Newspapers have also hired data journalists who
are experts in visualizing complex stories with charts and tables. In print, they
can produce interesting but static charts and graphs. That same information
366 Key Changes
placed online can be interactive as the user selects and reconfigures the in-
formation with a few clicks. For example, rather than providing national or
state views of the latest COVID-19 statistics alone, these data visualizations
enable online readers to select the county they want to see, the statistics they
are most interested in, and the time periods relevant to them. Not only have
Internet technologies enabled news organizations to offer access using dif-
ferent configurations and business models, but they also allowed the nature
of the content offered to change.
Innovations in the music industry have often led to structural changes be-
cause of quirks or loopholes in copyright law. The speed of technology
changes has increased constantly since the early twentieth century while the
speed of changes to the law has stayed much the same; and the gap between
the two has grown.
The first significant manifestation of that gap was the lack of copyright in
sound recordings, which we discussed in Chapter 4. That gap wasn’t closed
until Congress passed the Sound Recording Act of 1971, which established
federal copyright in sound recordings and likely contributed to record labels
surpassing music publishers as the dominant generators of music industry
revenue in the ensuing decade.
Yet the most longstanding manifestation of the gap between technology
changes and the law has been in radio. As we saw in Chapter 3, when radio
arose as a popular source of entertainment in the 1930s, regulations lim-
ited the ability of radio stations to play recorded music on the air. Those
restrictions faded away when television emerged in the 1940s and live
entertainers began fleeing radio for the new medium. From then on, laws
and regulations had the opposite effect: they enabled playing recorded music
on the air to become a highly lucrative and low-cost business.
Radio stations did not have to license the records they played for
public performance, because there was no copyright protection for sound
recordings. They did have to license the musical compositions being
Coda 367
performed, but the law did not require stations to seek permission from
music publishers in advance, and there was a competitive market for com-
position performance royalties after the establishment of BMI in 1939. But
even the Sound Recording Act of 1971 did not extend copyright protection
to radio broadcasts.
Although Congress established sound recording performance royalty
schemes for digital radio in the 1990s, to this day AM/FM broadcast radio
does not have to pay for the use of records on the air. The National Association
of Broadcasters’ success in fending off repeated attempts to establish perfor-
mance royalties on sound recordings for AM/FM radio broadcasters has
benefited the radio industry for nearly a century.
Copyright law’s limited ability to respond to technological innovations af-
fected the industry in a different way in the 1970s, when the convenience
of tape cassettes motivated innovations that turned a format originally in-
tended for voice recording into a high-fidelity medium for music. By the late
1970s, consumers could buy moderately priced tape decks and blank tapes
and make copies of LPs that could sound almost identical to the originals.
Copyright law forbade activities such as copying your friend’s albums so that
you didn’t have to buy them, even as its applicability to other activities, such
as taping your own albums so that you could play them in the car, was not
entirely clear.
But the law wasn’t effective as a tool against copying albums onto cassettes.
As home taping rose and record sales began to decrease in 1979, the in-
dustry made a series of attempts to ameliorate its effects, including pushing
Congress to enact a levy on blank tapes that would go to copyright owners.
This failed, as did the media industry’s attempt, in the Universal v. Sony
(“Betamax”) Supreme Court case, to find home taping of broadcasts to be a
violation of copyright law.
As personal computer technology and online services began to dis-
rupt the music industry in the 1990s, music industry lobbying forces tried
more aggressively to ensure that copyright law kept up with technological
developments. But even when the industry got Congress to pass laws that
extended copyright protection in the Internet environment, technologists
found ways around those laws.
The furthest-reaching example of this is the Digital Millennium Copyright
Act (DMCA), which was signed into law in 1998. Its purpose, among other
things, was to give the emerging online services at that time space to inno-
vate without having to be responsible for their users’ potential copyright
368 Key Changes
Many Internet startups live by the adage: “Ask for forgiveness; not permis-
sion.” Launching services designed to exploit legal or regulatory loopholes
has become an essential element of Silicon Valley culture. In music, that
meant skirting the precepts of copyright law, but for others seeking to dis-
rupt the status quo, it meant ignoring the constraints or financial impacts of
the tax code or labor laws. The so-called “gig economy” relies on the ability
to deploy software applications that match revenue opportunities with
providers who are treated as independent contractors rather than employees.
Such businesses not only avoid obligations around salary, benefits, and taxes
but also often require those contractors to use their own assets—whether
automobiles or rooms in their home—to satisfy the needs of customers.
There may be no better example of an upstart rule-breaker confronting a
highly regulated industry than Uber and taxicabs.xi Innovation has played
a role throughout the history of this transportation business. In fact, it gave
rise to the very term taxicab. The hansom cab was a light horse-drawn car-
riage; its name was a shortened version of the French term “cabriolet.” When
a carriage was equipped with a “taximeter” that measured time and distance
traveled, the “taxicab” was born. Overcharging for rides led to the imposition
of a licensing structure for drivers. The pollution caused by cars proved less
distasteful than what horses left behind, and automobiles soon became the
preferred vehicle for taxis. The Great Depression led numerous unemployed
people to seek work as taxi drivers, making it increasingly difficult for anyone
to make a decent wage. To constrain the number of taxis permitted to pick
up passengers, major cities, starting with New York, instituted a medallion
system to authorize cars in addition to licensing drivers.
Over time, cities kept a tight rein on minting additional medallions. This
drove enormous increases in prices.7 Larger companies amassed the re-
sources to own numerous medallions and then hired drivers who had to
earn enough to cover their “nut”—the substantial upfront fees for each shift.
The bureaucracy frustrated upward mobility for drivers and constrained the
supply for passengers, creating the perfect conditions for a free-wheeling8
entrant willing to defy the rules.
In 2009, Internet entrepreneur Garrett Camp, frustrated by the high
fees to hire a private car, came up with the idea for a “ride-sharing” appli-
cation. He was soon joined by Travis Kalanick, a serial entrepreneur who
had built and sold peer-to-peer file-sharing and content delivery network
370 Key Changes
startups. Kalanick became the public face of the new company and its aggres-
sive take-no-prisoners ethos. “Ubercab” launched in San Francisco without
any approvals from regulatory or government bodies. Despite offering only
luxury car rides priced higher than taxi fares at first, the convenience and
value of the service fueled rapid growth. Ubercab expanded to other cities
and added lower-cost options by allowing individuals to use their own
cars, subject to Uber’s background check, insurance policies, and vehicle
standards.
Cabs and their drivers had to follow the far more rigorous rules and
licensing requirements that Public Utilities Commissions enforced in each
city. And those institutions, prompted by complaints from the incumbents,
pressured Ubercab (and its “fast-follower” competitor Lyft) to live by those
same rules with a campaign of fines and tickets for the company and their
drivers. Some cities, such as Portland, Oregon, simply declared such services
illegal. Taxi drivers filed class action suits for unfairly taking their fares.
Uber was not deterred by any of these actions. Claiming that it was a tech-
nology company not subject to conventional taxi regulations, it rebranded
as “Uber.” It used some of its substantial venture capital cash to pay the fines.
It developed software called Greyball that used data to detect when gov-
ernment officials were using the app to catch Uber and its drivers in unau-
thorized activities: actual drivers quickly canceled the officials’ rides, while
Greyball showed vehicles approaching on their phone displays that simply
did not exist. If there were no rides, there were no violations to cite.
Lyft gained momentum by enticing riders with lower prices and allowing
drivers to align with both services. Uber responded by creating software to
create fake Lyft riders to gather data on the availability of their competitor’s
cars and drivers to improve their own service. In addition, the data it gathered
surreptitiously identified drivers working for both firms who could then be
targeted with special bonuses to become exclusive to Uber. These actions put
Uber at legal risk for pursuing unfair business practices.
Expanding the number of Uber drivers increased the availability of rides,
but it also created issues around passenger safety, as female passengers were
sometimes subject to harassment or worse. In 2014, Uber added a surcharge
of $1 on every trip as a “Safe Rides Fee.” However, the cash was not used to
improve safety in any way. Instead, in yet another case of unsavory practices,
the half- billion dollars collected were simply added to the company’s
bottom line.
Coda 371
Uber was, for a time, the most popular download in Apple’s app store.
As part of its extensive data collection, Uber’s software enabled it to iden-
tify individual iPhones even after a user deleted their Uber app, a practice
that violated Apple’s policies. Apple refused to approve an updated version
of the Uber app unless it abandoned its intrusive practices. Not wanting to
abandon the edge provided by the data, Uber crafted a devious solution to the
dilemma: it retained the secret data collection code but created a “geofence”
around Apple’s headquarters in Cupertino. Within that area, any Apple em-
ployee running the Uber app would observe strict adherence to the rules
with no problematic data collected. But if you ran the app outside of that re-
gion, all bets (and rules) were off. Apple eventually discovered the ruse, and
CEO Tim Cook threatened Kalanick with Uber’s complete removal from the
app store. Uber complied, although the entire kerfuffle proved to be nothing
more than a slap on the wrist.
One of the most successful elements of Uber’s pushback on govern-
mental efforts to rein it in was in the court of public opinion. It mobilized
its riders, who found the service superior to conventional taxis, to put pres-
sure on politicians. Uber obtained thousands of signatures on petitions that
it could deliver to the steps of City Hall, with drivers and riders cheering it
on. Ultimately, Uber’s tactics prevailed, and it was able to expand its services
in cities all over the world.
Many drivers appreciated the flexibility of scheduling when driving for
Uber, but their contributions to the company’s growth were not rewarded.
Treating them as contractors who drove their own vehicles gave Uber the
benefits of huge revenues without incurring the costs of salaries, benefits,
or car repairs. In the face of mounting driver complaints about compensa-
tion, Kalanick famously said that Uber would be better off financially when it
could replace the driver completely with self-driving vehicles.
Playing fast and loose with the rules did have consequences. The company’s
reputation took significant hits for its underhanded maneuvers (and its mi-
sogynistic workplace culture). When Uber went public in 2019, it was valued
at over $80 billion, worth more than GM and Ford combined. Still, by 2022,
even with the public exposure of all its questionable practices, the company’s
enterprise value declined only about five percent. Kalanick himself took a
much greater hit as his investors forced him to resign as CEO for the com-
pany that he had propelled to market dominance. His bank account did not
suffer, however, as his shares were worth billions after Uber’s IPO.
372 Key Changes
Music listeners have repeatedly shown that popular tastes evolve over time
and that they have a willingness, if not a desire, to connect with new genres
and artists. These shifts have become even more pronounced since World
War II and the rise of teenagers as a distinct demographic segment with
significant purchasing power. Each new wave of teens want music and art-
ists that represent their own particular generational concerns–and if their
parents found that music objectionable, so much the better.
These new musical styles and genres were repeatedly brought into the
world by African American artists. But the power structure of the industry
held the mistaken and biased view that the majority of their audience simply
would not accept these artists, so they deliberately limited the aperture for
this material to reach the broader market. They chose to keep the Black art-
ists in the background and placed their melodies and words in the hands
and voices of white performers to improve the perceived acceptability to
white fans.
The earliest recordings continued the minstrelsy practices prevalent in live
performances of the day and offered “race music” with white artists using
exaggerated Black speech patterns and demeaning language. Jazz evolved
out of earlier forms of African American music and was refined by Black
musicians in New Orleans. But the first recording of the genre featured
the Original Dixieland Jass Band,9 which was made up entirely of white
musicians. Paul Whiteman became known as the “King of Jazz,” while the
arrangements for his orchestra of white musicians toned down the more im-
provisational elements of the genre to broaden its “mainstream” (i.e., white)
appeal. Entrepreneurs created independent companies such as Okeh Records
Coda 373
that, unlike the established companies, were willing to feature artists such as
Mamie Smith and Louis Armstrong. Jazz performed by these and other art-
ists was marketed to both Black and white audiences, and the establishment’s
limited viewpoint proved unfounded. Fans of all colors bought the records,
and by 1928, Columbia Records purchased Okeh and its catalog.
Through the 1940s and into the 1950s, radio segregated music from
African American artists on separate stations. The industry periodical Radio
& Records did not even have a chart to report on the popularity of music from
Black artists at that time, rendering it invisible to readers.
Early rock & roll music was widely viewed as a corrupting influence on
young people, and the race of the founders of the genre contributed to that
impression for many in white America. One of those founding fathers of
rock, Fats Domino, released “Ain’t It a Shame,” a song he cowrote with Dave
Bartholomew, in April 1955. Within a month, a somewhat sanitized version
of the same song recorded by Pat Boone, now retitled as the more familiar
“Ain’t That a Shame,” followed. One would be hard-pressed to find a less ob-
jectionable musical artist to white households than Pat Boone. His co-opted
version gained national attention for the song (and publishing royalties for
Domino). Yet fans proved to be more accepting than many expected, and
Fats Domino’s version outsold Boone’s.
At the start of the music video era, MTV resisted playing videos by Black
artists because they did not think they would appeal to its core audience of
predominantly white rock fans. Michael Jackson’s “Billie Jean” gave the lie to
that perception after some arm-twisting by CBS Records resulted in a change
of policy. The popularity of Jackson’s videos and those from other Black art-
ists demonstrated once again that fans’ musical tastes were broader and more
diverse than expected.
A few years later, the mainstream record business was uninter-
ested in signing hip-hop artists when the genre emerged in urban Black
neighborhoods. The view that this was not “real music” even held them back
from offering “white-washed” versions. In this case, the technology available
provided an alternative path to reach fans. The availability of recordable tape
cassettes and, later, CDs, gave those artists a way, albeit on a limited scale, to
manufacture and to distribute their creative works. Soon independent labels
focusing on rap music provided the capital and infrastructure to expand. As
the popularity of the new genre grew, the major labels did adapt and brought
these smaller hip-hop labels into their corporate structures. However, the
genre still received short shrift in terms of marketing and promotion until
374 Key Changes
the birth of the Soundscan era. Having accurate sales data showed that rap
(and country) were far more popular than the industry believed. Fans across
the country—both Black and white—demonstrated their affinity for new
genres born in the African American community despite the hurdles placed
in the way.
indicative of the sport’s approach to race that, in the early part of his career,
he was referred to as “The Great White Hope” of the NBA.10
By the late 1990s, hip-hop’s cultural influence and importance to the music
business was well-established. Around that same time, Allen Iverson entered
the NBA with the basketball abilities to dominate any game he played de-
spite being only six feet tall. He was selected number one in the NBA draft
by the Philadelphia 76ers in 1996, and with his numerous tattoos and corn-
rows he resembled rappers in music videos more than superstar players of
the day such as Michael Jordan. “Gangsta rap,” which represented the milieu
of street gangs and culture, was seen by some as promoting criminality, and
Iverson’s personal history gave people more evidence to paint him with the
same brush: he had served time in jail while in high school over an alterca-
tion at a bowling alley, although the conviction was later reversed. On the
basketball court, Iverson did not adhere to the unwritten rules that required
deference to players like Michael Jordan. Iverson seemed to embody the very
image that white ownership wanted to avoid for its majority-Black league.
When he appeared on the cover of the NBA’s magazine, they airbrushed the
photo to remove his tattoos.xii
Another supremely talented player also entered the league in 1996, but in
contrast to Iverson, Kobe Bryant came with a resume containing all of the
establishment’s preferred attributes. Bryant had decidedly not grown up
“in the hood.” He was raised in the affluent area of the Philadelphia suburbs
known as the Main Line, and in Italy where his father Joe played profession-
ally after leaving the NBA. Not many basketball prospects from the inner city
spoke Italian fluently and played soccer. Despite strong grades and excellent
SAT scores, Bryant decided to skip college and made the leap to the NBA di-
rectly from high school.
Dr. Todd Boyd is the Katherine and Frank Price Endowed Chair for the
Study of Race and Popular Culture and Professor of Cinema and Media
Studies at USC. He wrote that “Kobe Bryant has often functioned as the
league’s de facto White Man, in that his upper-middle-class status is more
easily assimilated in the game’s overall fabric than the hip hop-inspired nar-
rative that Allen Iverson embodies.”xiii
Though the league might have preferred Bryant as a more “acceptable” star,
Iverson’s success on the court made him a fan favorite. The hip-hop aesthetic
that Iverson represented became an intrinsic part of the league. Rappers now
headline the performances at the NBA All Star Game, and hip-hop music is
played for the dance teams as entertainment in every arena. For a time, Jay-Z
376 Key Changes
owned a small share of the Brooklyn Nets.11 Despite the fears of white owner-
ship, this shift has driven the league to greater heights of popularity with both
white and Black fans.
The music industry created retail channels in the late 1920s when record
labels standardized on the 78-rpm disc. Once that happened, it was inevi-
table that independent stores would appear that sold records from all labels.
As we saw in Chapter 4, this in turn led to the proliferation of retail channels
for physical music products throughout the latter half of the twentieth cen-
tury, with all of their complexity: indie record stores, big discount chains,
distributors, rack jobbers, and so on. It also led to the swift exit of record
companies from the hardware business.
The reason why record labels adopted a standard format was simple: in-
crease the size of the overall market by making it easier for consumers to buy
more records. This is an extremely common and proven strategy in many
markets, from tires to mattresses to plumbing fixtures. Yet in the vast ma-
jority of cases, standardization comes with a tradeoff: suppliers of products
into standards-based markets lose control over retail channels. Goodyear,
Serta, and Kohler don’t control Mavis Discount Tire, Mattress Firm, or
Home Depot.
Control over retail channels amid format standardization became a per-
ennial struggle for the music industry; the history of the industry is one of
record labels’ repeated and mostly futile attempts to take back elements of
channel control. Antitrust law constrained record labels’ ability to control re-
tail channels in significant ways, so they resorted to other tactics that worked
around the edges.
One of these tactics was selling directly to consumers through record
clubs. This was moderately successful; but the labels had to cross-license titles
to each other’s clubs to compete effectively, especially against discounters like
Sam Goody that had also built formidable mail-order businesses. Otherwise,
the tactics that labels used to influence retail channels cost them money, such
Coda 377
Hollywood studios and television networks were able to watch the record
labels suffer through the Napster era at a relatively safe distance: Internet ac-
cess speeds and storage media in the late 1990s were not fast or large enough
to handle digital video. (For example, it took roughly 10 hours to download
a movie over a 56 kbps dialup line, and the resulting file would fill up the
hard drive of a typical PC of that era.) So while they knew that video on the
Internet was coming, they had some time to plan.
The major studios and networks tried to be more aggressive than the
record labels in taking control of online distribution channels. The studios
had bitter experience with the loss of channel control in the late 1940s, when
the Supreme Court’s decision in United States v. Paramount Pictures forced
them to divest the movie theaters they owned. And the television networks
were stymied in the 1950s by the FCC’s regulations that limited the number
of broadcast stations that a single company could own.
Hollywood’s first attempt to control online channels was a digital down-
load service called MovieLink, which was a joint venture of most of the major
studios. MovieLink launched in late 2002, about a year after the major labels
launched MusicNet and pressplay. It was the first Internet video service to
offer licensed content from all of the major studios.14
MovieLink was never more than a moderate success, and its competition
increased over time. By the mid-2000s, the video rental chain Blockbuster
had expanded into online video, as did a few startups and consumer elec-
tronics retailers. The studios licensed their content to those entities, and
for a little while it looked like services such as Blockbuster Online and Best
Buy CinemaNow might be the future of video with ample major-studio
catalogs of content for sale. Yet online movies and TV were slow to catch
on. Downloading was still a clunky process, and few people wanted to watch
movies on PCs. The resulting business that these services did was minimal.
The mood in Hollywood shifted in late 2005 when Apple launched video-
capable iPods and expanded the iTunes Store to sell movies and TV shows.
The studios saw that Apple was dominating the digital music world of the
time with its tightly integrated, easy-to-use products, so they attempted to
prevent it from doing the same thing in video. Their response was a tech-
nology called UltraViolet, the brainchild of Sony Pictures executive Mitch
Singer.15
380 Key Changes
them. And the Holy Grail of channel control continues to elude Hollywood
just as it continues to elude the music industry.
Conclusion
The history of the recorded music experience is certainly a long and winding
road. We’ve progressed from the indentations in a tinfoil cylinder to
terabytes of zeros and ones stored in the cloud. Instead of placing a needle
ever so carefully on the right spot on a record, music is now accessible by
the touch of a finger on a powerful handheld computer or a few spoken
words. Understanding how each of the advances in technology and formats
came about is essential to understanding today’s music industry. In his
2005 commencement address at Stanford University, Steve Jobs said, “You
can’t connect the dots looking forward; you can only connect them looking
backwards.” We think that we’ve done more than simply recite the history
of recorded music formats. We’ve provided a framework—the 6Cs—that
connects the dots and provides both a consistent way to examine that history
and provides perspective about the changes yet to come.
There are two incontrovertible lessons that we can take away from our ex-
amination of the industry from the earliest phonographs to streaming serv-
ices powered by AI algorithms. First, it is only a matter of time before some
yet-to-be-invented technology changes the way that we experience recorded
music again. And second, regardless of that experience, billions of people
will be listening. The artist Moby said, “A great song is a great song, whether
it’s on vinyl or CD or cassette or reel to reel or mp3. Then again, that might be
an overly optimistic view, but I do think that great music will transcend the
medium in which it is delivered.” We agree.
Afterword
Unchained Melody—Righteous Brothers
What’s the next technology that will disrupt the music industry? As we write
this in 2022, we’re pretty sure of at least one word in one answer to that ques-
tion: blockchain.
Blockchain adherents claim that the technology will disrupt many
industries as profoundly as the Internet itself did, starting back in the 1990s.
The amount of money, hype, and talent being poured into blockchain tech-
nology today has not been seen since the first Internet Bubble of the late
1990s; that alone virtually guarantees that the effects that the technology will
have on various facets of life will be profound.
The music industry has been engaging with blockchain and related
technologies for the past several years. Although the technology’s impact
isn’t likely to be truly clear for a few more years, the contours of its likely
effects on the structure of the industry are starting to come into view. We can
suggest some directions that it will take, as startups, artists, and others exper-
iment with it, and as the public starts to catch on.
the first cryptocurrency. People involved in the field soon understood that
the technology had many applications beyond cryptocurrency. Bitcoin uses
one blockchain; just as there can be arbitrarily many databases (or database
companies) in the world, there can also be arbitrarily many blockchains.
Several ancillary technologies have been invented to enable certain types
of blockchain applications; these, together with blockchain technology it-
self, are collectively known as “Web 3.0” or simply “Web3.” The term “Web3”
reflects the notion that these technologies are bringing about a third wave in
the development of the Internet, where “Web 1” was the original “Wild West”
Internet of the late 1990s through early 2000s and “Web 2” represents the
current environment dominated by tech behemoths like Alphabet (Google)
and Meta (Facebook, Instagram).
The most important of these Web3 technologies for these purposes is
smart contracts. A smart contract is a piece of code that runs on every in-
stance of a blockchain whenever a new record is added to it. Smart contracts
are so named because they are often used to implement rules of an agree-
ment (contract) among parties in transactions. Certain blockchains, such
as Ethereum, Solana, and Cardano, support smart contracts and are used in
music applications.
Blockchain applications for music began to emerge in the mid-2010s.
They have fallen into two categories, which we can refer to as B2B (business-
to-business) and B2C (business-to-consumer).i
set by law. The PROs representing those rights-holders should get perfor-
mance royalties according to the blanket licenses that the music service
negotiated with each PRO. And so on. The idea is that each time a song is
played, the music service adds a record to a blockchain that indicates the
identity of the song, along with other data indicating the type of use,1 and a
series of smart contracts executes to pay the royalties.
Such a scheme would have substantial advantages over the current status
quo, where each music service has to maintain its own highly complex infra-
structure for making royalty payments. It would reduce operating costs, not
just for music services but for everyone involved. Because the vast majorities
of streaming services’ music catalogs are identical, this scheme would elim-
inate massive redundancies. And it would provide consistency, verifiability,
and transparency for royalty transactions.
Yet this example is idealistic and currently unattainable, for a number
of reasons. One is that blockchain technology at this time of writing is not
capable of handling the billions of smart contract transactions per day that
today’s streaming volumes would generate. The fact that every new transac-
tion must be propagated to every copy of a blockchain severely limits scala-
bility. For example, Ethereum, the most widely used blockchain that supports
smart contracts, can currently process between 10 and 20 transactions per
second.ii A future version has been promised that will lift that number to
100,000, but even that is still not quite enough to support royalty transaction
volumes for streaming music—which itself is growing over time.iii
Another reason is that the accuracy of transaction processing is only as
good as the accuracy of the data that goes into it—the “Garbage In, Garbage
Out” (GIGO) principle that data and technology people have long under-
stood. This creates a decentralized data governance problem that blockchain
technology by itself does not solve. Yet another reason is that it would require
an unprecedented amount of highly detailed cross-industry cooperation re-
quired to implement such a scheme.
At least some of these problems are solvable, whether through technolog-
ical innovation or through sheer effort and cooperation motivated by the
potential benefits. Several startup companies are working on these types of
B2B solutions, and pilot projects are under way. The Open Music Initiative,
a standards organization working on metadata models and protocols that
can be implemented on blockchains, created a blueprint for a blockchain-
based solution to the streaming mechanicals problem that we discussed
in Chapter 9—although that was sidelined with the passage of the Music
386 Key Changes
But these B2B applications are virtually unknown outside of the in-
dustry. Far more attention—and hype—go towards B2C applications, which
purport to bypass the entire industry infrastructure by using blockchain
technology to sell music directly to fans. The buzzword that has been
most frequently attached to this activity over the past couple of years is
Afterword 387
complete the transaction. That’s one reason why fees for minting and buying
NFTs have been quite high, often exceeding $100.
The process is similarly complex for sellers of NFTs, as it also is for art-
ists who may want to participate in DAOs or other blockchain-based direct
royalty schemes. The complexity is a real barrier to participation: it’s worth
noting that even today, SoundExchange—the sound recording PRO for dig-
ital radio that we saw back in Chapter 3—still holds millions of dollars in un-
collected royalties on its books because so many artists have not yet provided
even conventional payment details to it.
Of course, NFT minting and buying processes should get simpler over
time and may well be much easier and cheaper by the time you read this.
Service providers are already starting to appear that promise to make it easier
for NFT sellers and buyers alike.6
Water & Music’s data shows that the median price of a music NFT during
2021 fluctuated wildly and settled down in the $800–$900 range by the start
of 2022. But by the end of the following year, prices had plummeted: the me-
dian price dropped below $40 by the end of 2022.
Yet waning interest immediately after the March 2021 sales spike is
only part of the story. The developers behind Ethereum had been working
for years to solve the problem of the skyrocketing costs of computing
power and electricity required to validate transactions being added to the
blockchain—which has caused serious environmental harm. In September
2022, Ethereum switched to a type of algorithm for validating transactions
called “proof of stake” that requires far less computing power than the pre-
vious “proof of work” algorithm. The switch obviated the need for high gas
fees; this made it worthwhile for creators to sell NFTs at lower prices. It re-
mains to be seen where NFT prices will ultimately settle once fees become
less relevant, but certainly there are plenty of music-related NFTs available at
this time of writing for less than the typical price of vinyl.
This leads us to look at the music NFT market in another way as it moves
into its next phase of development. Figure A.1 shows a classic hype cycle
curve, including the number of artists minting NFTs and the total number
of NFT campaigns launched each month from mid-2021 through the end of
2022. It shows a classic hype cycle curve, including an upward trend in both
numbers that coincides with Ethereum’s September 2022 switch to proof-
of-stake. It’s likely that this trend will continue as the market converges on
pricing and release configurations for music NFTs that resonate with fans.
Once that happens, we will see whether music NFTs grow to become a major
390 Key Changes
400
350
NFT Campaigns
300
250
200
150
100
Artists
50
0
Ju 1
Au 1
Se 1
O 1
N 1
D 21
Ja 1
Fe 2
M 22
Ap 22
M 22
Ju 2
Ju 2
Au 2
Se 2
O 2
N 2
D 22
2
2
l-2
2
2
-2
-2
-2
2
l-2
2
2
-2
-2
n-
g-
p-
n-
b-
-
r-
n-
g-
p-
-
ct
ov
ec
ar
ay
ct
ov
ec
Ju
Figure A.1 2022 music NFT campaigns and artists. Water & Music.
Bill Rosenblatt.
music distribution format, like cassettes or CDs, or a small niche format that
fades away over time, like PlayTapes or MiniDiscs, or something else entirely.
The notion of “music NFTs” has evolved beyond simply records of pur-
chase of digital music files, although many of those exist. Artists are
exploring various possibilities. One is to mint multiple NFTs for a given
piece of music, akin to a limited-edition vinyl or CD release. This is a pop-
ular choice; the median number of copies of each music NFT at this time of
writing is ten.viii Rapper Tory Lanez took this to an extreme in August 2021
by minting a million NFTs of his album When It’s Dark at $1 each,ix and
then, after the million NFTs sold out in less than a minute, minting a second
batch of NFTs.x
Another possibility is to combine digital music with physical items, such as
actual limited-edition vinyl or merch, or with artist experiences. The rock band
Kings of Leon was an early NFT adopter, issuing a series of NFTs in March
Afterword 391
2021 around their album When You See Yourself that included limited-edition
vinyl and lifetime front-row seats at their live shows.xi Katy Perry auctioned
off an NFT package that included the Golden Lion, a huge stage prop that she
rode on in her 2015 Super Bowl halftime performance. (The package sold for
$600,000.) In early 2022, John Lennon’s son Julian auctioned off NFTs of some
of his dad’s guitars and items of clothing that John wore in Beatles movies,
which sold for prices in the $10,000–$20,000 range. A startup called POAP
(for Proof of Attendance Protocol) is building technology for POAPs (pro-
nounced “poh-app”), which are NFTs that commemorate a fan’s attendance at
a concert or other event, like a digital ticket stub on a blockchain. The company
announced a partnership with Warner Music Group in April 2022.xii
Yet other use cases lie in borrowing ideas from the larger and more active
world of visual art NFTs.7 So-called generative NFTs are those that use AI to
create dynamic content that emerges or changes in real time according to a
set of artistic parameters that could depend on the buyer as well as the cre-
ator. Generative techniques are making their way into music—along the lines
that we discussed in Chapter 11—specifically for NFTs. Some musical artists
are also selling NFTs for visual instead of audio content, such as Canadian
singer-songwriter Shawn Mendes, who has a line of NFTs for 2D avatars that
represent signature items such as his guitar and items of clothing.xiii And
many audio NFTs also come with visual artworks, often created by noted art-
ists and/or generative AI.
So what do fans actually get with NFTs, other than any physical items or
experiences? That’s a little hard to pin down. For NFTs on digital assets such
as music files, in most cases, the purchaser doesn’t get exclusive rights to the
assets; they are usually not restricted with DRM or some other technological
scheme, and they can be copied at will.8 (One advantage of generative NFTs
is that the content is not trivially easy to copy.) NFT skeptics often use the
phrase “bragging rights” to describe what you get. After all, NFTs are really
just publicly viewable entries in a database, as opposed to private purchase
records stored by a traditional retailer; and they are verifiable records stored
in your crypto wallet that you can show to anyone.
But there’s more to it than that. Proponents of NFTs say that they enable
artists to build more direct relationships with their fans. For artists, NFTs can
392 Key Changes
be more personal in nature than mere music files; and the combination of
high prices and complex buying processes means that, financial speculators
aside, NFT buyers are likely to be among the most devoted fans. Yet the same
has been said about various technologies—such as websites and peer-to-peer
file-sharing—since the Web 1 era. It’s not clear that NFTs will have more
than those technologies’ marginal effects on artist–fan relationships once
the novelty wears off, prices and complexity come down, and the number
of artists who use the technology increases. In addition, proponents’ claims
that NFTs eliminate intermediaries (such as record labels) seem dubious: the
complexities of NFT commerce and the growing number of NFTs virtually
guarantee that intermediaries will emerge, albeit most likely new and dif-
ferent ones.
The success of NFTs ultimately depends on their legitimacy as an
“ownership” mechanism, that is, on the public’s collective willingness
to perceive these records in cyberspace as evidence of “ownership” or
“collectability.” In Chapter 4 we discussed how the desire for ownership
has fueled the vinyl revival; it remains to be seen if NFTs satisfy that same
desire. For digital music files, it particularly depends on fans’ willingness
to see them as indicia of “ownership” that they do not believe exist for the
piles of encoded bits that they used to buy on iTunes. As NFT adherents
like to point out, today’s fiat currencies—no longer backed by gold or
other physical assets—also depend on societies’ collective willingness to
believe in their value.
the original publisher can have no involvement in further sales (or rentals,
loans, etc.).9
Both creators and consumers have been uncertain about what content
rights NFTs convey. Some consumers wonder why they should pay what
may be inflated prices for things to which they don’t have exclusive access
(whether by law or by technology). And some creators wonder why they
should trust this notion of “artificial scarcity” when it could reflect negatively
on their reputations if it should backfire or evaporate.
Meanwhile, the current ferment of experimentation with NFTs is
extending to their copyright law ramifications. A few people have engaged
in boundary-testing projects, such as creating NFTs on works that are in the
public domain or to which others own the rights. In perhaps the most brazen
such scheme for music to date, a startup called HitPiece launched briefly
in February 2022 with offers of unique NFTs of a large number of albums,
though it had licensed none of the intellectual property associated with
them—not the music, the trademarks, nor the cover images. After only a few
days, the RIAA threatened a lawsuit and it shut down.xiv
On the other hand, lawyers are also seeing opportunities in the NFT
world; they have started drafting NFT agreements that spell out rights for
artists, sellers, and buyers;10 and a growing number of entities are starting to
mint NFTs that include contracts with such terms. As the stakes grow higher
for NFTs, so does the need for clarification11 and solutions to these copyright
conundrums. Both will surely come in time, as will common understandings
of what is and isn’t permissible.12
Conclusion
Chapter 1
1. Another music video sits at no. 1, but it would be difficult to describe “Baby Shark” as
featuring major recording artists.
Chapter 2
The least expensive unit cost $600 (or over $9000 today), so it was far from a mass-
market product, though the technology established the foundation for the next gen-
eration of jukeboxes. VE-10-50 X. The Victor Victrola Page, http://www.victor-victr
ola.com/index.html. Accessed May 9, 2022.
14. In 1906, Sousa published a lengthy opinion piece in Appleton’s magazine titled “The
Menace of Mechanical Music,” detailing his concerns.
15. Some theaters tried playing a record of Caruso singing to accompany the film, but
keeping the two in sync was nearly impossible.
16. Though it took 70 years, Scott Joplin did eventually receive the popular acclaim he
warranted, as several of his compositions were featured in the film score for The Sting,
which climbed to the top of the Billboard Charts in 1974.
17. And, of course, the audience was not asked to distinguish between the artist per-
forming alone and the recording playing by itself in a “blind” listening test. Such tests
would come later on when the technology was more up to the job.
18. The purchase of automobiles made up half of that debt.
19. The same concept applies today to student discount plans for streaming services.
20. As we’ll see in the next chapter, the term “race records” was used in the industry for
decades until it was replaced by “rhythm & blues” in the 1950s.
21. The label was named to honor Elizabeth Greenfield who was born enslaved and be-
came the best-known black concert artist of the nineteenth century. She was called
the “Black Swan” in reference to the “Swedish Nightingale,” world-famous soprano
Jenny Lind.
22. This less-than-complimentary terminology lasted into the 1950s as well, when
“country and western music” and subsequently “country music” became the ac-
cepted name.
23. For example, Santana’s “Black Magic Woman” is arguably a better version than
Fleetwood Mac’s original. The same cannot be said for Pat Boone’s take on Little
Richard’s “Tutti Frutti.” Though in both cases, the covers outsold the originals.
Chapter 3
1. The NBC Symphony disbanded in 1954, but radio orchestras persist to this day in
countries with government funding for the arts. In Germany, Sweden, the United
Kingdom, Japan, and elsewhere, the orchestras of state-run broadcasters are among
the finest in their countries.
2. Savage Beast Technologies was founded in 2000 to bring the MGP to market, its name
inspired by the often-misquoted William Congreve line “Music has charms to soothe
the savage breast.” The technology was originally intended to assist retail stores in sel-
ling music before the company pivoted to streaming.
3. It wouldn’t have made much difference. The band’s follow-up album, The Long Run,
wouldn’t appear until September 1979, a full two years after Hotel California had
faded from the charts.
Notes 397
4. These services originally employed rooms full of people, often college students, who
would listen to radio stations and write down the songs they heard. Later on, acoustic
fingerprinting technology (see Chapter 8) took over this task.
5. Classic Rock is a radio format that began as an offshoot of AOR and spread to dozens
of stations nationwide in the mid-1980s through consultants such as Fred Jacobs of
Jacobs Media. But Classic Rock never got its own chart in R&R, presumably because it
featured older material rather than current releases.
6. Most trade magazines and tip sheets refused to publish R&B/soul charts or picks until
then because, as they claimed, that part of the industry was too payola-ridden for the
data to be trustworthy. Denisoff, Solid Gold, p. 260.
7. These regulations were quietly abandoned in 2009 by agreement between the NAB
and record labels.
8. China was on this list until it passed legislation providing for royalties for radio play
of recorded music, which went into effect in June 2021.
9. Plus Rep. Ted Deutch of Florida. Miami is another important center of music in-
dustry activity.
10. Some music publishers might disagree with this point; but as far as we know, it has not
been tested in court.
11. A few ad-hoc solutions are being advanced among private parties. For example,
Universal Music Publishing has offered a blanket license to its entire catalog to
Wondery, a podcast publisher that is now owned by Amazon.
12. Industry gossip held that Davis was “thrown under the bus” for reasons that remain
a mystery to this day. Sentiment was that Davis wouldn’t have been fired merely for
spending $94,000—on anything.
13. There was a sense in the industry that DJs on Black radio stations were more suscep-
tible to payola because they weren’t paid as well as their white counterparts.
14. Because it was hard for radio stations to buy shellac albums due to shellac shortages.
Denisoff, Solid Gold, 221.
15. The FCC has five Commissioners who vote on regulatory matters. Typically, three of
them (including the Chair) are of the party of the sitting president and the other two
are of the opposite party.
16. Even this is a simplified example, because (among other things) it omits the roles of
music publishers in PRO affiliations and splits. There are also some cases where the
record is released but the splits are not defined or are in dispute.
17. Except small stations with low revenues, which need only pay nominal annual flat fees
to SoundExchange.
18. Digital recording technology, which would make this type of process far less time-
consuming, came a few years later.
19. “Like a Rolling Stone” wasn’t the first song to be chopped into halves and pressed on
two sides of a 45. Ray Charles’s “What’d I Say” and the Isley Brothers’ “Shout” both got
the same treatment in 1959.
20. In the 2020 edition of Jacobs Media’s annual TechSurvey of radio listeners, re-
flecting listening habits before the COVID-19 pandemic forced everyone to stay
inside, 53 percent of respondents stated that all or most of their AM/FM listening
398 Notes
was in the car. For Gen Z and Millennials, the figure was 68 percent or more.
Techsurvey 2020 Results, Jacobs Media, https://jacobsmedia.com/techsurvey-2020-
results/. Accessed May 18, 2023.
21. In 1981, when the Rolling Stones released their Tattoo You album and embarked on a
massive world tour, WYSP in Philadelphia played Stones music almost half the time
during the run-up to the band’s two gigs at JFK Stadium.
Chapter 4
1. Recording engineers eventually figured out how to extend the capacities of vinyl sides
through severe audio compression that diminished sound quality. The likely longest
LP and 45 sides were both prog-rock tracks from the mid-1970s. For LPs, Todd
Rundgren’s side-long epic “A Treatise on Cosmic Fire” clocked in at over 35 minutes
on his 1975 album Initiation. For 45s, a 10 minute, 27 second live version of “Lunar
Sea” by the British prog-rock band Camel was the B-side of their single “Another
Night” in 1976.
2. The term “vinyl” likely arose as a short form of Vinylite, a name for polyvinyl chloride
that was trademarked by Carbide and Carbon Chemicals Corporation, a subsidiary
of Union Carbide.
3. Many old turntables had a fourth speed: 16. 16 2/3-rpm records were sometimes used
to distribute prerecorded programs to radio stations and for audiobook recordings
for the visually disabled. Old professional turntables with 16 speeds in college radio
stations have served as perennial sources of amusement for late-night college DJs.
4. The name EP has been used since then to describe various other formats that are be-
tween singles and LPs in length.
5. Rosenblatt’s first piece of “serious” audio equipment, a parental hand-me-down.
6. A fourth, Denon’s UD-4, was introduced in Japan, Europe, and the United Kingdom.
7. The CD-4 quad system’s requirement for phono cartridges with greater frequency re-
sponse and lower tracking force led to improvements in cartridge design that made
their way into standard stereo cartridges.
8. And of course there have been numerous innovations in turntable design over
the years.
9. Those return policies weren’t all bad for the labels. Occasionally they would manu-
facture and ship to retail many more records than they expected to sell, so that they
could get RIAA Gold or Platinum certifications—and executives could earn higher
bonuses—which were based on shipments, not sales, in the days before sales were
measured with some accuracy. This gave rise to the saying, “Ship Platinum, re-
turn Gold.”
10. See https://www.youtube.com/watch?v=SRjl_nIRSLk.
11. Ironically, Sony Music and BMG Music would end up merging—as would the two
record clubs—in the mid-2000s.
12. Amazon is likely the no. 3 used vinyl seller. Amazon does not report used vinyl sales.
Notes 399
13. Some overlap exists between indie retail sales and online sales, because indie record
stores sometimes sell their inventory through Discogs. This phenomenon increased
when the COVID-19 pandemic forced brick-and-mortar record stores to close for a
while—and coincided with a dramatic increase in RIAA-reported vinyl sales in 2021.
14. The right to resell digital files has been a hot topic of debate and the subject of a few
noteworthy lawsuits, such as Capitol Records v. ReDigi, where the record company
sued a startup that built a marketplace for “used” iTunes music files. ReDigi lost the
case and the subsequent appeal.
15. Record Store Day (RSD) is an annual event that celebrates independent record stores.
Hundreds of limited-edition albums are released specifically for RSD each year. Given
that RSD started in 2007, just after vinyl sales hit bottom, it’s often credited as a cata-
lyst for the vinyl revival.
16. Those were the three best-selling albums on Discogs in December 2017.
17. Only one of the ten best-selling albums of 2022 on Discogs was a classic title (David
Bowie’s 1977 album Low, at no. 9); the rest were new releases.
18. That the album was available in several versions, including four standard versions and
an exclusive version for Target stores, certainly helped boost this figure.
19. Outside of Western classical music, Indian sitar master Ravi Shankar released his first
LP, Three Ragas, on the British label His Master’s Voice (HMV) in 1956. “Raga Jog,”
over 28 minutes long, took up an entire side of the album. Ragas are largely impro-
vised and can last for hours when performed live.
20. The rotation speed of a record is constant, but the linear speed is proportional to the
square of the radius at any given point, so it decreases from the outer rim of a record
towards the center. That’s the reason why RCA’s 7-inch 45s were designed with a large
hole in the middle: so that the rotation speed toward the end of the side wouldn’t de-
crease enough to harm the sound quality too much.
21. Not to be confused with Hipgnosis Songs Fund, an investment fund for song rights
set up in 2018, which took its name from the design firm but has no relation-
ship to it.
22. The first artist-owned label was Reprise, which Frank Sinatra launched in 1960. It
wasn’t a vanity label but one that Sinatra started himself after he became dissatisfied
with Capitol. Reprise quickly became an important label with a big-name artist roster
and was acquired by Warner Bros. in 1963.
23. This is an estimate; total annual revenue figures for music publishing from that period
aren’t available.
24. The eminent jazz critic Nat Hentoff testified at a 1961 Congressional hearing that
“when [a jazz musician] performs on a record, his improvisation is what makes the
tune quite a new one” and therefore that there should be no question of the per-
forming artist’s creativity.
25. Possibly because the labels knew that public performance rights in sound recordings
were doomed to fail due to the radio industry’s influence, as we saw in Chapter 3.
26. Under that standard, collections of hit songs performed by cover bands with names
like The Realistics and Top of the Poppers, which were sold on late-night and UHF
television in the 1970s, would have infringed copyright. Hallmark, Columbia House,
and others put these albums out to capitalize on current hits by taking advantage of
400 Notes
compulsory mechanical licenses for the compositions and not having to pay the orig-
inal artists’ record labels.
27. Columbia eventually released most of the Dylan sessions—with The Band as backing
musicians—as The Basement Tapes in 1975. Decca/London responded to the Stones
bootleg by releasing an album of other performances from the same tour as Get Yer
Ya-Ya’s Out! in 1970.
28. ARMADA was eventually supplanted by the National Association of Independent
Record Distributors (NAIRD) and then by the American Association of Independent
Music (A2IM).
29. Lobbying around copyright law in the 1960s and 1970s was limited to industry “in-
side baseball” and was a much lower-key affair than it would become in the Internet
era. The Copyright Office routinely puts out requests for input related to its studies on
copyright law reform; in those days, the numbers of responses to those requests num-
bered in the low tens, while today they number in the thousands.
Chapter 5
1. Crosby sold one of his Ampex Model 200s to guitarist Les Paul, who used it to invent
multitrack recording.
2. Studio recorders of the 1960s and 1970s typically used tape with eighth-inch track
widths, such as 1-inch tape for 8-track machines and 2-inch tape for 16-track
machines, as compared to four tracks packed into quarter-inch-wide tape (1/16 inch
per track) in commercial prerecorded tapes.
3. For example, it wasn’t until 1962 that Radio Shack’s catalog listed any component
stereo systems that included tape decks. Radio Shack’s rival Lafayette waited until the
1970s when reel-to-reel had been superseded by 8-tracks and cassettes.
4. RCA introduced a quadraphonic 8-track cartridge format called Quad-8 or Q8 in
1970. Q8 was simply 8-track reconfigured as two sets of four channels instead of
four stereo pairs. Q8 also predated the popular quad vinyl formats and faded away
after a few years. Quad cassettes (four channels instead of two stereo pairs) were also
attempted but never caught on.
5. For example, note the large hole in the upper right corner of the Fidelipac cart in
Figure 5.2: this was for a pinch roller built into the player that would swing up into
place when a cart was inserted. Each 8-track tape each had its own pinch roller,
meaning that no such mechanism had to be built into players.
6. The slight taper at one end of the 8-track made it easier to determine which end to in-
sert into the machine without having to take one’s eyes off the road.
7. Another attempt at a commercial cartridge tape format was PlayTape, introduced in
1966. PlayTape was like a mini-Fidelipac, a small endless-loop mono cartridge meant
for cheap portable players. A few thousand commercial releases were made in the
format, but it never caught on, mainly because its 24-minute maximum capacity
meant that it could never hold entire albums. PlayTape also met its demise in 1970.
Notes 401
8. Portable battery-powered 8-track players eventually appeared in the 1970s; they were
never very popular.
9. Other Dolby flavors (Dolby C, S, and HX Pro) came later, but Dolby B has remained
the most commonly used.
10. CrO2 tape typically increases high-end frequency response by 2–3 KHz, so that, for
example, a tape deck with 30–14 KHz frequency response with ferric oxide tape
might offer response out to 16 or 17 KHz with CrO2.
11. Other designs with enhanced sound quality came later, including “metal”
formulations that used pure iron instead of iron oxide particles.
12. Howie Singer owned an Advent 201 tape deck. Bill Rosenblatt owned a pair of Advent
Loudspeakers.
13. Including separate heads for recording, playback, and erase instead of a single head
for everything, and a dual-capstan drive mechanism that kept tape speed more con-
stant, minimizing wow and flutter distortion.
14. Such as proprietary noise reduction and tape bias settings.
15. The “Nak” became a status symbol for wealthy audiophiles. For example, in the recent
HBO TV series Vinyl, about a record label in the 1970s, Nakamichi decks were osten-
tatiously mounted on the walls of both the label head’s office and the living room in
his palatial suburban home.
16. One more attempt at a new analog tape format came in the late 1970s: the Elcaset,
created by Sony, Panasonic, and Teac. It looked like an overgrown cassette and
contained one-quarter-inch tape that ran at 3.75 IPS. Its makers claimed sound
quality equal to the best open-reel tape. But by that time, fine audio quality was al-
ready available from cassette decks that sold for half the prices of Elcaset machines;
and record labels did not embrace the format for commercial releases. The Elcaset
disappeared quickly.
17. Of course, a more complete explanation would take up far more space than it
warrants here. Certainly recording your friends’ albums so that you didn’t have to buy
them was considered unlawful. Also of note is the difference between US/UK law and
Continental European law regarding so-called private copying; see footnote 22.
18. As we’ll see shortly, the question of the legality of taping broadcasts at home was de-
cided by the Supreme Court much later, in 1984.
19. In rough terms, this meant flat frequency response extending past 15 KHz, signal-to-
noise ratio of at least 60 dB, and wow/flutter at less than 0.1 percent. By the late 1970s,
those specs were widely available in midline cassette decks costing less than $300.
20. It was not the only one. Other likely factors included a long list of late-1970s super-
star artists who didn’t release new albums in 1979 (including The Rolling Stones, The
Who, Queen, Fleetwood Mac, and many others), the Oil Crisis in the wake of the
Iranian Revolution driving vinyl prices higher while depressing discretionary in-
come, and the inferior audio quality of many commercial cassettes.
21. The decision in CBS Songs Ltd v Amstrad Consumer Electronics Plc was, in a way, a har-
binger of the 2005 Grokster case in the United States, which we discuss in Chapter 9.
The House of Lords held that Amstrad did not induce or incite copyright infringe-
ment and therefore was not liable for any infringements caused by its customers.
402 Notes
22. Some countries in Continental Europe had by then enacted levies on blank media
to compensate for consumer copying, and eventually almost all of them did. But
the law in Continental Europe is different from that of the United Kingdom. In
Europe, unlike in Britain (and the United States), copying content for personal use
is allowed without advance permission from copyright holders, although consumers
are expected to pay royalties on those copies. So-called private copying levies are
intended as mechanisms to ensure that such royalties get paid. See International
Survey on Private Copying, Stichting de Thuiskopie and World Intellectual Property
Organization, WIPO Publication No. 1037E/16, 2015. https://www.wipo.int/edocs/
pubdocs/en/wipo_pub_1037_2016.pdf. Accessed May 27, 2022.
23. Consumer dual-cassette machines may have already been available in Japan when the
Amstrad TS55 came out.
24. Actually, the case was appealed first to the Ninth Circuit, which reversed the district
court; but then the Supreme Court reversed the Ninth Circuit on the relevant points.
25. Note that Betamax was concerned with home taping of freely available broadcast tel-
evision, not copying of prerecorded tapes. Commercial prerecorded video tapes did
not appear until after the original lawsuit was filed.
26. Fans of Frank Zappa will recognize this as an intended pun.
27. The increase in jazz releases on cassette was also surely due to the increasing pop-
ularity during the 1970s of jazz-rock fusion music that crossed over to the rock
audience.
28. It is now known as CDSA (Content Delivery and Storage Association) and run by
Guy Finley, Larry’s son.
29. After the decline of cassettes, Moore moved on to CD-Rs, then to digital releases on
various platforms.
30. Leaving aside that the parties may have taken place at venues that didn’t pay perfor-
mance royalties to PROs such as ASCAP and BMI.
31. The idea that a use of copyrighted material can be fair use if it is “transformative”
comes from a highly influential 1990 law journal article by Pierre Leval, then a US
district court judge. The article influenced subsequent court decisions about sam-
pling, such as Campbell v. Acuff-Rose, discussed later in this chapter. Pierre N. Leval,
“Toward a Fair Use Standard,” Harvard Law Review 103, no. 5 (1990): 1105–36,
https://doi.org/10.2307/1341457. Accessed April 4, 2023.
32. This could still require a license for the composition from the songwriter/music pub-
lisher, though not from the record label. In the case of “Planet Rock,” Bambaataa and
Kraftwerk worked out a royalty without going to court.
33. This behemoth was badged as the Audioton TBS9300, Dynasty HT-959, Elta 6930
(shown in Figure 5.5), ESC JC-2000, Helix Wheely 5000, International MS-959,
Matsuki MS-959, Orbex MS-959, Shadow JB750, and Technidyne TD10000. See Wiki
Boombox, Analog Alley, https://www.wikiboombox.com/tiki-index.php.
34. Raheem’s boombox, a Promax Super Jumbo with dual cassette decks, is on display
at the Smithsonian’s National Museum of African American History and Culture in
Washington, DC.
Notes 403
35. Sony also envisioned the use case of two people listening together. Early Walkman
models, including the TPS-L2, had dual headphone jacks. They also had a HOT LINE
button that would mute the music and turn on a built-in microphone so that the two
listeners could talk to each other without taking headphones off. Sony removed these
features in later models, but the use case was revived decades later in the age of earbud
sharing.
36. An exception: Radio Shack, which called its product the StereoMate. Sony itself used
the names Stowaway, Soundabout, and Freestyle in various countries before settling
on Walkman worldwide.
37. Along with audiophile disc formats such as DVD-Audio and SACD.
38. Few of which mention Guardians of the Galaxy.
Chapter 6
1. Pluggers were known to bribe the store’s salesgirls with perfume to push their tunes to
customers—perhaps the first instance of “payola” in the music business.
2. Search for “Soundies” on YouTube and you will be able to watch performances by
many of these artists.
3. Although the program gave the impression that the countdown was based on
record sales, the final songs selected were under the control of the ad agency of the
sponsor: Lucky Strike cigarettes.
4. Miller was a successful A&R executive and record producer who signed Johnny
Mathis, Tony Bennett, and Patti Page, to name just a few. He loved pop but thought
rock & roll was mediocre and famously passed on signing Buddy Holly, Elvis Presley,
and the Beatles.
5. The night that the Beatles first appeared on Ed Sullivan, the variety show featured a
musical performance from the hit musical Oliver as well. The original Broadway cast
that night included “David” Jones as the Artful Dodger. Like so many other people
who saw that Beatles performance, Jones decided that rock & roll would be his future.
6. Their hits included contributions from many of the top songwriters of the rock era
including Carole King & Gerry Goffin and Neil Diamond.
7. QUBE was a bold early experiment in interactive television that presaged many later
developments—including consumer privacy concerns that helped sink the service in
the early 1980s.
8. This same pattern would be repeated with Apple and Steve Jobs after Napster’s deci-
mation of the business, to be discussed in Chapter 8.
9. Malone is now Chairman of Liberty Media, whose assets include Sirius XM Satellite
Radio, Pandora, and part of iHeartMedia. See Chapter 3.
10. The baby boomers would eventually get their “own” music video channel when MTV
launched VH-1 in 1985.
11. Young even wrote the song “This Note’s for You” as a criticism of corporate adver-
tising. Despite those sentiments, it won MTV’s Video of the Year award in 1989.
404 Notes
12. You can search for and watch the video on YouTube, where it has 5.6 million views at
this writing.
13. Thriller is the only album in history to sell more than 50 million copies.
14. History repeated itself almost 20 years later, when the AOL-Time Warner merger
turned out so poorly that the company sold the then-struggling Warner Music Group
to private investors who later made huge profits.
15. A summary of the use cases and payments for synchronization can be found in this
Continuing Legal Education session at South by Southwest in 2014: Bob Donnelly,
Danielle Aguirre, Kris Muñoz, and Andrew Sparkler Donnelly, “Flipping Publishing
Micro-Pennies Into a Bucket,” SXSW 2014 CLE, https://foxrothschild.gjassets.com/
content/uploads/2017/03/Flipping-Publishing-Mirco-Pennies-into-a-Bucket.pdf.
Accessed May 4, 2022.
16. In many other cases, rights holders have refused to entertain synch licenses on any
terms at all. For example, the film Almost Famous, a paean to 1970s rock released
in 2000, tried to license the ultimate 1970s rock song, “Stairway to Heaven,” for the
soundtrack, but Led Zeppelin refused. More recently, former Pink Floyd bassist/
songwriter Roger Waters refused to license Floyd’s “Another Brick in the Wall” to
Mark Zuckerberg to promote Instagram.
Chapter 7
1. The other colors of the “rainbow” describe the later specifications for discs, including
the writeable and rewriteable variations.
2. Rashomon is a 1950 psychological thriller by Akira Kurosawa known for its characters
providing their different subjective views of the same crime.
3. You can listen to these recordings at the Smithsonian website: https://www.si.edu/
newsdesk/releases/playback-130-year-old-sounds-revealed. Accessed May 18, 2023.
4. Higher sampling rates, larger sample sizes, and various sampling optimization
techniques have been used to provide even higher quality digital versions in both op-
tical disc and file download formats that appeal to audiophiles. For example, DVD-
Audio discs contain audio that is sampled at as much as 192 kHz at up to 24 bits per
sample.
5. Recall from Chapter 3 that FM radio sound quality doesn’t quite have the frequency
response range of LPs.
6. The Beatles were slow in making their music available in both downloads and
streaming as well.
7. Another factor that led to Thriller’s phenomenal sales was its crossover appeal to white
audiences, thanks in part to Eddie Van Halen’s shredding guitar solo on “Beat It.”
8. The title refers to Neil Young’s 1972 song “The Needle and the Damage Done” about
the perils of heroin addiction for musicians. Replacing “needle” with “CD” is a clever
reference to the change in formats.
Notes 405
9. Young would be even more upset when the even poorer quality of the MP3 file came
to prominence.
10. In fact, equivalents to this law in countries outside the United States are known as
copyright exhaustion instead of first sale.
11. Japan never enacted similar legislation, allowing a CD rental business to become a
thriving channel there for many years. However, the cultural norms there have dis-
couraged unauthorized copying.
Chapter 8
1. File Transfer Protocol (FTP) is a standard for transmitting data files over computer
networks that was invented in the early 1970s. It’s widely used today, though not by
casual users.
2. Integrated Services Data Network (ISDN), a telecommunications protocol created
in the late 1980s, was about 50 times faster than the dial-up modems of the time. It
was slow to take off and was eventually eclipsed by DSL and cable. Once broadband
Internet began to catch on, ISDN became known as “It Still Does Nothing.”
3. Audio File Size Calculations, AudioMountain.com, https://museumofportablesound.
com/mp3/. Accessed May 23, 2022.
4. For example, a typical Dell desktop PC in 1995 included a 1 GB hard disk drive and
sold for around $2,000, or $3,700 in today’s dollars.
5. For example, Albhy Galuten, a Grammy award winning producer, inventor of the
“drum loop,” and technologist with numerous US patents, worked at Universal Music
at that time.
6. This was an exemplar of Metcalfe’s Law, a sequel to Moore’s Law that states that the
value of a network is proportional to the square of the number of devices attached to
the network.
7. Some of the other major record labels engaged in their own trials, around the same
timeframe, with other CD copy-protection schemes that were arguably less harmful
to users’ PCs; but the firestorm around the technology that Sony BMG adopted
caused those efforts to be abandoned quickly.
8. Pun intended.
9. This wasn’t the case in many other countries, which enacted levies (taxes) on devices
with digital storage, including personal computers. See Chapter 5, note 22.
10. Copyright law doesn’t require a plaintiff to prove actual damages, i.e., actual financial
harm it claims to have suffered from the defendant’s behavior. Instead, a copyright
plaintiff can ask for statutory damages, which are amounts set in law rather than by
economic damage assessments. At the time of the Napster case, the limit on statutory
damages for willful infringement was $100,000 per copyrighted work (plus attorneys’
fees); it has since been raised to $150,000.
11. The central directory indicated willful infringement, which could entitle the plaintiffs
to higher damages. In addition, the court found that Napster’s central directory
406 Notes
rendered it liable for so-called vicarious infringement, which essentially means
“getting paid to look the other way.”
12. Reciprocal, a Softbank/Microsoft funded start-up offering secure digital distribu-
tion services, was one of the companies that made such a proposal to Napster. Howie
Singer was an executive at Reciprocal.
13. The highest damage award in a court case stemming from failure to qualify for DMCA
safe harbors was $1 billion, awarded in 2019 in Sony Music Entertainment et al. v.
Cox Communications, a case that the major record labels brought against Cox, a large
cable ISP. At this writing, the damage award is on appeal.
14. Justice David Souter adapted the concept of inducement of copyright infringement
from patent law, where it had long been established. A company can sell a product that
infringes a patent when someone who buys the product uses it. Even though the user
is the “direct infringer” (the one who is technically liable for patent infringement), the
inducement principle makes it possible for patent owners to go after companies that
“induce” users to infringe their patents.
15. There were two exceptions to this. Apple licensed the technology to HP, which
released a line of iPod clones in 2004. It also licensed the technology to Motorola
for use with Motorola’s ROKR MP3-playing mobile phones starting in 2005. None of
these was successful; the candy-bar-style ROKR was eclipsed by the iPhone in 2007.
16. The pioneering streaming service Rhapsody, which we’ll meet in Chapter 9, acquired
this version of Napster in 2011. In 2016, the company phased out the Rhapsody brand
name and rebranded the service (yet again) as Napster.
17. For a detailed and dispassionate look at the history and technology of FairPlay from
one of its engineers, see Rod Schultz, “The Many Facades of DRM,” 2012. Available at
https://web.archive.org/web/20150105090945/http://fortunedotcom.files.wordpress.
com/2014/12/2012_misc_drm.pdf. Accessed May 18, 2023.
18. It is actually a gross overstatement to say that DRM for music has gone away. It has
been eliminated for paid downloads, but all of the interactive streaming services,
discussed in Chapter 9, use forms of DRM, including for “offline listening,” a.k.a.
tethered downloads. It is more accurate to say that a larger portion of recorded music
industry revenue comes from DRM-based channels nowadays than in 2007. See
https://copyrightandtechnology.com/2017/02/16/the-myth-of-drm-free-music-
revisited/.
19. The album only content was often limited in time, and eventually all the tracks be-
came available as singles.
20. The labels unsuccessfully tried to create an open standard for a similar enhanced
album concept called CMX that would work on all devices including mobile phones.
21. The previous agreement between the companies had confined Jobs and Apple to use
the Apple trademark for computer-related products but not for music, to avoid confu-
sion with the music label of the same name founded and owned by the Beatles.
22. Superstar artists are sometimes able to negotiate rates as high as 25 percent.
23. As a matter of copyright law, a download purchase is a license, not a transfer of own-
ership as with a CD or LP. However, these cases turned primarily on language in the
terms of contracts between labels and artists.
Notes 407
24. Neil Young later created his own high-resolution music player called Pono and an as-
sociated download store to create a superior audio experience.
25. Elaborate schemes were set up to steal CDs from factories before they were released
to stores to ensure that the newest songs were available before the discs were even on a
retail shelf.
26. Some of which were artifacts of tactics to evade technologies that detected unauthor-
ized copies on the networks.
27. The methodology did not examine causality, and the correlation of interest in music
could have provided the same results.
Chapter 9
1. Universal Music Group’s IPO filing even went so far as to give “Music as a Service” the
acronym of “MaaS.”
2. For example, the early version of Microsoft’s DRM that Rhapsody used for downloads
did not support transfer of music files to portable devices.
3. For further details on possible shifts, see Howie Singer, “In 2022, ‘Fan-Centric’
Accounting Will Bring Emerging Artists More Money from Streaming Music,” dot.
LA, https://dot.la/what-streaming-music-services-pay-2656063221.html. Accessed
July 17, 2022.
4. During the download era, Apple was rumored to be contemplating the purchase of a
label, but that also never came to fruition.
5. The attacks on Ukraine have led many services to halt their operations in Russia.
6. You can find the Key Changes playlist for the song titles included as part of each
chapter heading in this book at https://spoti.fi/3InaDsO.
7. Although, as often happens, friction in a user experience leads to opportunities for
entrepreneurs. Startups such as SongShift and Soundiiz have developed tools for
sharing playlists across music services.
8. In November 2022, Deutsche Grammophon, the iconic label owned by Universal
Music Group, announced STAGE+, a higher quality streaming service catering to
classical fans featuring audio, video, live performances, interviews, and other exclu-
sive content. And in March 2023, Apple launched Apple Music Classical, a separate
app from Apple Music for classical fans with over 5 million tracks in its catalog. As
we explain in Chapter 10, classical music is especially attractive as a niche service
because of its unique metadata requirements compared to those of mainstream
services.
9. After an initial climb up the Country charts, Billboard reconsidered this classification
and removed the song from that list, though it did not blunt its momentum on the
genre-less Hot 100.
10. Frank was the Senior VP for Global Streaming Marketing at Universal Music Group.
After his untimely passing at the age of 47 in 2019, Billboard recognized his thought
leadership in digital music by creating an award in his name.
408 Notes
11. The forms of fraud we discuss here involve gaming the system to increase streams
rather than the separate matter of giving financial considerations directly to streaming
services (as payola did for radio stations).
12. When labels were reissuing albums on CDs in the 1980s and 1990s, they had to re-
clear the mecha\nicals with music publishers. A backlog of uncleared mechanicals
developed, which became the subject of lawsuits between labels and publishers that
were eventually settled.
13. Epic Games prevailed over Apple at the district court level in November 2022; Apple
is appealing. The case against Google is ongoing at this time of writing.
Chapter 10
1. As a rough rule of thumb, compressed music files take up a megabyte of storage space
per minute of audio, while compressed video takes up a gigabyte per hour, both at decent
quality levels. This means that video takes up about 16 times the amount of space as audio.
2. YouTube abandoned Flash around 2010 once the HTML standard began to incorpo-
rate video natively (HTML5).
3. The major mobile operators had completed their buildouts of 3G wireless by the
2007–2008 timeframe, as we saw in Chapter 9, but while 3G was fast enough to sup-
port streaming audio, it wasn’t sufficient for most video.
4. As the technologist for Warner Music’s strategy group, Howie Singer negotiated the
specifics of this arrangement with YouTube.
5. Of course, this is not an apples-to-apples comparison. Many songs have multiple
videos on YouTube, and many music videos uploaded to YouTube are not of commer-
cially released music.
6. There are other highly popular apps that enable sharing of short-form video, such as
Byte and the video features built into Instagram and Snapchat. But none of these par-
ticularly feature music.
7. YouTube had taken similar steps: its original length limit was 5 minutes, then 10
minutes, then all limits were removed.
8. The number three was never specified in law, but the baseball analogy had a certain
ring to it and became conventional, if not standard, among online services.
9. YouTube users can earn ad revenue shares if they exceed certain thresholds of user
video views and channel subscribers.
10. Fair use is an affirmative defense in copyright law—it’s an excuse you give in court
when you’re sued for copyright infringement, and it’s something that ultimately
only a court can decide. There are many reasons why a particular use of content is or
isn’t likely to be fair use; for example, criticism and scholarly research are generally
accepted as fair use. Further discussion of this is beyond the scope of this book.
11. Google has never entered the MIREX competition. In a much less scientific test, Bill
Rosenblatt recently uploaded to YouTube a set of classic rock covers from an audience
recording of a live performance by a band he plays in; Content ID flagged only a third
of them. Yet when he uploaded studio recordings of the band’s classic rock covers a
few months later, Content ID identified all of the compositions correctly.
Notes 409
12. SoundCloud also fulfills this function, for similar reasons. But YouTube had a three-
year head start on SoundCloud, so YouTube is used more often for this purpose, while
SoundCloud is more popular for indie artists who want to eschew record labels and
release their music directly to fans.
13. As mentioned earlier in this chapter, YouTube has a “copyright strike” system, in
which it will terminate a user’s account if videos the user posts receive three valid and
unresolved takedown notices. This does not prevent other users from reposting the
same content.
14. “Notice and staydown” is controversial, but the details—though quite interesting—
are beyond the scope of this book.
15. The Copyright Office’s study included several suggestions for DMCA reform, but “no-
tice and staydown” was not among them.
16. Except in small niche markets. For example, WeAreTheHits.com has a library of a few
million musical compositions that are fully precleared for use on YouTube. Musicians
can post videos of cover versions of those compositions through WeAreTheHits.com
and earn ad revenue shares (less WeAreTheHits.com’s commission).
17. Pun intended.
18. It’s worth noting that record labels were, at that time, still working feverishly to digi-
tize their entire catalogs to make them available online. So the term “full catalog” was
something of a moving target.
19. Laws in countries outside the United States vary, but “notice and takedown” is fairly
common as a pragmatic implementation of some national laws, not to mention that
it’s easier to export takedown notice–processing systems around the world than
it is to implement different systems that reflect the legal nuances in each different
country.
20. Technically, lyric videos posted without authorization could be found to infringe
copyrights in the lyrics. But there is no history of artists or labels suing indi-
vidual users over lyric videos; at most, they use the DMCA process to have them
taken down.
21. See the Guardian article cited in the previous sentence. This number could be
reconciled with the data in that article by recognizing that YouTube pays out ad rev-
enue shares to many independent artists that do not collect royalties through paid
subscription services.
22. Typing “Arabian Prince” into YouTube brings up videos of Saudi Crown Prince
Mohammed bin Salman as well as the hip-hop pioneer whom we interviewed for
this book.
23. Services that focus on classical music expend even more effort on this to create and
clean up metadata on composers, conductors, soloists, movements, and so on. One
such service, Primephonic, was acquired by Apple in August 2021; Apple launched
Apple Music Classical in March 2023 with Primephonic’s metadata and editorial
talent at the heart of the service.
24. YouTube Music still falls short of curated opt-in services in showing discographies of
longtime artists such as Miles Davis and Elvis Presley who have large catalogs that are
replete with reissues, special editions, anthologies, and so on.
410 Notes
25. For example, in Apodaca’s video, he drank from a bottle of Ocean Spray cranberry
juice; in gratitude, Ocean Spray sent him a brand new pickup truck loaded with Cran-
Raspberry juice.
26. TikTok also has text hashtags that are similar to those on Twitter, LinkedIn, and so on.
27. Not to be confused with the late jazz soprano saxophonist of the same name.
Chapter 11
1. “abcdefu,” the 2022 single by Gayle, was mastered with an AI-based tool called
LANDR and nominated for a Song of the Year Grammy.
2. No relation to the author.
3. One of the early developers of speech recognition software for PCs was Ray Kurzweil,
who also invented one of the first commercially available digital sampling synthesizers
in collaboration with Stevie Wonder.
4. Accordingly at this time of writing there is no native support on Apple’s HomePod
smart speakers for Spotify or YouTube Music.
5. A focus track is akin to a single: it’s the track from an artist’s output that the label
wants to push at a given point in time.
6. The set of standards is called DDEX (Digital Data Exchange). DDEX also includes
standards for communicating more fundamental information about track releases
from labels to music services and information about song plays and record sales in
the opposite direction.
7. Collaborative filtering dates back to the mid-1990s, and a music recommendation
system called Firefly was the first commercial application.
8. There is an emerging field of synthetic voices that imitate celebrities with their (or
their estates’) permission, so that their voices can do more without their personal
involvement.
9. A work for hire is one that is created under contract to someone else, with the implica-
tion that the creator relinquishes rights to that work under copyright law.
10. Data mining is a process for analyzing large sets of data to extract patterns using sta-
tistical and other techniques. It is often used in training AI systems.
11. For further details on Copyright and Artificial Intelligence, see Briggitte Vezina
and Diane Peters, “Why We’re Advocating for a Cautious Approach to Copyright
and Artificial Intelligence,” Creative Commons, February 20, 2020, https://creative
commons.org/2020/02/20/cautious-approach-to-copyright-and-artificial-intelligence/.
Accessed June 2, 2022.11.
12. Some music companies have taken the position that audio fingerprints are derivative
works for purposes of negotiations with fingerprinting technology vendors; but this
theory has not been tested in court.
13. Notably, the lawsuit does not accuse the creators of Copilot of copyright infringe-
ment; it merely accuses them of violating terms of service, that is, of breach of license
(contract). The plaintiff may have done this strategically so that the defendants cannot
raise a fair use defense—which is possible under copyright but not contract law.
Notes 411
Chapter 12
1. Pun intended.
2. Will Page, the former Chief Economist at Spotify, used this phrase as the title of his
recent book on digital disruption.
3. Creative Commons is a scheme for providing licenses to digital content that gives
licensees rights beyond those normally granted under copyright law. Creative
Commons licenses for OER typically allow a licensee to copy the content freely and
merely require attribution of the content to its author.
4. “Shovelware” is text material traditionally published in print that is converted to
a digital format “as is,” with little or no reworking to make it digital-friendly. The
resulting product is usually not a great user experience, especially on mobile devices.
5. Craigslist did eventually begin to charge for some categories of listings, such as em-
ployment opportunities.
6. There are myriad legitimate uses of BitTorrent. For example, several online game
publishers use BitTorrent to distribute the enormous collections of software and
media files that make up their games; Facebook and Twitter use it to distribute server
software updates.
7. Pun intended again.
8. Pun intended again.
9. The group was later renamed to use the word “Jazz,” reflecting a shift to the more rec-
ognizable terminology.
10. This term originated when segregated America and a biased press was searching for
a boxer—any boxer—who could defeat Jack Johnson, who had become the first Black
athlete to win the world heavyweight championship in 1908.
11. Jay-Z had to sell his ownership stake because he started a sports agency.
12. Sony Music was an exception to the rule about labels not launching their own formats
and digital music stores—because, of course, it was a sibling to a consumer electronics
giant. Sony made a couple of attempts to launch its own digital music technology stack
and retail presence, often in partnership with other labels and electronics makers.
None of these succeeded.
13. The number will surely be significantly higher by the time you read this.
14. Disney and 20th Century Fox did not participate in the joint venture but did license
their content to MovieLink.
15. No relation to the author.
Afterword
1. Such data could include geography (country), type of user account (paid subscription
vs. free), bit rate, and so on.
2. Another scalability constraint with current blockchains is that they don’t have the ca-
pacity to store music or other digital content files themselves, although innovators are
addressing that limitation too.
412 Notes
3. The RIAA’s vinyl sales figures do not count some independent releases and are US
only; Water & Music’s NFT sales database does not necessarily track every music
NFT ever minted but is not limited to the United States. None of these figures counts
resales.
4. Although almost half of this is from a single NFT campaign by Snoop Dogg in
February 2022, which netted over $44 million from 8,934 NFTs that sold for $5,000
apiece.
5. This curve shape is so common that the eminent technology market research firm
Gartner calls it the Hype Cycle and has built an entire research methodology around
it. See https://www.gartner.com/en/research/methodologies/gartner-hype-cycle.
6. In exchange for some combination of more fees or commissions, and/or with features
designed to keep users glued to their services instead of competitors’.
7. The NFTs that have set price records have all been for visual artworks. At this writing,
the highest price paid for an NFT was $91.8 million for Merge, a digital artwork by
the artist known as Pak, in December 2021. In comparison, the most expensive single
music NFT to date is one by electronic music artist 3LAU, which went for $1.33 mil-
lion in March 2021.
8. NFTs with content access control technologies have begun to appear, though the
purveyors of such technologies studiously avoid using the term “DRM.”
9. Other countries such as the United Kingdom and France have resale royalties set in
law, which apply to unique or limited-edition works of art sold at auction or by pro-
fessional dealers. Attempts to pass similar laws in the United States have failed.
10. See for example “20 Questions: An Artist’s Checklist for an NFT Pitch,” https://musict
ech.solutions/2022/03/01/20-questions-an-artists-checklist-for-an-nft-pitch/, from
music tech attorney Chris Castle. Accessed May 17, 2023.
11. At this writing, the best explanation of the relationship between NFTs and copyright
that we have seen is a piece by the noted Internet law scholar James Grimmelmann
and two colleagues at Cornell University and the Initiative for Cryptocurrencies and
Contracts (IC3); see James Grimmelmann, Yan Ji, and Tyler Kell, “The Tangled Truth
about NFTs and Copyright,” The Verge, https://www.theverge.com/23139793/nft-cry
pto-copyright-ownership-primer-cornell-ic3. Accessed May 17, 2023.
12. The legal scholar Lawrence Lessig proposed a compelling framework for under-
standing how the forces of law, social norms, the market, and technology work to-
gether to regulate environments such as the Internet in his book Code: And Other
Laws of Cyberspace (New York: Basic Books, 1999). This framework, which has come
to be known as the Pathetic Dot Theory, will surely be used to analyze the progress of
Web3 in the years to come.
References
Chapter 1
i Thomas Ricker, “First Click: Remember When Steve Jobs Said Even Jesus
Couldn’t Sell Music Subscriptions?” The Verge, June 8, 2015, https://www.
theverge.com/2015/6/8/8744963/steve-jobs-jesus-people-dont-want-music-
subscriptions. Accessed June 28, 2022.
ii “Nikki Sixx Quotes,” Inspirational Stories, https://www.inspirationalstories.
com/quotes/t/nikki-sixx/. Accessed June 28, 2022.
iii Jem Aswad, “Global Music Biz Revenue to Double to $131 Billion in 2030, Says
Bullish Goldman Sachs Report,” Variety, June 14, 2022, https://variety.com/
2022/music/news/global-music-revenue-double-goldman-sachs-report-123
5293827/. Accessed June 28, 2022.
iv Bill Rosenblatt, “The Short, Unhappy Life of Music Downloads,” Forbes.com,
May 7, 2018, https://www.forbes.com/sites/billrosenblatt/2018/05/07/the-
short-unhappy-life-of-music-downloads/. Accessed June 5, 2022.
Chapter 2
i Vinylmint, “History of the Record Industry 1877–1920s,” Medium, June 7,
2014, https://medium.com/@Vinylmint/history-of-the-record-industry-
1877-1920s-48deacb4c4c3. Accessed May 9, 2022.
ii David J. Steffen, From Edison to Marconi: The First Thirty Years of Recorded
Music (Jefferson, NC: McFarland, 2005), 23–25.
iii David L. Morton Jr., Sound Recording: The Life Story of a Technology
(Baltimore: Johns Hopkins University Press, 2004), 13–17.
iv Steffen, From Edison to Marconi, 27–30.
v Randall Stross, The Wizard of Menlo Park (New York: Broadway Books, 2007),
157–63.
vi Emile Berliner and Frederic Wile, Emile Berliner, Maker of the Microphone
(Indianapolis, IN: Bobbs-Merrill, 1926), 171–73.
vii Morton, Sound Recording, 31–42.
viii Roland Gelatt, The Fabulous Phonograph 1877–1977, Revised 2nd Edition
(New York: MacMillan Publishing, 1977), 82–86.
ix Andre Millard, America on Record: A History of Recorded Sound, 2nd Edition
(Cambridge: Cambridge University Press, 2005), 125–28.
x “History of the Record Industry 1877–1920s.”
xi David L. Suisman, Selling Sounds: The Commercial Revolution in American
Music (Cambridge, MA: Harvard University Press, 2009). 94–96.
414 References
Chapter 3
i Marc Fisher, Something in the Air: Radio, Rock, and the Revolution That
Shaped a Generation (New York: Random House, 2007), 11–14; Christopher
H. Sterling and John Michael Kittross, Stay Tuned: A History of American
Broadcasting (Milton Park: Routledge, 2001), 63, 72.
ii Lawrence Lessig, “Laws that Choke Creativity,” TED Talks, https://www.ted.
com/talks/lawrence_lessig_laws_that_choke_creativity. Accessed April
29, 2022.
iii Charles Fitch, “How FM Stereo Came to Life,” Radio World, January 27, 2016,
https://www.radioworld.com/columns-and-views/roots-of-radio/how-fm-
stereo-came-to-life. Accessed August 3, 2022.
iv “The Infinite Dial 2022,” Edison Research and Triton Digital, http://www.edi
sonresearch.com/wp-content/uploads/2022/03/Infinite-Dial-2022-Webinar-
revised.pdf. Accessed May 25, 2022.
v Sterling and Kitross, Stay Tuned, 339–41.
vi Peter Fornatale and Joshua Mills, Radio in the Television Age
(New York: Overlook Press, 1980), 125–26.
vii Serge Denisoff, Solid Gold: The Popular Record Industry (Piscataway,
NJ: Transaction Publishers, 1975), 257–58.
viii Paul Rappaport, interview with the author, October 2020.
ix Denisoff, Solid Gold, 267–69.
x Rappaport, interview with the author.
416 References
xi Mark Hugo Lopez and Daniel Dockterman, “U.S. Hispanic Country-of-
Origin Counts for Nation, Top 30 Metropolitan Areas,” Pew Research Center,
May 26, 2011, https://assets.pewresearch.org/wp-content/uploads/sites/7/
reports/142.pdf. Accessed April 5, 2023.
xii Subcommittee on Courts, Civil Liberties, and the Administration of
Justice of the Committee on the Judiciary, House of Representatives,
95th Congress, Performance Rights in Sound Recordings. Washington,
DC: US Government Printing Office, 1978, quoting Hearings on Economic
Conditions in the Performing Arts Before the Select Subcommittee on
Education of the House Committee on Education and Labor, 87th Cong.,
1st and 2d Sess. (1961–62).
xiii Fornatale and Mills, Radio in the Television Age, 13.
xiv Performance Rights in Sound Recordings.
xv Fisher, Something in the Air, 279–80.
xvi “The Infinite Dial 2022.”
xvii Larry Miller, “Sync or Swim—Licensing Music for Podcasts.” Musonomics
(podcast), September 28, 2020, http://musonomics.org/sync-or-swim-licens
ing-music-for-podcasts/. Accessed April 29, 2022.
xviii Interactive Advertising Bureau, “IAB U.S. Podcast Advertising Revenue
Study: Full-Year 2021 Results & 2022–2024 Growth Projections,” Interactive
Advertising Bureau, https://www.iab.com/wp-content/uploads/2022/05/IAB-
FY-2021-Podcast-Ad-Revenue-and-2022-2024-Growth-Projections_FINAL.
pdf. Accessed November 29, 2022.
xix Richard Neer, FM: The Rise and Fall of Rock Radio (New York: Villard,
2001), 31–32.
xx Ronald H. Coase, “Payola in Radio and Television Broadcasting.” Journal of
Law and Economics 22, no. 2 (October 1979), 269–328.
xxi Fisher, Something in the Air, 91.
xxii Denisoff, Solid Gold, 276–278.
xxiii Frederic Dannen, Hit Men: Power Brokers and Fast Money Inside the Music
Business (New York: Vintage, 1991), 115–19.
xxiv John Jackson, A House on Fire: The Rise and Fall of Philadelphia Soul
(Oxford: Oxford University Press, 2004), 177–80, 191–94.
xxv Neer, FM, 153.
xxvi Dannen, Hit Men, 201, 206.
xxvii Peter Boyer, “CBS Assails NBC News over a Payola Report,” New York Times,
April 3, 1986, https://www.nytimes.com/1986/04/03/arts/cbs-assails-nbc-
news-over-a-payola-report.html. Accessed July 12, 2022; Dannen, Hit Men,
286–88.
xxviii Penny Pagano and William Knoedelsleder Jr., “Senate Plans Record Industry
Payola Probe,” The Los Angeles Times, April 3, 1986, https://www.latimes.com/
archives/la-xpm-1986-04-03-mn-2589-story.html. Accessed April 4, 2023;
Dannen, Hit Men, 291, 302.
xxix William Knoedelsleder, “$1 Million in Suspected ‘New Payola’ Is Probed,”
Los Angeles Times, December 22, 1987, https://www.latimes.com/archives/la-
xpm-1987-12-22-mn-30675-story.html. Accessed April 3, 2023; Dannen, Hit
Men, 296.
xxx Dannen, Hit Men, 286–312.
References 417
xxxi Lola Ogunnaike, “Record Labels Must Pay Shortchanged Performers,”
New York Times, May 5, 2004, https://www.nytimes.com/2004/05/05/arts/rec
ord-labels-must-pay-shortchanged-performers.html. Accessed April 4, 2023.
xxxii Jeff Leeds, “EMI Agrees to Fine to Resolve Payola Case,” New York Times, June
16, 2006, https://www.nytimes.com/2006/06/16/business/worldbusiness/
emi-agrees-to-fine-to-resolve-payola-case.html. Accessed April 4, 2023.
xxxiii Jeff Leeds, “Radio Broadcasters Agree to Fine over Payola,” New York Times,
March 6, 2007, https://www.nytimes.com/2007/03/06/technology/06iht-pay
ola.4812569.html. Accessed April 4, 2023.
xxxiv Elias Leight, “Want to Get on the Radio? Have $50,000?” Rolling Stone, August
6, 2019, https://www.rollingstone.com/pro/features/radio-stations-hit-pay-
for-play-867825/. Accessed April 4, 2023.
xxxv Greg Milner, Perfecting Sound Forever: An Aural History of Recorded Music
(New York: Farrar, Straus and Giroux, 2010), 153–56.
xxxvi Buck Owens and Randy Poe, Buck ‘Em!: The Autobiography of Buck Owens
(Lanman, MD: Backbeat, 2013), 141.
xxxvii Peter Asher, interview with the authors, September 2020.
xxxviii Milner, Perfecting Sound Forever, 129–32.
xxxix Milner, Perfecting Sound Forever, 237–39.
xl Milner, Perfecting Sound Forever, 284–86.
xli Greil Marcus, Like a Rolling Stone: Bob Dylan at the Crossroads
(New York: PublicAffairs, 2006), 141.
xlii Jim McKeon, interview with the author, September 2020.
xliii Jac Holzman, interview with the authors, September 2020.
xliv McKeon, interview.
xlv McKeon, interview.
xlvi “Nielsen Report Puts Radio at the Center of the Audio Universe,” InsideRadio,
June 30, 2022, https://www.insideradio.com/nielsen-report-puts-radio-at-
the-center-of-the-audio-universe/article_a3188d52-f7f1-11ec-8617-6b758
6059cd3.html. Accessed January 30, 2023.
xlvii Techsurvey 2022 Results, Jacobs Media, https://jacobsmedia.com/techsurvey-
2022-results/. Accessed November 22, 2022.
xlviii The Infinite Dial 2022.
xlix “Network Radio Averaged 16 Minutes of Spots per Hour, as Rates Crept Up,”
InsideRadio, November 29, 2018, http://www.insideradio.com/free/network-
radio-averaged-16-minutes-of-spots-per-hour-as-rates-crept-up/article_9
7ff6ae0-f3a8-11e8-b201-8770a1fdf2ed.html. Accessed April 29, 2022.
l “What Do Ad Loads Look Like on Internet Radio?” Marketing Charts, https://
www.marketingcharts.com/digital-66030. Accessed April 29, 2022.
Chapter 4
i Robert Levine, “For the Record: How Vinyl Got its Groove Back—to the Tune
of a Billion Dollars.” Billboard, May 25, 2022, https://www.billboard.com/pro/
vinyl-boom-analysis-for-the-record. Accessed July 21, 2022.
ii Roland Gelatt, The Fabulous Phonograph, 1877–1977 (New York: MacMillan,
1977), 290.
418 References
iii Scott Thill, “1948: Columbia’s Microgroove LP Makes Albums Sound Good,”
Wired, June 21, 2010, https://www.wired.com/2010/06/0621first-lp-released/.
Accessed May 22, 2022.
iv Gelatt, The Fabulous Phonograph, 293.
v Sean Wilentz, 360 Sound: The Columbia Records Story (San Francisco: Chronicle
Books, 2012), 127.
vi Gelatt, The Fabulous Phonograph, 292; Jac Holzman, interview with the
authors, September 2020.
vii Wilentz, 360 Sound, 171.
viii Vinylmint, “History of the Record Industry, 1920—1950s,” Medium, June 8,
2014, https://medium.com/@Vinylmint/history-of-the-record-industry-
1920-1950s-6d491d7cb606. Accessed May 25, 2022.
ix Gelatt, The Fabulous Phonograph, 295.
x Peter Copeland, “Manual of Analogue Sound Restoration Techniques,” The
British Library. https://web.archive.org/web/20110409161239/http:/www.
bl.uk/reshelp/findhelprestype/sound/anaudio/analoguesoundrestoration.
pdf. Accessed April 29, 2022.
xi Emory Cook, “Binaural Disks,” High Fidelity, November-December
1952, 33–35.
xii Andre Millard, America on Record: A History of Recorded Sound
(Cambridge: Cambridge University Press, 2005), 192.
xiii Millard, America on Record, 212.
xiv Gelatt, The Fabulous Phonograph, 317.
xv Daniel Duggan, “Investor Buys Former Handleman Co. HQ in Troy for $3
Million,” Crain’s Detroit Business, February 24, 2010, https://www.crainsdetr
oit.com/article/20100224/DM01/302249988/investor-buys-former-handle
man-co-hq-in-troy-for-3-million. Accessed May 25, 2022.
xvi Denisoff, Solid Gold, 191–92.
xvii Denisoff, Solid Gold, 194–98.
xviii Denisoff, Solid Gold, 201–7.
xix Denisoff, Solid Gold, 210–12.
xx Russell and David Sanjek, American Popular Music in the 20th Century
(Oxford: Oxford University Press, 1991), 86.
xxi Sanjek, American Popular Music in the 20th Century, 130.
xxii Sanjek, American Popular Music in the 20th Century, 129–30.
xxiii Daniel Brockman and Jason W. Smith, “The Rise and Fall of the Columbia
House record Club: And How We Learned to Steal Music,” Boston Phoenix,
November 18, 2011, https://thephoenix.com/Boston/music/129722-rise-and-
fall-of-the-columbia-house-record-clu/. Accessed May 25, 2022.
xxiv Sanjek, American Popular Music in the 20th Century, 130.
xxv Denisoff, Solid Gold, 191.
xxvi Sanjek, American Popular Music in the 20th Century, 144–45.
xxvii Sanjek, American Popular Music in the 20th Century, 204.
xxviii Sanjek, American Popular Music in the 20th Century, 217.
xxix “A Short History of How Jukeboxes Changed the World,” Rock-Ola. https://
www.rock-ola.com/blogs/news/a-short-history-of-how-jukeboxes-changed-
the-world. Accessed May 25, 2022.
References 419
xxx David van Etten, “American Jukebox History.” https://www.jukeboxhistory.
info/. Accessed May 25, 2022.
xxxi Van Etten, “American Jukebox History.”
xxxii James Mishra, “The Dark History of the Jukebox: How the Mafia Used Murder
to Build Music Machine Empires,” Click Track, May 15, 2020, https://www.
clicktrack.fm/p/the-dark-history-of-the-jukebox-how. Accessed November
12, 2022.
xxxiii Robert Shelton, “Happy Tunes on Cash Registers,” New York Times, March 16,
1958, https://www.nytimes.com/1958/03/16/archives/happy-tunes-on-cash-
registers-record-industry-sees-tape-and-stereo.html. Accessed April 4, 2023.
xxxiv Associated Press, “Vinyl music gives record stores a boost in a digital world,”
ABC News, April 19, 2017, https://web.archive.org/web/20180207210919/
https://abcnews.go.com/amp/Entertainment/wireStory/vinyl-music-record-
stores-boost-digital-world-46887652. Accessed April 29, 2022.
xxxv Lyndsey Havens, “Why Did Vinyl Subscriptions Spike This Year—And Will
It Continue?” Billboard, December 18, 2020, https://www.billboard.com/artic
les/business/9501250/vinyl-subscriptions-spike-2020-bright-eyes/. Accessed
April 3, 2023.
xxxvi Bill Rosenblatt, “Vinyl Is Bigger Than We Thought. Much Bigger,” Forbes,
September 18, 2018, https://www.forbes.com/sites/billrosenblatt/2018/09/18/
vinyl-is-bigger-than-we-thought-much-bigger/. Accessed April 29, 2022.
xxxvii 2018 figures, eBay (used), RIAA (new).
xxxviii Amina Niasse, “Vinyl Record Sales Climb Just 1% After Years of Rapid
Growth,” Bloomberg, July 14, 2022, https://www.bloomberg.com/news/artic
les/2022-07-14/vinyl-record-sales-climb-just-1-after-years-of-rapid-growth.
Accessed July 21, 2022.
xxxix Millard, America on Record, 209.
xl Gelatt, The Fabulous Phonograph, 297.
xli Daniel Shannon, “Sales of Stereos Are Less Sound,” New York Times, May 23,
1982, https://www.nytimes.com/1982/05/23/business/sales-of-stereos-are-
less-sound-as-audio-industry-s-markets-slip.html. Accessed April 4, 2023.
xlii Aaron Perzanowski and Jason Schultz, The End of Ownership: Personal
Property in the Digital Economy (Cambridge, MA: MIT Press, 2016), 90–97.
xliii Dave Grohl, “A Statement from the Desk of Record Store Day 2015
Ambassador Dave Grohl,” Record Store Day, https://recordstoreday.com/Cus
tomPage/3882 . Accessed May 25, 2022.
xliv “U.S. Year-End Music Report for 2022,” Luminate, https://luminatedata.com/
reports/luminate-2022-u-s-year-end-report/. Accessed April 5, 2023.
xlv Keith Caulfield, “1 of Every 25 Vinyl Albums Sold in U.S. in 2022 Was by
Taylor Swift,” Billboard, January 11, 2023, https://www.billboard.com/pro/tay
lor-swift-vinyl-albums-1-of-every-25-sold/. Accessed January 31, 2023.
xlvi Shelton, “Happy Tunes on Cash Registers.”
xlvii RIAA figures.
xlviii Copyright law revision: hearings on H.R. 2223, before the Subcommittee on
Courts, Civil Liberties, and the Administration of Justice of the Committee
on the Judiciary, House of Representatives, Ninety-fourth Congress, 1976,
https://archive.org/details/copyrightlawrevi03unit. Accessed May 25, 2022.
420 References
xlix “SME and WMG the biggest market share winners in 2021,” Music &
Copyright’s Blog, April 5, 2022, https://musicandcopyright.wordpress.com/
2022/04/05/sme-and-wmg-t he-biggest-market-share-winners-in-2021/ .
Accessed May 25, 2022.
l Barbara Ringer, The Unauthorized Duplication of Sound Recordings
(Copyright Law Revision Study No. 26) (Washington, DC: U.S. Copyright
Office, 1961); Subcommittee on Courts, Civil Liberties, and the Administration
of Justice of the Committee on the Judiciary, House of Representatives,
95th Congress, Performance Rights in Sound Recordings; Oman, Ralph
(Register of Copyrights), Report on Copyright Implications of Digital Audio
Transmission Services (Washington, DC: U.S. Copyright Office, 1991).
li Performance Rights in Sound Recordings.
lii Sanjek, American Popular Music in the 20th Century, 144.
liii Sanjek, American Popular Music in the 20th Century, 142.
liv Sanjek, American Popular Music in the 20th Century, 170.
lv Stanley Green, Jukebox Piracy. The Atlantic Monthly, April 1962, 136.
lvi Sanjek, American Popular Music in the 20th Century, 229.
lvii “History of the Jukebox License Office,” Jukebox Licensing Office, http://www.
jukeboxlicense.org/history.htm. Accessed April 29, 2022.
Chapter 5
i Friedrich Engel, Peter Hammar, and Richard L. Hess, “A Selected History of
Magnetic Recording,” https://www.richardhess.com/tape/history/Engel_
Hammar—Magnetic_Tape_History.pdf, 2–3. Accessed April 29, 2022.
ii Toby Mountain, “The Birth of Stereo Recording,” Northeastern Digital,
https://www.northeasterndigital.com/post/the-birth-of-stereo-recording.
Accessed April 29, 2022.
iii Engel et al., “Magnetic Recording,” 4–5.
iv Engel et al., “Magnetic Recording,” 8.
v “AMPEX Model 200/200A Tape Machine,” History of Recording, https://
www.historyofrecording.com/AMPEX_Model_200.html. Accessed May
25, 2022; John Leslie and Ross Snyder, “History of The Early Days of Ampex
Corporation,” AES Historical Committee, December 17, 2010, https://www.
aes.org/aeshc/docs/company.histories/ampex/leslie_snyder_early-days-of-
ampex.pdf. Accessed May 25, 2022.
vi Andre Millard, America on Record: A History of Recorded Sound
(Cambridge: Cambridge University Press, 2005), 201–8.
vii Maxine Roth, “The Story of Music on Tape,” Tape Recording (January
1968), 44–48.
viii C. J. LeBel, “Tape or Disc?” High Fidelity (October 1959), 56+.
ix Brad Hardisty, “The Reel History of Analog Tape Recording,” Performer, May
4, 2015, https://performermag.com/home-recording/the-reel-history-of-ana
log-tape-recording/. Accessed April 29, 2022.
x David L. Morton, Sound Recording: The Life Story of a Technology
(Baltimore: Johns Hopkins University Press, 2004), 146.
xi Roth, “The Story of Music on Tape.”
References 421
xii Adrian Wu, “Open Reel Tape— The Ultimate Analog Source?,” Copper
Magazine, https://www.psaudio.com/copper/article/open-reel-tape-the-
ultimate-analog-source/. Accessed April 29, 2022; David Sarser, “Tapes, Disks,
and Coexistence.” High Fidelity (March 1955), 44+.
xiii “Audio Vendor Brochure and Manual,” Cousino Inc., http://lcweb2.loc.
gov/master/mbrs/recording_preser vation/manua ls/C ousino%20Univer
sal%20Audio%20Vendor.pdf. Accessed May 15, 2022.
xiv Eric D. Daniel, C. Dennis Mee, and Mark H. Clark, Magnetic Recording: The
First 100 Years (Piscataway, NJ: IEEE Press, 1999), 98–99.
xv Morton, Sound Recording, 158.
xvi Daniel et al., Magnetic Recording, 99.
xvii Morton, Sound Recording, 160.
xviii Stereo Review, HiFi/Stereo Review’s Tape Recorder Annual 1968.
xix Neil Genzlinger, “Lou Ottens, Father of Countless Mixtapes, Is Dead at 94,”
The New York Times, March 11, 2021, https://www.nytimes.com/2021/03/11/
arts/music/lou-ottens-dead.html. Accessed April 4, 2023.
xx Daniel et al., Magnetic Recording, 102.
xxi John Komurki, Cassette Cultures: Past and Present of a Musical Icon
(Salenstein: Benteli, 2019), 15.
xxii Stereo Review, HiFi/Stereo Review’s Tape Recorder Annual 1968.
xxiii Daniel et al., Magnetic Recording, 103.
xxiv Robert Angus and Norman Eisenberg, “Are Cassettes Here to Stay?,” High
Fidelity (July 1, 1969), 46–53.
xxv Julian Hirsch, “Julian Hirsch Hits the Cassette Decks.” Stereo Review
(November 1970), 56–69.
xxvi Julian Hirsch, “Advent 201 Cassette Tape Deck.” Stereo Review (October
1971), 40–42.
xxvii Stereo Review, Stereo Review’s Tape Recording and Buying Guide 1974.
xxviii Sarser, “Tapes, Disks, and Coexistence.”
xxix Roth, “The Story of Music on Tape.”
xxx Stereo Review, HiFi/Stereo Review’s Tape Recorder Annual 1968, 31.
xxxi Arthur J. Zuckerman, “Copy-Cat Tape Decks,” Popular Mechanics (October
1983), 94+.
xxxii Alan Sugar, What You See Is What You Get: My Autobiography
(London: Macmillan, 2010), 205–8.
xxxiii Clinton Heylin, Bootleg: The Secret History of the Other Recording Industry
(New York: St. Martin’s Griffin, 1994), 238.
xxxiv Richard Harrington, “The Record Industry Goes To War On Home Taping,”
Washington Post, June 15, 1980.
xxxv Stereo Review, Stereo Review Annual Product Guide, 1982–1986.
xxxvi Harrington, “Home Taping”; Heylin, Bootleg, 232–33.
xxxvii Ralph Blumenthal, “Goody Chain and Ex-Official Are Sentenced in Tapes
Case,” New York Times, November 6, 1982, 29.
xxxviii Bill Holland, “Audio- Only Home Taping Bill Readied on Senate Side,”
Billboard, September 7, 1985, 1+.
xxxix Barry Miles, Zappa: A Biography (New York: Grove Press, 2004), 332–38;
Harrington, “Home Taping.”
xl Schwann Publications, Schwann Record and Tape Guide, November 1971.
422 References
xli Schwann Publications, Schwann Record and Tape Guide, 1971–1978.
xlii “Finley to Use 4 and 8 Track Units,” Billboard, July 31, 1965, 3.
xliii “ITA Takes Giant Steps to Push Quality Tape,” Billboard, May 20, 1972, 37.
Verna, Paul, “ITA at 25,” Billboard, March 11, 1995, 73.
xliv James F. Peltz, “Cetec Gauss Bets Standard Tape Can Survive Digital Threat,”
Los Angeles Times, August 11, 1987, https://www.latimes.com/archives/la-
xpm-1987-09-01-fi-5454-story.html. Accessed April 4, 2023.
xlv Jon Pareles, “Record-It-Yourself Music on Cassette,” New York Times, May 11,
1987, https://www.nytimes.com/1987/05/11/arts/record-it-yourself-music-
on-cassette.html. Accessed April 4, 2023.
xlvi Simon Reynolds, Rip It Up and Start Again: Postpunk 1978– 1984
(London: Penguin, 2006), 196.
xlvii Amanda Sewell, “How Copyright Affected the Musical Style and Critical
Reception of Sample-Based Hip-Hop,” Journal of Popular Music Studies 26,
nos. 2–3 (2014): 295–320, 298, quoting producer Vinroc.
xlviii Sewell, “Sample-Based Hip-Hop,” 302–3.
xlix Jeff Chang, Can’t Stop Won’t Stop: A History of the Hip-Hop Generation
(London: Picador, 2005), 387–91.
l Millard, America on Record, 211–12.
li Shelton, “Happy Tunes on Cash Registers.”
lii Sam Goody ads in the New York Daily News, 1973–1981.
liii Chris Willman, “Adele’s ‘30’ Has the Biggest Bow of 2021 With 839,000 Album
Units,” Variety, November 28, 2021, https://variety.com/2021/music/news/
adele-30-biggest-number-one-album-year-1235120896/. Accessed April
4, 2023.
liv See for example Iain Taylor, “Audio Cassettes: Despite Being ‘a Bit Rubbish’, Sales
Have Doubled During the Pandemic—Here’s Why,” The Conversation, https://
theconversation.com/audio-cassettes-despite-being-a-bit-rubbish-sales-have-
doubled-during-the-pandemic-heres-why-157097. Accessed April 29, 2022.
lv See generally Komurki, Cassette Cultures.
lvi Komurki, Cassette Cultures, 69–75.
Chapter 6
i Jack Banks, Monopoly: MTV’s Quest to Control the Music (Boulder,
CO: Westview Press, 1996), 41.
ii William Ruhlmann, “Your Hit Parade: 1957 Review,” All Music Guide, https://
www.allmusic.com/album/your-hit-parade-1957-mw0000884963. Accessed
May 4, 2022.
iii Stephanie Nolasco, “Elvis Presley’s First Appearance on the Ed Sullivan Show
Remembered 64 Years Later,” Foxnews.com, September 9, 2020, https://www.
foxnews.com/entertainment/elvis-presley-ed-sullivan-show-anniversary.
Accessed May 4, 2022.
iv Banks, Monopoly, 61–65.
v Robert Fontenot, “The History of American Bandstand,” Liveabout.com, June
12, 2019, https://www.liveabout.com/american-bandstand-important-eve
nts-timeline-2523794. Accessed May 4, 2022.
References 423
vi Craig Marks and Rob Tannenbaum, I Want My MTV (New York: Dutton,
2011), 20.
vii “The Beatles,” EdSullivan.com, https://www.edsullivan.com/artists/the-beat
les/. Accessed May 4, 2022.
viii Bill Roedy and David Fisher, What Makes Business Rock (New York: Wiley,
2011), 42.
ix R. Serge Denisoff, Inside MTV (Milton Park: Routledge, 1988), 73.
x Denisoff, Inside MTV, 45–46.
xi Marks and Tannenbaum, I Want My MTV, 16.
xii Banks, Monopoly, 67.
xiii Marks and Tannenbaum, I Want My MTV, 27.
xiv Banks, Monopoly, 36.
xv Andrew Goodwin, Dancing in the Distraction Factory (Minneapolis: University
of Minnesota Press, 1992), 58.
xvi Denisoff, Inside MTV, 95.
xvii Nick Rhodes, interview with the authors, November 30, 2020.
xviii Banks, Monopoly, 36.
xix Marks and Tannenbaum, I Want My MTV, 74, 102.
xx Jordan Rost (Former Vice President of Research, Warner Amex Satellite
Entertainment), interview with the author, January 13, 2021.
xxi Marks and Tannenbaum, I Want My MTV, 130.
xxii Banks, Monopoly, 78.
xxiii “March 1982: ‘I Want My MTV!’ Campaign Launched,” Totally80s.com,
https://www.totally80s.com/article/march-1982-i-want-my-mtv-campaign-
launched. Accessed May 4, 2022.
xxiv Banks, Monopoly, 38.
xxv Marks and Tannenbaum, I Want My MTV, 80.
xxvi Greg Prato, “David Coverdale Tells the Story behind Whitesnake’s Iconic ‘Here
I Go Again’ Video,” Consequenceofsound.com, March 13, 2019, https://cons
equence.net/2019/03/david-coverdale-whitesnake-here-i-go-again-video/.
Accessed May 4, 2022.
xxvii Jerry Casale, interview with the authors, January 10, 2021.
xxviii Marks and Tannenbaum, I Want My MTV, 125.
xxix Marks and Tannenbaum, I Want My MTV, XLIV.
xxx David Benjamin, interview with the authors, January 28, 2021.
xxxi Jim Impoco, “The Beatles Suck. Yeah, We Said That,” Newsweek, February
3, 2014, https://www.newsweek.com/beatles-suck-yeah-we-said-227748.
Accessed May 4, 2022.
xxxii Denisoff, Inside MTV, 37.
xxxiii Goodwin, Dancing in the Distraction Factory, 54.
xxxiv Marks and Tannenbaum, I Want My MTV, 242.
xxxv Denisoff, Inside MTV, 157–58.
xxxvi Banks, Monopoly, 101.
xxxvii Marks and Tannenbaum, I Want My MTV, 243.
xxxviii Banks, Monopoly, 84.
xxxix Michael A. Hiltzik, “Viacom to Buy MTV and Showtime in Deal Worth $667.5
Million,” Los Angeles Times, August 27, 1985, https://www.latimes.com/archi
ves/la-xpm-1985-08-27-fi-25404-story.html. Accessed April 3, 2023.
424 References
xl Roedy and Fisher, What Makes Business Rock, 44.
xli Roedy and Fisher, What Makes Business Rock, 40.
xlii Marks and Tannenbaum, I Want My MTV, 564.
xliii Bobby Owsinski, “The Streaming Performance Royalty Explained,” Music 3.0
Blog, November 13, 2019, https://music3point0.com/2016/05/19/streaming-
performance-royalty/#ixzz6souR0TTQ. Accessed May 4, 2022.
xliv “‘Start Me Up’: the $3 Million Anthem That Launched Microsoft’s Windows
95,” 4As, https://www.aaaa.org/timeline-event/start-3-million-anthem-launc
hed-microsofts-windows-95/?cn-reloaded=1. Accessed May 4, 2022.
xlv Banks, Monopoly, 106–8.
xlvi Roedy and Fisher, What Makes Business Rock, 53, 134.
xlvii Roedy and Fisher, What Makes Business Rock, 161.
xlviii Chris Molanphy, “A Deal with the TV Gods,” Hit Parade (podcast), June 18,
2022, https://slate.com/podcasts/hit-parade/2022/06/television-is-a-hitmak
ing-jukebox. Accessed July 5, 2022.
xlix Denisoff, Inside MTV, 117.
Chapter 7
i Kees A. Immink, “The Compact Disc Story,” Journal of the Audio Engineering
Society, 46, no. 5 (May 1998): 458–465.
ii Kees A. Immink, “Shannon, Beethoven and the Compact Disc.” IEEE
Information Theory Society Newsletter (December 2007), 42–46.
iii Marc Finer, interview with the authors, December 10, 2020.
iv Kieran Prendiville, “Tomorrow’s World: The Compact Disc,” BBC, February
24, 2015, https://youtu.be/bMp1pSVxoqw. Accessed May 23, 2022.
v J. B. H. Peek and J. P. Sinjou, “The CD System as Standardized by Philips
and Sony,” in Origins and Successors of the Compact Disc: Contributions of
Philips to Optical Storage, ed. J. B. H. Peek, J. W. M. Bergmans, J. A. M. M. van
Haaren, Frank Toolenaar, and S. G. Stan (New York: Springer, 2009), 53–136.
vi “Portable Music Players Enter the Spin Zone,” Wired, September 10, 2009,
https://www.wired.com/2009/09/1001first-cd-players/. Accessed May
23. 2022.
vii “Play/Pause: A Look at 30 Years of the Compact Disc,” Business Week, October
1, 2012, https://www.bloomberg.com/news/photo-essays/2012-10-01/play-
pause-a-look-at-30-years-of-the-compact-disc. Accessed May 23, 2002.
viii Luke Dormehl, “38 Years ago, CDs Rewrote Our Relationship With Music and
Primed Us For 2020,” Digital Trends, October 12, 2020, https://www.digita
ltrends.com/features/how-cds-prepared-us-for-the-future/. Accessed May
23, 2022.
ix Edward Rasen, “Compact Discs: Sound of the Future,” Spin (May 1985), 28–32.
x Steve Knopper, Appetite for Self-Destruction (Seattle: CreateSpace Independent
Publishing Platform, 2017), 41.
xi Ryan Waniata, “The Life and Times of the Late, Great CD,” Digital Trends,
February 7, 2018, https://www.digitaltrends.com/music/the-history-of-the-
cds-rise-and-fall/. Accessed May 23.2022.
xii Milner, Perfecting Sound Forever, 211–19.
References 425
xiii Kara Swisher, “Circuit City Plugs into Visions of an Electronics Dream World,”
Washington Post, February 1, 1993, https://www.washingtonpost.com/arch
ive/business/1993/02/01/circuit-city-plugs-in-to-visions-of-an-electron
ics-dream-world/2f67d9fd-c21c-4931-99d3-1c3a13c8db43/. Accessed May
23, 2022.
xiv Bob Harvilla, “How Soundscan Changed Everything We Knew About Popular
Music,” The Ringer, May 25, 2021, https://www.theringer.com/music/2021/
5/25/22452539/soundscan-billboard-charts-streaming-numbers. Accessed
May 23, 2022.
xv Dan Margolis, “Music to a Customers Ears: CD Buyers Can Hear Them First
in Stores,” Los Angeles Times, June 14, 1995, https://www.latimes.com/archi
ves/la-xpm-1995-06-14-fi-13105-story.html. Accessed April 4, 2023.
xvi Knopper, Appetite for Self-Destruction, 36–40.
xvii Jon Pareles, “Now on CDs, First 4 Beatles Albums,” New York Times, February,
25, 2002, https://www.nytimes.com/1987/02/25/arts/now-on-cd-s-first-4-
beatles-albums.html. Accessed May 23, 2022.
xviii Donald Passman, All You Need to Know About the Music Business, 10th Edition
(New York: Simon & Schuster, 2019), 81–83.
xix Jay L. Frank, Futurehit, DNA (Pennsauken, NJ: Futurehit, Inc, 2009), 98–103.
xx Chris Molanphy, “Hit Parade: The War Against the Single,” Slate Podcasts,
September 29, 2017, https://chris.molanphy.com/hit-parade-the-great-war-
against-the-single-edition/. Accessed May 23, 2022.
xxi Knopper, Appetite for Self-Destruction, 42.
xxii Neil Young, “The CD and the Damage Done,” Harper’s Magazine, July
1992, 23–24.
xxiii Nick Rhodes, interview with the authors, November 30, 2020.
xxiv Jac Holzman, interview with the authors, September 4, 2020.
xxv Peter Asher, interview with the authors, September 27, 2020.
xxvi Albhy Galuten, interview with the authors, December 12, 2020.
xxvii “The Loudness Wars: Why Music Sounds Worse,” NPR, December 31, 2009,
https://www.npr.org/2009/12/31/122114058/the-loudness-wars-why-music-
sounds-worse. Accessed May 23, 2022.
xxviii Jane Birnbaum, “The Media Business: Without a Scratch, Used CD’s Rise
Again,” New York Times, September 6, 1993, https://www.nytimes.com/1993/
09/06/business/the-media-business-without-a-scratch-used-cds-rise-again.
html. Accessed May 23, 2022.
xxix Chris O’Malley, “A New Spin,” Time, August 24, 1998, https://content.time.
com/time/subscriber/article/0,33009,988955,00.html. Accessed May 23, 2022.
xxx Chris Nelson, “Music Piracy Shifts to Recordable CDs, Record Industry
Says,” MTV.com News, August 19, 1999, https://www.mtv.com/news/g5btet/
music-piracy-shifts-to-recordable-cds-record-industry-says/. Accessed April
5, 2023.
xxxi Eric Espe, “Music Industry Sees Threat in CD-copying Technology,” Silicon
Valley Business Journal, August 30, 1998, https://www.bizjournals.com/sanj
ose/stories/1998/08/31/story4.html. Accessed May 23, 2022.
xxxii Chris Nelson, “Recordable CDs Undercut Traditional Bootleg Market,” MTV.
com News, May 21, 1998, https://www.mtv.com/news/dsuz61/recordable-cds-
undercut-traditional-bootleg-market/. Accessed April 5, 2023.
426 References
xxxiii Barbara Fox Kraut, “The Audio Home Recording Act,” DePaul Journal of Art,
Technology, and Intellectual Property Law 2, no. 2 (Spring 1992): 38–40, https://
via.library.depaul.edu/cgi/viewcontent.cgi?article=1495&context=jatip.
xxxiv “About Piracy,” Record Industry Association of America, https://www.riaa.
com/resources-learning/about-piracy/. Accessed May 23, 2022.
xxxv Stephen Labaton, “5 Music Companies Settle Federal Case On CD Price-
Fixing,” New York Times, May 11, 2000, https://www.nytimes.com/2000/05/
11/business/5-music-companies-settle-federal-case-on-cd-price-fixing.html.
Accessed April 4, 2023.
Chapter 8
i “IBM PC AT,” Old-Computers.com, http://www.old-computers.com/museum/
computer.asp?st=1&c=185. Accessed May 23, 2022.
ii Paul Resinkoff, “A Proposal to Build Spotify (from 1982),” Digital Music News,
June 21, 2012, https://www.digitalmusicnews.com/2012/06/21/proposes/.
Accessed November 26, 2022.
iii Devon Schiff, “GO AEROSMITH: How ‘Head First’ Became the First Digitally
Downloadable Song 20 Years Ago Today,” Vice.com, June 27, 2014, https://www.
vice.com/en/article/6vapxr/go-aerosmith-how-head-first-became-the-first-
song-available-for-digital-download-20-years-ago-today. Accessed May 23, 2022.
iv Stephen Witt, How Music Got Free: A Story of Obsession and Invention
(London: Penguin Books, 2016), 6–17.
v Jonathan Sterne, MP3: The Meaning of a Format (Durham, NC: Duke
University Press, 2012), 117–129.
vi Witt, How Music Got Free, 27–28.
vii Ian Corbett, “What Data Compression Does to Your Music,” Sound On Sound,
April 2012, https://www.soundonsound.com/techniques/what-data-comp
ression-does-your-music. Accessed June 7, 2022.
viii Sterne, MP3, 232.
ix Seth Mnookin, “Universal’s CEO Once Called iPod Users Thieves: Now He’s
Giving Songs Away,” Wired, November 27, 2007, https://www.wired.com/
2007/11/mf-morris/. Accessed June 7, 2022.
x Po Bronson, “Rebootlegger,” Wired, July 1, 1998, https://www.wired.com/
1998/07/newmedia-4/. Accessed June 17,2022.
xi Jennifer Sullivan, “The Sound of Busic (sic),” Wired, February 26, 1999, https://
www.wired.com/1999/02/the-sound-of-busic/. Accessed June 7, 2022.
xii Barry Fox, “Out of the Bell Tower,” New Scientist, May 6, 1995, https://www.
computerhistory.org/timeline/memory-storage/. Accessed June 7, 2022.
xiii “Rio PMP300,” Wikipedia, https://en.wikipedia.org/wiki/Rio_PMP300.
Accessed June 17, 2022.
xiv Sam Costello, “History of the iPod: From the First iPod to the iPod Classic,”
Lifewire, January 12, 2020, https://www.lifewire.com/history-ipod-classic-
original-2000732. Accessed May 23, 2022.
xv Leander Kahney, “Straight Dope on the iPod’s Birth,” Wired, October 17,
2006, https://www.wired.com/2006/10/straight-dope-on-the-ipods-birth/.
Accessed June 27, 2022.
References 427
xvi Jennifer Sullivan, “More Popular Than Sex,” Wired, October 14, 1999, https://
www.wired.com/1999/10/more-popular-than-sex/. Accessed April 5, 2023.
xvii Matt Flint, “A File-Sharing Timeline: From the Creation of MP3 to the Trial of
The Pirate Bay,” The Takeaway, April 17, 2009, https://www.wnycstudios.org/
podcasts/takeaway/articles/10560-file-sharing-timeline-creation-mp3-trial-
pirate-bay. Accessed October 15, 2022.
xviii “RIAA vs Diamond Multimedia Decision,” Cyber Harvard, https://cyber.harv
ard.edu/property00/MP3/rio.html. Accessed May 23, 2022.
xix John Borland, “Metallica Fingers 335,435 Napster Users,” CNET, January 2,
2002, https://www.cnet.com/tech/services-and-software/metallica-fingers-
335435-napster-users/. Accessed May 23, 2022.
xx John Borland, “Napster Boots Dr. Dre Fans From Service,” CNET, January 2,
2002, https://www.cnet.com/news/napster-boots-dr-dre-fans-from-service/.
Accessed June 7, 2022.
xxi Borland, John, “Judges Issues Injunction Against Napster,” CNET, January 2,
2002, https://www.cnet.com/tech/services-and-software/judge-issues-injunct
ion-against-napster/. Accessed May 23, 2022.
xxii David Teather, “Napster Wins New Friend,” Guardian, November 1, 2000,
https://www.theguardian.com/technology/2000/nov/01/internetnews.busin
ess. Accessed June 7, 2022.
xxiii Mike Bebel, interview with the authors, June 27, 2022.
xxiv Brian Hiatt, “Napster Offers Labels $1 Billion in Licensing Fees,” MTV News,
February 20, 2001, https://www.mtv.com/news/5r4iyc/napster-offers-labels-
1-billion-in-licensing-fees. Accessed April 3, 2023.
xxv Ted Cohen, interview with the author, July 27, 2020.
xxvi Jim McDermott, “5 Reasons the Major Labels Didn’t Really Blow it With
Napster,” Hypebot, https://www.hypebot.com/hypebot/2015/05/five-reasons-
the-major-labels-didnt-blow-it-with-napster.html. Accessed June 7, 2022.
xxvii “Napster Settlement Offer Rejected,” CBS News, February 11, 2001, https://
www.cbsnews.com/news/napster-settlement-offer-rejected/. Accessed May
23, 2022.
xxviii Matt Richtel, “The Napster Decision: Appellate Judges Back Limitations on
Copying Music,” New York Times, February 13, 2001, https://www.nytimes.
com/2001/02/13/business/napster-decision-overv iew-appell ate-judges-
back-limitations-copying-music.html. Accessed June 17, 2022.
xxix T. S. Spangler, “Morpheus, Other File-Sharing Services Sued by RIAA,”
ExtremeTech, October 3, 2001, https://www.extremetech.com/extreme/
58270-morpheus-other-filesharing-services-sued-by-riaa. Accessed June
17, 2022.
xxx Eric Schumacher-Rasmussen, “Record Industry Sues Morpheus and Other
Decentralized File Sharing Services,” MTV News, October 3, 2001, https://
www.mtv.com/news/gpun33/record-industry-sues-morpheus-and-other-de-
centralized-file-sharing-services/. Accessed April 4, 2023.
xxxi Matt Richtel, “File Sharing Sites Found Not Liable For Infringement,” New York
Times, August 20, 2004, https://www.nytimes.com/2004/08/20/business/tec
hnology-file-sharing-sites-found-not-liable-for-infringement.html. Accessed
April 4, 2023.
428 References
xxxii “Web Music Sharers Under Attack,” CBS News, July 14, 2003, https://www.
cbsnews.com/news/web-music-sharers-under-attack/. Accessed June 7, 2022.
xxxiii “Public Knowledge Responds to RIAA’s plans to sue P2P File Traders,” Public
Knowledge press release, September 7, 2004, https://www.publicknowledge.
org/press-release/public-knowledge-responds-to-riaas-plans-to-sue-p2p-
file-traders/. Accessed June 7, 2022.
xxxiv Gil Kaufman, “File- Sharing Networks Can Be Liable for Copyright
Infringement, Supreme Court Rules,” MTV News, June 27, 2005, https://www.
mtv.com/news/n3nkvo/file-sharing-networks-can-be-liable-for-copyright-
infringements-supreme-court-rules/. Accessed April 3, 2023.
xxxv Josh Halliday, “Limewire Shut Down by Federal Court,” Guardian, October
27, 2010, https://www.theguardian.com/technology/2010/oct/27/limewire-
shut-down. Accessed June 27, 2022.
xxxvi Fred Goodman, “MP3 Technology Poised to Redefine Music Industry,” Rolling
Stone, March 9, 1999, https://www.rollingstone.com/music/music-news/
mp3-technology-poised-to-redefi ne-music-industry-97515/. Accessed June
17, 2022.
xxxvii Jodi Mardesich, “How the Internet Hits Big Music,” CNN Money, May 10,
1999, https://money.cnn.com/magazines/fortune/fortune_archive/1999/05/
10/259548/index.html. Accessed June 7, 2022.
xxxviii Brett Atwood, “Capitol to Sell Digital Singles,” Billboard, September 13, 1997.
Accessed June 27, 2022.
xxxix Jon Pareles, “Trying to Get in Tune With the Digital Age; Recording Industry
Seeks a Standard For Distributing Music on the Web,” New York Times,
February 1,1999, https://www.nytimes.com/1999/02/01/business/trying-get-
tune-with-digital-age-recording-industry-seeks-standard-for.html. Accessed
June 17, 2022.
xl Goodman, “MP3 Technology Poised to Redefine Music Industry.”
xli Eric Scheirer, “The End of SDMI,” MP3.com, October 15, 1999, https://web.
archive.org/web/20000229055832/http://w ww.mp3.com/news/394.html.
Accessed June 7, 2022.
xlii “MusicNet Launches Battle for Fans,” BBC, December 4, 2001, http://news.
bbc.co.uk/2/hi/entertainment/1691108.stm. Accessed June 7, 2022.
xliii Dan Tynan, “The 25 Worst Tech Products of All-Time,” PC World, May 26,
2006, https://www.pcworld.com/article/535838/worst_products_ever.html.
Accessed June 17, 2022.
xliv Matt Richtel, “Plans to Sell Music on the Internet Raise Antitrust Concerns,”
New York Times, August 7, 2001, https://www.nytimes.com/2001/08/07/busin
ess/technology-plans-to-sell-music-on-the-internet-raise-antitrust-conce
rns.html/. Accessed June 27, 2022.
xlv Steve Knopper, “iTunes’ 10th Anniversary: How Steve Jobs Turned the
Industry Upside Down,” Rolling Stone, April 26, 2013, https://www.rollingst
one.com/culture/culture-news/itunes-10th-anniversary-how-steve-jobs-tur
ned-the-industry-upside-down-68985/. Accessed June 27, 2022
xlvi Nathan Ingraham, “iTunes Store at 10: How Apple Built a Digital Media
Juggernaut,” The Verge, April 26, 2013, https://www.theverge.com/2013/4/
26/4265172/itunes-store-at-10-how-apple-built-a-digital-media-juggernaut.
Accessed June 7, 2022.
References 429
xlvii Paul Vidich, interview with the authors, December 7, 2020.
xlviii Jack Denton, “A Requiem for the Microsoft Zune,” Fast Company, June 15,
2022, https://www.fastcompany.com/90761088/a-requiem-for-the-micros
oft-zune-the-little-gadget-that-couldnt. Accessed June 27, 2022.
xlix Saul Hansell, “Amazon Launches a Music Store, Not a Service,” New York
Times, September 25, 2007, https://bits.blogs.nytimes.com/2007/09/25/ama
zon-launches-a-music-store-not-a-service. Accessed June 7, 2022.
l Steve Jobs, “Thoughts on Music,” Apple, February 6, 2007, https://web.archive.
org/web/20071223160841/https://www.apple.com/hotnews/thoughtsonmu
sic/. Accessed June 7, 2022.
li Michael Smith and Rahul Telang, “Assessing the Academic Literature
Regarding the Impact of Media Piracy on Sales,” Carnegie-Mellon
University,August 19, 2012, Available at http://ssrn.com/abstract=
2132153. Accessed May 18, 2023.
lii Jay Frank, Futurehit.DNA (New York: Futurehit, Inc, 2009), 118–120.
liii Maria Aspan, “For Shakira, First Came the Album, Then Came the Single,”
New York Times, June 6, 2006, https://www.nytimes.com/2006/06/12/busin
ess/media/12shakira.html. Accessed April 4, 2023.
liv “Ace of Base to Release New Album Worldwide,” The Local, February 24,
2015, https://www.thelocal.se/20150224/ace-of-base-to-release-new-album.
Accessed June 7, 2022.
lv Yinka Adegoke, “Labels Bet on Flexibility With iTunes New Pricing,” Reuters,
April 8, 2009, https://www.reuters.com/article/us-apple-itunes/labels-bet-
on-flexibility-with-itunes-new-pricing-idUSN0641102320090408. Accessed
June 7, 2022.
lvi “iTunes Store and DRM-Free Music: What You Need to Know,” Macworld,
January 7, 2009, https://www.macworld.com/article/1138000/drm-faq.html.
Accessed June 27, 2022.
lvii Adrian Covert, “A Decade of iTunes Singles Killed the Music Industry,” CNN
Business, April 25, 2013, https://money.cnn.com/2013/04/25/technology/itu
nes-music-decline/index.html. Accessed June 7, 2022.
lviii Michaels. Sean, “Most Music Didn’t Sell a Single Copy in 2008,” Guardian,
December 23, 2008, https://www.theguardian.com/music/2008/dec/23/
music-sell-sales. Accessed June 7, 2022.
lix Helienne Lindvall, “Behind the Music: Is the Long Tail a Myth,” Guardian,
January 8, 2009, https://www.theguardian.com/music/musicblog/2009/jan/
08/long-tail-myth-download. Accessed June 7, 2022.
lx “Madonna Swears at Music Pirates,” BBC News, April 22, 2003, http://news.
bbc.co.uk/2/hi/technology/2962475.stm. Accessed June 7, 2022.
lxi Goodman, “MP3 Technology Poised to Redefine Music Industry.”
lxii Bob Tarantino, “The Ninth Circuit’s Eminem License vs Sale Decision,”
Entertainment & Media Law Signal, September 9, 2010, http://www.entertai
nmentmedialawsignal.com/the-ninth-circuits-eminem-license-vs-sale-decis
ion/. Accessed June 7, 2022.
lxiii Marc Hogan, “Universal Settles Influential Eminem Digital- Revenues
Lawsuit,” Spin, October 31, 2012, https://www.spin.com/2012/10/universal-
settles-influential-eminem-digital-revenue-lawsuit/. Accessed June 7, 2022.
lxiv Nick Rhodes, interview with the authors, November 30, 2020.
430 References
lxv Lydia Warren, “Steve Jobs Listened to Vinyl at Home—Because it Sounded
Better Than His iPod,” Daily Mail, January 31, 2012, https://www.dailymail.
co.uk/news/article-2094590/Steve-Jobs-listened-vinyl-home-iPod.html.
Accessed June 7, 2022.
lxvi Bootie Cosgrove-Mather, “Poll: Young Say File-Sharing OK,” CBS News,
September 18, 2003, https://www.cbsnews.com/news/poll-young-say-file-
sharing-ok/. Accessed June 7, 2022.
lxvii Andrew Everard, “Never Mind the Lack of MP3 Quality, Feel the Lack of MP3
Breadth,” What Hi-Fi?, April 1, 2009, https://www.whathifi.com/news/usa-
never-mind-lack-mp3-quality-feel-lack-mp3-breadth. Accessed June 7, 2022.
lxviii “Where Do Music Collections Come From?” American Assembly, October
15, 2012, http://piracy.americanassembly.org/where-do-music-collections-
come-from/. Accessed June 7, 2022.
lxix Goodman, “MP3 Technology Poised to Redefine Music Industry.”
Chapter 9
i Maureen Farrell and Anne Steele, “Tencent Music Files for U.S. IPO,” Wall
Street Journal, October 2, 2018, https://www.wsj.com/articles/tencent-music-
files-for-u-s-ipo-1538496869. Accessed April 4, 2023.
ii Maureen Farrell, “Tencent Music Rises in Trading Debut,” Wall Street Journal,
December 12, 2018, https://www.wsj.com/articles/tencent-music-opens-hig
her-in-trading-debut-11544632668. Accessed April 4, 2023.
iii Glenn Peoples, “Cloud Village Music Subscribers Grew 51%, Revenues Up
38.6% in Q1,” Billboard, May 24, 2022, https://www.billboard.com/pro/cloud-
village-q1-2022-earnings/. Accessed April 4, 2023.
iv Andre Paine, “Tencent Music Paying Users Up 52% in Q2,” Music Week,
August 11, 2020, https://www.musicweek.com/digital/read/tencent-music-
paying-users-up-52-in-q2/080734. Accessed June 5, 2022.
v Dylan Smith,” Tencent Music Subscribers Grow 50% to 42.7 Million—Wall
Street Isn’t Impressed,” Digital Music News, May 12, 2020, https://www.digit
almusicnews.com/2020/05/12/tencent-music-subscribers-growth/. Accessed
June 7, 2022.
vi Shawn Knight, “Spotify Hits 205 Million Premium Subscribers, but Financial
Losses Swell,” Techspot, January 31, 2023. https://www.techspot.com/news/
97444-spotify-hits-205-million-premium-subscribers-but-financial.html.
Accessed January 31, 2023.
vii Sven Carlsson and Jona Leijonhufvud, The Spotify Play: How CEO and Founder
Daniel Ek Beat Apple, Google, and Amazon in the Race for Audio Dominance
(New York: Diversion Books, 2021), 248–256.
viii Ashley King, “For the First Time Ever, Annual Music Streams Top 1 Trillion—
in the US Alone,” Digital Music News, November 30, 2022, https://www.digit
almusicnews.com/2022/11/30/annual-music-streams-top-one-trillion-us-
alone. Accessed January 31, 2023.
ix Charles Mann, “The Hot New Bad Idea,” Inside Magazine, December, 2000.
x Chuck Melvin, “AT&T Creates ‘Ultimate Jukebox’ for Rock Hall,” Cleveland
Plain Dealer, April 1, 1998, 1F.
References 431
xi Jim Hu, “Pressplay Comes to Life after Long Wait,” CNET, April 14, 2002,
https://www.cnet.com/tech/services-and-software/pressplay-comes-to-life-
after-long-wait/. Accessed June 7, 2022.
xii Tim Ingham, “It’s Happened: 100,000 Tracks Are Now Being Uploaded
to Streaming Services Like Spotify Each Day,” Music Business Worldwide,
October 6, 2022, https://www.musicbusinessworldwide.com/its-happened-
100000-tracks-are-now-being-uploaded/. Accessed December 4, 2022.
xiii Melanie Uy, “3G vs. 4G Technology,” Lifewire, June 10, 2020, https://www.
lifewire.com/how-fast-are-4g-and-3g-internet-speeds-3974470. Accessed
June 5, 2022.
xiv “3G vs 4G,” Diffen, https://www.diff en.com/diff erence/3G_vs_4G. Accessed
June 5, 2022.
xv Matt Buchanan, “RIAA Tech Chief: DRM Not Dead, Will Become More
Powerful than You Can Possibly Imagine,” Gizmodo, May 8, 2008, https://gizm
odo.com/riaa-tech-chief-drm-not-dead-will-become-more-powerfu-388
648. Accessed June 5, 2022.
xvi Andrea Bossi, “There Were More Than One Trillion Music Streams in
U.S. during 2019,” Forbes, January 10, 2010, https://www.forbes.com/sites/
andreabossi/2020/01/10/over-one-trillion-music-streams-in-us-during-
2019-record-high. Accessed June 5, 2022.
xvii “Streaming Forward 2020 Report,” Digital Media Association, https://dima.
org/streaming-forward-report/. Accessed June 5, 2022.
xviii “About Royalty Rates,” Trichordist, 2021, https://thetrichordist.com/category/
royalty-rates-2/. Accessed May 25, 2022.
xix Stuart Dredge, “Thom Yorke Calls Spotify ‘The Last Desperate Fart of a Dying
Corpse.’” Guardian, October 7, 2018, https://www.theguardian.com/technol
ogy/2013/oct/07/spotify-thom-yorke-dying-corpse. Accessed June 13, 2022.
xx Paul Resnikoff, “Spotify Executive Calls Artist ‘Entitled’ for Requesting
Payment of One Penny Per Stream,” Digital Music News, June 29, 2021,
https://www.digitalmusicnews.com/2021/06/29/spotify-executive-entitled-
pay-penny-per-stream/. Accessed June 13, 2022.
xxi Shapiro, Ariel, “Warner Music Adopts SoundCloud’s User-centric Revenue
Model.” Verge, July 21, 2022, https://www.theverge.com/2022/7/21/23272
548/warner-music-soundcloud-user-centric-model-spotify. Accessed July
22, 2022.
xxii Ben Homewood, “UMG and Tidal Plot New Streaming Model to Benefit
Artists and Fans,” Music Week, January 31, 2023, https://www.musicweek.
com/digital/read/umg-and-tidal-plot-new-streaming-model-to-benefit-arti
sts-and-fans. Accessed January 31, 2023.
xxiii “Loud & Clear,” Spotify, https://loudandclear.byspotify.com/?question=user-
centric-model. Accessed June 13, 2022.
xxiv Karl Matchett, “Barcelona Agree £235m Camp Nou and Kit Sponsorship Deal
with Spotify,” Yahoo News, March 15, 2022, https://news.yahoo.com/barcel
ona-agree-235m-camp-nou-204133143.html. Accessed June 7, 2022.
xxv Tim Ingham, “‘Fake Artists’ Have Billions of Streams on Spotify. Is Sony Now
Playing the Service at Its Own Game?” Rolling Stone, May 15, 2019, https://
www.rolli n gst one.com/ pro/ f eatu res/ f ake- a rti s ts- h ave- billi ons- of- s tre
432 References
ams-on-spotify-is-sony-now-playing-the-service-at-its-own-game-834746/.
Accessed June 5, 2022.
xxvi “More to Love: Target and Michael Bublé Reveal Exclusive Bonus Tracks on
His New Album ‘To Be Loved,’” CNN, April 24, 2018, https://www.cnn.com/
2018/04/24/us/monkey-selfie-peta-appeal/index.html. Accessed June 5, 2022.
xxvii Tim Ingham, “Yes, Lucian Grainge Has Banned All Exclusives. No, It’s Not
All about Frank Ocean,” Music Business Worldwide, August 25, 2016, https://
www.musicbusinessworldwide.com/yes-lucian-grainge-banned-streaming-
exclusives-umg-no-not-frank-ocean/. Accessed June 5, 2022.
xxviii Ben Sisario, “A New Spotify Initiative Makes the Big Record Labels Nervous,”
New York Times, September 6, 2018, https://www.nytimes.com/2018/09/06/
business/media/spotify-music-industry-record-labels.html. Accessed June
5, 2022.
xxix Colin Stutz, “Spotify CEO Daniel Ek Says Direct Licensing With Artists
‘Doesn’t Make Us a Label,’” Billboard, July 26, 2018, https://www.billboard.
com/music/music-news/spotify-daniel-ek-direct-licensing-artists-not-label-
earnings-call-8467283/. Accessed June 5, 2022.
xxx Mark Mulligan, “Did July 1st 2019 Mark the End of Spotify’s Music Creator
Dream?” Midia Music Industry Blog, October 8, 2021,https://www.midiaresea
rch.com/blog/did-july-1st-2019-mark-the-end-of-spotif ys-music-creator-
dream. Accessed June 7, 2022.
xxxi Jon Russell, “Spotify Says It Paid $340M to Buy Gimlet and Anchor,”
Techcrunch, February 15, 2019, https://techcrunch.com/2019/02/14/spotify-
gimlet-anchor-340-million/. Accessed June 5, 2022.
xxxii Ashley Carman, “Spotify Acquires Another Podcast Network to Keep
Building Its Original Show Catalog,” The Verge, May 26, 2019, https://www.
theverge.com/2019/3/26/18282301/spot ify-podcasts-parcast-acquisition-
gimlet-media-anchor. Accessed June 5, 2022.
xxxiii Kif Leswing, “Apple Music Has Reportedly Passed Spotify in Paid Subscribers
in the US,” CNBC, April 5, 2019, https://www.cnbc.com/2019/04/05/apple-
music-has-reportedly-passed-spotify-in-paid-subscribers-in-the-us.html.
Accessed June 5, 2022.
xxxiv Manish Singh, “Google invests $4.5 billion in India’s Reliance Jio Platforms,”
TechCrunch, July 15, 2020, https://techcrunch.com/2020/07/15/google-inve
sts-4-5-billion-in-indias-reliance-jio-platforms/. Accessed May 23, 2022.
xxxv Sankalp Phirtiyal, “Google Expands Jio partnership with Indian Smartphone,
Cloud Tie-ups,” Reuters, June 24, 2021, https://www.reuters.com/technology/
google-says-cloud-partnership-with-indias-jio-boost-5g-plans-2021-06-24/.
Accessed June 13, 2022.
xxxvi Daniel Keyes, “Amazon Has Surpassed 150 million Prime Subscribers
Globally,” Business Insider, February 3, 2020, https://www.businessinsider.
com/amazon-surpasses-150-million-prime-subscribers-2020-2. Accessed
June 5, 2022.
xxxvii Marissa Matozzo, “Amazon Announced a New Perk for Prime Members—
Here’s How to Get It,” Yahoo Life, November 16, 2022, https://www.yahoo.
com/lifestyle/amazon-announced-perk-prime-members-140028030.html.
Accessed November 18, 2022.
References 433
xxxviii Eli Blumenthal, “Hulu and Spotify Partner to Bundle TV and Music Streaming
for $9.99 per Month,” USA Today, March 12, 2019, https://www.usatoday.
com/story/tech/talkingtech/2019/03/12/apple-looming-hulu-and-spotify-
partner-9-99-streaming-bundle/3138377002/. Accessed November 18, 2022.
xxxix Harry Guiness, “Spotify Free vs. Premium: Is it Worth Upgrading? Spotify
Free vs. Premium: Is it Worth Upgrading?,” How to Geek, April 27, 2018,
https://www.howtogeek.com/350288/spotify-free-vs-premium-is-it-worth-
upgrading/. Accessed June 5, 2022.
xl Carlsson and Leijonhufvud, The Spotify Play, 54–66.
xli Jacob Key (formerly VP Digital Strategy Europe, Warner Music Group), inter-
view with the author, March 15, 2021.
xlii Carlsson and Leijonhufvud, The Spotify Play, 100–104,172–73.
xliii “Where Is Spotify Available,” Spotify, https://support.spotify.com/us/article/
where-spotify-is-available/. Accessed June 5, 2022.
xliv “Apple Services Now Available in More Countries Around the World,” Apple,
April 21, 2020, https://www.apple.com/newsroom/2020/04/apple-services-
now-available-in-more-countries-around-the-world/. Accessed June 5, 2022.
xlv “IFPI Releases Engaging with Music 2021,” International Federation of the
Phonographic Industry, October 21, 2021, https://www.ifpi.org/ifpi-releases-
engaging-with-music-2021/. Accessed June 5, 2022.
xlvi Murray Stassen, “JioSaavn’s Paying Subcriber Base Has Quadrupled Since
Early 2019,” Music Business Worldwide, May 28, 2020, https://www.musicbu
sinessworldwide.com/jiosaavns-paying-subscriber-base-has-quadrupled-
since-early-2019/. Accessed June 5, 2022.
xlvii Stuart Dredge, “African Streaming service Boomplay Now Has 75m Users,”
Music Ally, June 3, 2020, https://musically.com/2020/06/03/african-stream
ing-service-boomplay-now-has-75m-users/. Accessed June 5, 2022.
xlviii Thomas Røssel and Camilla Louise Brandt, “Polaris Nordic: Digital Music in
the Nordics,” Polaris, January 2020, https://polarismusichub.com/wp-cont
ent/uploads/2020/05/Full_Report_Polaris_Nordic_Digital-Music-in-the-
Nordics-2020.pdf. Accessed June 5, 2022.
xlix Murray Stassen, “ Apple Music Just Raised Its Subscription Price to $10.99
in the U.S. Will Spotify be Next?” Music Business Worldwide, October 24,
2022, https://www.musicbusinessworldwide.com/apple-music-just-raised-
its-subscription-price-to-10-99-in-the-us-will-spotify-be-next1/. Accessed
October 25, 2022.
l Dylan Smith, “ Spotify CEO Daniel Ek Confirms 2023 Price Increase Plans
Following Apple Music Raises,” Digital Music News, October 27, 2022, https://
www.digit a lmu s icn e ws.com/ 2 022/ 1 0/ 2 7/ spot i fy- price- i ncre a se- 2 023/ .
Accessed October 27, 2022.
li Simon Cohen, “Amazon Music Unlimited Is Getting More Expensive in 2023,”
Digital Trends, January 20, 2023, https://www.digitaltrends.com/home-thea
ter/amazon-music-unlimited-february-2023-price-hike/. Accessed January
31, 2023.
lii International Federation of the Phonographic Industry, “Engaging with Music
2022,” https://www.ifpi.org/wp-content/uploads/2022/11/Engaging-with-
Music-2022_full-report-1.pdf/. Accessed November 28, 2022.
434 References
liii Eammon Forde, “Spotify’s Discover Weekly Has Driven 2.3bn Hours of
Listening,” Music Ally, July 20, 2020, https://musically.com/2020/07/10/spoti
fys-discover-weekly-has-driven-2-3bn-hours-of-listening/. Accessed June
5, 2022.
liv Luis Aguiar Wicht and Joel Waldfogel, “Platforms, Promotion, and Product
Discovery: Evidence from Spotify Playlists,” EU Science Hub, June 5, 2018,
https://joint-research-centre.ec.europa.eu/publications/platforms-promot
ion-and-product-discovery-evidence-spotify-playlists_en. Accessed June
5, 2022.
lv “Today’s Top Hits Is the World’s Destination for the Very Best in Music,”
Spotify: For the Record, June 24, 2021, https://newsroom.spotify.com/2021-06-
24/todays-top-hits-is-the-worlds-destination-for-the-very-best-in-music/.
Accessed June 13, 2022.
lvi Paul Resnikoff, “What Succeeds on Spotify’s Rap Caviar— A Statistical
Analysis,” Digital Music News, August 18, 2020, https://www.digitalmusicn
ews.com/2020/08/18/spotify-rapcaviar-analysis. Accessed June 13, 2022.
lvii Y. Heisler, “Steve Jobs Was Never a Fan of Subscription Music,” Engadget, May
19, 2014, https://www.engadget.com/2014-05-19-steve-jobs-was-never-a-
fan-of-subscription-music.html. Accessed June 13, 2022.
lviii Madeline Buxton, “Spotify’s New Free Version Is So Good You Might Cancel
Your Subscription,” Refinery 29, April 24, 2018, https://www.refi nery29.com/
en-us/2018/04/197232/spotify-for-free-upgrade. Accessed June 5, 2022.
lix Adrian Pope, “Clamp Down on Free Spotify? Now is Not the Time for Knee-
Jerk Decisions,” PIAS: The Independent Blog, Undated, https://www.piasgroup.
net/blog/clamp-down-on-free-spotify-now-is-not-the-time-for-knee-jerk-
decisions/. Accessed June 5, 2022.
lx Andre Paine, “Tencent Music Paying Users Up 52% in Q2.”
lxi Bobby Owsinski, “The Music Long Tail Might Not Be That Long After All,”
Music 3.0 Blog, November 13, 2019, https://music3point0.com/2019/11/13/
music-long-tail-rebuff ed/#ixzz6QtTtdmzJ. Accessed June 5, 2022.
lxii Glenn Peoples, “1 in Every 60 U.S. Streams Last Week Was Drake,” Billboard,
June 28, 2022, https://www.billboard.com/pro/drake-honestly-nevermind-
streams-marketshare/. Accessed June 29, 2022.
lxiii Tim Ingham, “Over 66% of All Music Listening in the US Is Now of Catalog Records
Rather Than New Releases,” Music Business Worldwide, July 13, 2021, https://www.
musicbusinessworldwide.com/over-66-of-all-music-listening-in-the-us-is-now-
of-catalog-records-rather-than-new-releases/. Accessed June 7, 2022.
lxiv Stephen Dowling, “The Most Iconic Long Intros,” BBC, April 8, 2019, https://
www.bbc.com/culture/article/20190407-the-most-iconic-long-song-intros-
in-pop-music. Accessed June 7, 2022.
lxv Elias Leight, “Think You Have a Hit? Make Sure It’s the First Song on Your
Album,” Rolling Stone, March 7, 2019. https://www.rollingstone.com/music/
music- featu res/ w hy- t he- f irst- s ong- on- t he- a lbum- is- t he- b est- 8 03 2 83/ .
Accessed June 7, 2022.
lxvi Zachary Mack, “How Streaming Affects the Lengths of Songs,” The Verge, May
28, 2019, https://www.theverge.com/2019/5/28/18642978/music-streaming-
spotify-song-length-distribution-production-switched-on-pop-vergecast-
interview. Accessed June 5, 2022.
References 435
lxvii Mark Mulligan, “How the DNA of a Hit Has Changed Over 20 Years,” Midia
Research, July 13, 2020, https://www.midiaresearch.com/blog/how-the-dna-
of-a-hit-has-changed-over-20-years. Accessed June 13, 2022.
lxviii Alexis Petridis, “Mark Ronson: ‘I Was Floundering. I Was Drinking Too Much
and Giving Orders,’” Guardian, February 22, 2019. https://www.theguardian.
com/music/2019/feb/22/mark-ronson-i-was-floundering-i-was-drinking-
too-much-and-giving-orders. Accessed June 13, 2022.
lxix Michael Tauberg, “Music Is Getting Shorter,” Medium, April 27, 2018, https://
michaeltauberg.medium.com/music-and-our-attention-spans-are-getting-
shorter-8be37b5c2d67. Accessed June 7, 2022.
lxx “Are Songs Getting Even Shorter in 2021?” Hit Songs Deconstructed, Undated,
https:// w ww.hitson g sde c ons t ruc t ed.com/ h sd_ w i re/ t he- s hort- g ame/ .
Accessed July 8, 2022.
lxxi Neil Shah, “Migos Album ‘Culture II’ Tops Charts but Sparks Debate Over Its
Length,” Wall Street Journal, February 5, 2018, https://www.wsj.com/articles/
migos-album-culture-ii-tops-charts-but-sparks-debate-over-its-length-151
7848564. Accessed June 7, 2022.
lxxii Erielle Sudario, “Spotify Changes Playlist Rules Post Pocket Gods’ Protest
Album,” We Got This Covered, May 23, 2022, https://wegotthiscovered.com/
music/spotify-changes-playlist-r ules-p ost-p ocket-gods-protest-a lbum/.
Accessed June 7, 2022.
lxxiii Leight, “Think You Have a Hit?”
lxxiv Jason Lipshutz, “RIP Lead Singles: Why Hip-Hop Titans Are Dropping
Full Albums All at Once,” Billboard, June 21, 2022, https://www.billboard.
com/pro/drake-future-album-release-strategy-hip-hop-singles/. Accessed
November 1, 2022.
lxxv Corinne Reichert, “Taylor Swift Album ‘Midnights’ Sweeps All 10 Top Spots
on Billboard Hot 100,” CNet, October 31, 2022, https://www.cnet.com/cult
ure/entertainment/taylor-swift-album-midnights-sweeps-all-10-top-spots-
on-billboard-hot-100/. Accessed November 1, 2022.
lxxvi John Gapper, “Taylor Swift Is Perfect at the Three-Minute Spotify Song,”
Financial Times, November 25, 2022, https://www.ft.com/content/afdd5bfe-
c5ae-411c-900f-8c5eab846944. Accessed November, 26, 2022.
lxxvii ”Official Charts Company Shakes Up Rules to Stop the Ed Sheeran Effect,”
Guardian, June 27, 2017, https://www.theguardian.com/music/2017/jun/27/
official-charts-company-shakes-up-rules-support-new-talent-ed-sheeran-eff
ect. Accessed November 1, 2022.
lxxviii Wicht and Waldfogel, “Platforms, Promotion, and Product Discovery.”
lxxix Neil Shah, “Didn’t Like That New Album? Another One Is Coming Before You
Know It,” Wall Street Journal, March 26, 2018, https://www.wsj.com/articles/
didnt-like-that-new-album-another-one-is-coming-before-you-know-it-152
2076320. Accessed June 7, 2022.
lxxx Ashley King, “German Pop Artist Mike Singer Uses ‘Stream-to-Unlock’ Trick
to Boost Streaming Algorithms,” Digital Music News, July 14, 2020, https://
www.digitalmusicne ws.com/2020/07/14/mike-singer-stream-to-unlo ck-
campaign/. Accessed June 7, 2022.
lxxxi “Spotify Flagship Playlist ¡Viva Latino! Hits 10 Million Followers,” Spotify: For
the Record, July 26, 2019, https://newsroom.spotify.com/2019-07-26/spot
436 References
ify-flagship-playlist-viva-latino-hits-10-million-followers/. Accessed June
5, 2022.
lxxxii “Popular Music Is More Collaborative Than Ever,” Economist, February 2,
2018, https://www.economist.com/graphic-detail/2018/02/02/popular-
music-is-more-collaborative-than-ever. Accessed June 5, 2022.
lxxxiii “Sorry (Justin Bieber Song),” Wikipedia, https://en.wikipedia.org/wiki/Sor
ry_(Justin_Bieber_song). Accessed June 5, 2022.
lxxxiv Stuart Dredge, “Spotify CEO Talks Covid-19, Artist Incomes and Podcasting,”
Music Ally, July 30, 2020, https://musically.com/2020/07/30/spotify-ceo-
talks-covid-19-artist-incomes-and-podcasting-interview/. Accessed June
7, 2022.
lxxxv Lake Schatz, “Spotify CEO to Artists: ‘You Can’t Record Music Once Every
Three to Four Years and Think That’s Going to Be Enough,’” Consequence of
Sound, August 1, 2020, https://consequence.net/2020/08/spotify-daniel-ek-
artist-recording-comments/. Accessed June 13, 2022.
lxxxvi Jay Frank, Futurehit.DNA (New York: Futurehit, Inc, 2009), 30–34, 105–110,
131–135.
lxxxvii Joann Pan, “Grooveshark Circumvents Mobile Bans by Launching an HTML5
Player,” Mashable, September 5, 2012, https://mashable.com/archive/groo
veshark-html5-player. Accessed June 7, 2022.
lxxxviii Peter Kafka, “Grooveshark, the Free Music Service That Used to Scare the
Labels Gives Up,” Vox, April 30, 2015, https://www.vox.com/2015/4/30/11562
064/grooveshark-the-free-music-service-that-used-to-scare-the-big-labels.
Accessed June 7, 2022.
lxxxix F-1 Registration, Spotify, March 31, 2018, https://www.sec.gov/Archives/
edgar/data/1639920/000119312518063434/d494294df1.htm#rom494294_7.
Accessed June 7, 2022.
xc Tim Ingham, “The Great Big Spotify Scam: Did a Bulgarian Playlister Swindle
Their Way to a Fortune on Streaming Service,” Music Business Worldwide,
November 20, 2018, https://www.musicbusinessworldwide.com/great-big-
spotify-scam-bulgarian-playlister-swindle-way-fortune-streaming-service/.
Accessed June 7, 2022.
xci Murray Stassen, “BTS Set Spotify Record with 20.9M Day One Plays, but
10M Nearly of Them Are Discounted on Platform’s Chart,” Music Business
Worldwide, May 24, 2021, https://www.musicbusinessworldwide.com/bts-
set-new-spotify-streaming-record-with-20-9m-day-one-plays-but-nearly-
10m-of-them-are-discounted-on-platforms-chart/. Accessed June 7, 2022.
xcii Leight, “Fake Streams Could Be Costing Artists $300 Million a Year,” Rolling
Stone, June 18, 2019, https://www.rollingstone.com/pro/features/fake-stre
ams-indie-labels-spotify-tidal-846641/. Accessed June 7, 2022.
xciii Peter Slattery, “Scammers Are Gaming Spotify by Faking Collaborations With
Famous Artists,” OneZero, June 1, 2020, https://onezero.medium.com/scamm
ers-are-gaming-spotify-by-faking-collaborations-with-famous-artists-42d12
7e370dc. Accessed June 7, 2022.
xciv Landon Groves, “Hate Your Favorite Band’s Latest Song? On Spotify, It
Might Be a fake,” Input Mag, July 17, 2020, https://www.inputmag.com/cult
ure/spotify-is-letting-scammers-rip-artists-off-in-plain-sight. Accessed
June 7, 2022.
References 437
xcv Bruce Houghton, “Spotify Cuts Off Playlist Payola Service SpotLister,”
Hypebot, March 19, 2018, https://www.hypebot.com/hypebot/2018/03/spot
ify-cuts-off-playlist-payola-service-spotlister.html. Accessed June 7, 2022.
xcvi Tatiana Cirisano, “Record Industry Groups Take Action Against Fake Music
Streams in Germany,” Billboard, August 24, 2020, https://www.billboard.com/
pro/ifpi-bvmi-take-action-against-fake-music-streams-germany/. Accessed
June 7, 2022.
xcvii Ed Christman, “Spotify and Publishing Group Reach $30 Million Settlement
Agreement Over Unpaid Royalties,” Billboard, March 17, 2016, https://www.
billboard.com/music/music-news/spotify-nmpa-publishing-30-million-set
tlement-unpaid-royalties-7263747/. Accessed June 7, 2022.
xcviii Sarah Jeong, “A $1.6 Billion Spotify Lawsuit is Based on a Law Made for Player
Pianos,” The Verge, March 14, 2018, https://www.theverge.com/2018/3/14/
17117160/spotify-mechanical-license-copyright-wixen-explainer. Accessed
June 7, 2022.
xcix Dani Deahl, “The Music Modernization Act Has Been Signed Into Law,”
The Verge, October 11, 2018, https://www.theverge.com/2018/10/11/17963
804/music-modernization-act-mma-copyrig ht-law-bill-labels-congress.
Accessed June 7, 2022.
c “About the MLC,” Mechanical Licensing Collective, https://www.themlc.com/
our-story. Accessed June 7, 2022.
ci “How Can YOU Play Your Part™?” Mechanical Licensing Collective, https://
www.themlc.com/play-your-part. Accessed July 17, 2022.
cii Natasha Lomas, “Apple Pay and iOS App Store Under Formal Antitrust Probe
in Europe,” Techcrunch, June 16, 2016, https://techcrunch.com/2020/06/16/
apple-p ay-and-ios-app-store-under-for mal-antitr ust-probe-in-europe/.
Accessed June 7, 2022.
ciii Jack Nicas and David McCabe, “Their Businesses Went Virtual. Then Apple
Wanted a Cut,” New York Times, July 28, 2020, https://www.nytimes.com/
2020/07/28/technology/apple-app-store-airbnb-classpass.html. Accessed
June 7, 2022.
Chapter 10
i John Horrigan, “Home Broadband Adoption 2006: Part 1. Broadband
Adoption in the United States,” Pew Research Center, May 28, 2006, https://
www.pewresearch.org/internet/2006/05/28/part-1-broadband-adoption-in-
the-united-states/. Accessed May 25, 2022.
ii Martyn Williams, “IMesh P-to-P provider gets Sony Music license,” PCWorld,
July 12, 2005, https://www.pcworld.idg.com.au/article/135515/imesh_p-to-
p_provider_gets_sony_music_license/. Accessed May 25, 2022.
iii John Borland, “Legal P2P opens for business,” CNET, September 26, 2006,
https://www.cnet.com/tech/home-entertainment/legal-p2p-opens-for-busin
ess/. Accessed May 25, 2022.
iv “Sony BMG Music Entertainment Signs Content License Agreement
with YouTube,” Sony Music press release, October 8, 2006, https://www.
sonymu s ic.com/ s onymu s ic/ s ony- bmg- music-e ntert ainme nt-signs-cont
438 References
ent-license-agreement-with-youtube/. Accessed May 25, 2022; Nate
Anderson, “YouTube signs up Universal, Sony BMG, CBS, and . . . Google?,”
Ars Technica, October 9, 2006, https://arstechnica.com/information-technol
ogy/2006/10/7935/. Accessed May 25, 2022.
v “EMI Music, Google and YouTube strike milestone partnership,” EMI press
release, May 31, 2007, https://web.archive.org/web/20070602142911/http:/
www.emigroup.com/Press/2007/press38.htm. Accessed May 25, 2022.
vi Andrew Ross Sorkin and Jeff Leeds, “Music Companies Grab a Share of the
YouTube Sale,” New York Times, October 19, 2006, https://www.nytimes.com/
2006/10/19/technology/music-companies-grab-a-share-of-the-youtube-sale.
html. Accessed May 25, 2002.
vii Michael Wesch, “YouTube Statistics,” Digital Ethnography @ Kansas State
University, March 18, 2008, https://web.archive.org/web/20140723075938/
http://mediatedcultures.net/thoughts/youtube-statistics/. Accessed May
25, 2022.
viii “Thanks a Billion!,” TikTok, September 7, 2021, https://newsroom.tiktok.com/
en-us/1-billion-people-on-tiktok. Accessed May 25, 2022.
ix “TikTok Users Worldwide (2020–2025),” Insider Intelligence, June 1, 2022,
https:// w ww.insi d eri ntel l ige n ce.com/ c ha r ts/ g lo b al- t ik t ok- u ser- s tats/ .
Accessed November 16, 2022. eMarketer explains that its methodology for
counting users is more conservative than the methodology that TikTok itself
used to generate its 1 billion estimate.
x “About Spotify,” Spotify, May 25, 2022, https://newsroom.spotify.com/comp
any-info/. Accessed May 25, 2022.
“Essential YouTube stats for 2022,” Datareportal, February 23, 2022, https://
datareportal.com/essential-youtube-stats. Accessed May 25, 2022.
xi “TikTok Users Worldwide (2020–2025).” eMarketer predicted in late 2021 that
TikTok’s growth will slow significantly starting in 2022.
xii “TikTok Is Hitting Revenue Milestones at Lightning Speed,” Chartr, June 29,
2022, https://www.chartr.co/stories/2022-06-29-2-tiktok-is-hitting-milesto
nes-quickly. Accessed July 14, 2022.
xiii MIREX2020 Results, MIREX, https://www.music-ir.org/mirex/wiki/
2020:MIREX2020_Results. Accessed August 5, 2020.
xiv Andy Baio, “The Tangled Issue of Cover Song Copyright on YouTube,” Wired,
March 5, 2012, https://www.wired.co.uk/article/cover-song-licensing-on-
youtube. Accessed May 25, 2022.
xv Jessica Toonkel, Anne Steele, and Salvador Rodriguez, “TikTok Parent
ByteDance Plans Music-Streaming Expansion,” Wall Street Journal, October
12, 2022, https://www.wsj.com/articles/tiktok-parent-bytedance-plans-
music-streaming-expansion-11665602937. Accessed November 16, 2022.
xvi Directive (EU) 2019/790 of the European Parliament and of the Council of
17 April 2019 on copyright and related rights in the Digital Single Market
and amending Directives 96/9/EC and 2001/29/EC. Official Journal of the
European Union, https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/
?uri=CELEX:32019L0790&rid=1. Accessed May 25, 2022.
xvii See for example Bill Rosenblatt, “EU Article 13 (Now Article 17) Passes After
More Changes, Making Copyright Filtering More Likely,” Copyright and
Technology, April 1, 2019, https://copyrightandtechnology.com/2019/04/01/
References 439
eu-article-13-now-article-17-passes-after-more-changes-making-copyright-
filtering-more-likely/. Accessed May 25, 2022.
xviii Maria Strong (Acting Registry of Copyrights), Section 512 of Title 17
(Washington, DC: United States Copyright Office, May 2020). For a brief sum-
mary, see Bill Rosenblatt, “Copyright Office’s Section 512 Report Finds the
Balance Askew,” Copyright and Technology, May 26, 2020, https://copyrightan
dtechnology.com/2020/05/26/copyright-offi ce-releases-section-512-report/.
Accessed May 25, 2022.
xix Shelby Carpenter, “Inside Vevo’s Plan To Beat YouTube And Become The
MTV Of The Digital Age,” Forbes, October 9, 2016, https://www.forbes.com/
sites/shelbycarpenter/2016/10/19/vevo-play-beat-youtube-mtv-digital-age/.
Accessed May 25, 2022.
xx “VEVO Was Most Trafficked U.S. Entertainment-Music Web Network in
December 2009,” Vevo press release, January 13, 2010, https://www.prnewsw
ire.com/news-releases/vevo-was-most-trafficked-us-entertainment-music-
web-network-in-december-2009-81347087.html. Accessed May 25, 2022.
xxi Ethan Smith, “MTV Overtakes Vevo as Top Online Music Destination,” Wall
Street Journal, September 8, 2010, https://www.wsj.com/articles/BL-SEB-
46396. Accessed July 25, 2022.
xxii “The World’s Leading Music Video Network,” Vevo, Undated, https://www.
hq.vevo.com/. Accessed July 25, 2022.
xxiii Jack Nicas, “YouTube Tops 1 Billion Hours of Video a Day, on Pace to Eclipse
TV,” Wall Street Journal, February 27, 2017, https://www.wsj.com/articles/
youtube-tops-1-billion-hours-of-video-a-day-on-pace-to-eclipse-tv-148
8220851. Accessed April 4, 2023.
xxiv “Music Became Even More Valuable on YouTube in 2019,” Pex, June 4,
2019, https://pex.com/blog/state-of-youtube-2019-music-more-valuable/.
Accessed May 25, 2022.
xxv The Infinite Dial 2015, Edison Research and Triton Digital, March 4, 2015,
https://www.edisonresearch.com/the-infinite-dial-2015/. Accessed May
25, 2022.
xxvi The Infinite Dial 2022. Edison Research and Triton Digital, March 23,2022,
https://www.edisonresearch.com/the-infinite-dial-2022/. Accessed March
25, 2022.
xxvii “Rewarding Creativity: Fixing the Value Gap (excerpt from Global Music
Report 2017),” International Federation of the Phonographic Industry, April
25, 2017, https://web.archive.org/web/20190221033533/https:/www.ifpi.org/
downloads/GMR2017_ValueGap.pdf. Accessed May 25, 2022.
xxviii Ethan Smith and Jessica E. Vascellaro, “Warner Pulls Music From YouTube,”
Wall Street Journal, December 22, 2008, https://www.wsj.com/articles/SB122
980193788724073. Accessed May 25, 2022.
xxix See for example: Stuart Dredge, “How Much Do Musicians Really Make from
Spotify, iTunes and YouTube?,” Guardian, April 3, 2015, https://www.theguard
ian.com/technology/2015/apr/03/how-much-musicians-make-spotify-itu
nes-youtube. Accessed May, 25, 2022.
xxx Paul Resnikoff, “YouTube Music Says It Pays the Same Royalty Rate as
Spotify—At Least on Its Subscription Streams,” Digital Music News, October 9,
440 References
2019, https://www.digitalmusicnews.com/2019/10/09/youtube-music-prem
ium-subscription/. Accessed May 25, 2022.
xxxi Tim Ingham, “So . . . How Much Did TikTok Actually Pay the Music Industry
From Its $4bn in Revenues Last Year?,” Music Business Worldwide, July 14,
2022, https://www.musicbusinessworldwide.com/so-how-much-did-tik
tok-actually-pay-the-music-industry-from-its-4bn-in-revenues-last-year/.
Accessed July 25, 2022.
xxxii Elias Leight, “TikTok Pays Artists ‘Almost Nothing’ in Music Royalties—And
the Industry is Losing Patience,” Billboard, November 7, 2022, https://www.
billboard.com/pro/tiktok-pays-artists-little-music-royalties-insiders-fed-up/.
Accessed November 16, 2022.
xxxiii Dan Whateley, “How TikTok is changing the music industry,” Business Insider,
July 7, 2022, https://www.businessinsider.com/how-tiktok-is-changing-the-
music-industry-marketing-discovery-2021-7. Accessed July 14, 2022.
xxxiv Rania Aniftos, “Here’s a Timeline of the Viral ‘Dreams’ TikTok, From
Cranberry Juice Gifts to Stevie Nicks’ Recreation,” Billboard, October 14, 2020,
https://w ww.billboard.com/music/music-news/viral-dreams-tiktok-timel
ine-9465600/. Accessed July 14, 2022.
xxxv Whateley, “How TikTok Is Changing the Music Industry.”
xxxvi Chris Molanphy, “Feat. Don’t Fail Me Now,” Slate, July 31, 2015, https://slate.
com/culture/2015/07/the-history-of-featured-rappers-and-other-featured-
artists-in-pop-songs.html. Accessed May 25, 2022.
xxxvii Ashley King, “YouTube Music Exec Calls TikTok Short Videos ‘Junk Food’—
‘It Has to Lead You to Long-Form Content So It’s Not Empty Calories,’” Digital
Music News, July 8, 2022, https://www.digitalmusicnews.com/2022/07/08/
youtube-music-exec-tiktok-junk-food/. Accessed July 14, 2022.
xxxviii Elias Leight, “Sped-Up Songs Are Taking Over TikTok and Driving Songs Up
the Charts,” Billboard, November 10, 2022, https://www.billboard.com/pro/
sped-up-songs-tiktok-streaming-charts/. Accessed November 16, 2022.
xxxix Tamas Bodzsar, “How Long Should a Chorus Be,” Songwriting Essentials,
Undated, https://www.howtowritebettersongs.com/how-long-should-a-cho
rus-be/. Accessed May 25, 2022.
xl David Farrell, “Stingray Buys Qello Concerts. Now No. 1 SVOD Supplier,” FYI
Music News, January 4, 2018, https://www.fyimusicnews.ca/articles/2018/01/04/
stingray-buys-qello-concerts-now-no-1-svod-supplier. Accessed May 25, 2022.
xli Julian Clover, “Les Echos-Le Parisien buys into Medici.TV,” Broadband TV
News, February 25, 2020, https://www.broadbandtvnews.com/2020/02/25/
les-echos-le-parisien-buys-into-medici-tv/. Accessed May 25, 2022.
Chapter 11
i Bernard Marr, “The Key Definitions of Artificial Intelligence That Explains Its
Importance,” Forbes, February 14, 2014, https://www.forbes.com/sites/bern
ardmarr/2018/02/14/the-key-definitions-of-art ificial-intelligence-ai-that-
explain-its-importance/. Accessed June 7, 2022.
ii Murray Stassen, “Warner Is Signing Double the Number of Artists via AI
Driven Tool Sodatone,” Music Business Worldwide, November 24, 2020,
References 441
https://www.musicbusinessworldwide.com/warner-is-signing-double-the-
numb er-of-artists-via-ai-driven-ar-tool-s odatone-than-it-did-last-year-
now-its-hired-a-global-head-of-data-science/. Accessed June 7, 2022.
iii “What is Machine Learning?” IBM Cloud Education, https://www.ibm.com/
cloud/learn/machine-learning. Accessed June 2, 2022.
iv B. H. Huang and Lawrence Rabiner, Fundamentals of Speech Recognition
(London: Pearson College Division, 1993), 1–28.
v Bernadette Jackson, “How Amazon Echo Works?” How Stuff Works, https://
electronics.howstuff works.com/gadgets/high-tech-gadgets/amazon-echo.
htm#. Accessed April 4, 2023.
vi “First Recording of Computer-Generated Music—Created by Alan Turing—
Restored,” Guardian, September 26, 2016, https://www.theguardian.com/scie
nce/2016/sep/26/first-recording-computer-generated-music-created-alan-
turing-restored-enigma-code. Accessed June 2, 2022.
vii Tim Adams, “Interview David Cope: ‘You Pushed the Button and Out Came
Hundreds and Thousands of Sonatas,’” Guardian, July 11, 2010, https://www.
theguardian.com/technology/2010/jul/11/david-cope-computer-composer.
Accessed June 2, 2022.
viii “List of Most Streamed YouTube Videos,” Wikipedia, https://en.wikipedia.org/
wiki/List_of_most-viewed_YouTube_videos. Accessed June 2, 2022.
ix Andrea Bossi, “There Were More Than One Trillion Music Streams In U.S.
During 2019, Record High,” Forbes, January 10, 2020, https://www.forbes.
com/sites/andreabossi/2020/01/10/over-one-trillion-music-streams-in-us-
during-2019-record-high. Accessed June 13, 2022.
x “List of Most Followed Twitter Accounts,” Wikipedia, https://en.wikipedia.
org/wiki/List_of_most-followed_Twitter_accounts. Accessed March 31, 2023.
xi Linda Andress, “Top Instagram Accounts in 2023,” How Sociable, January 24,
2020, https://howsociable.com/charts-instagram-top-accounts/. Accessed
March 31, 2023.
xii Rutger Rosenberg and Jason Joven, “How To Understand and Use Music
Data Analytics Tools,” Hypebot, Undated, https://www.hypebot.com/hype
bot/2020/06/how-to-understand-and-use-music-data-analytics-tools.html.
Accessed June 2, 2022.
xiii “Making Music and Art Using Machine Learning,” Google Magenta AI, https://
magenta.tensorfl ow.org/. Accessed June 2, 2022.
xiv “Open AI Jukebox,” Open AI, https://openai.com/blog/jukebox/. Accessed
June 2, 2022.
xv Smart Audio Report, NPR and Edison Research, June 17, 2022, https://www.
edisonresearch.com/smart-audio-report-2022-from-npr-and-edison-resea
rch/. Accessed June 17, 2022.
xvi Stuart Dredge, “The Future of Smart Speakers: How Voice Tech is Impacting
the Music Business,” Midem: Music Industry Insights, October 10, 2019,
https://blog.midem.com/2019/10/the-future-of-smart-speakers/. Accessed
June 2, 2022.
xvii James Shotwell, “36% of Americans Now Own Smart Speakers: Here’s What
That Means for Musicians,” Haulix, February 26, 2019, https://haulixdaily.
com/2019/02/smart-speakers-music-business. Accessed June 2, 2022.
442 References
xviii “Smart Speaker Use Case Frequency,” Statista, https://www.statista.com/sta
tistics/994696/united-states-smart-speaker-use-case-frequency/. Accessed
June 2, 2022.
xix Tim Ingham, “‘Fake Artists’ Have Billions of Streams on Spotify. Is Sony Now
Playing the Service at Its Own Game?”, Rolling Stone, May 15, 2019, https://
www.rollingstone.com/pro/features/fake-artists-have-billions-of-streams-
on-spot ify-is-s ony-now-playing-t he-s ervice-at-its-own-game-834746/ .
Accessed June 13, 2022.
xx Murray Stassen, “Over 1,000 Songs with Human Mimicking AI Vocals Have
Been released by Tencent Music in China. One of Them Has 100M Streams,”
Digital Music News, November 15, 2022, https://www.musicbusinessworldw
ide.com/over-1000-songs-human-mimicking-ai-vocals-have-been-released-
by-tencent-music-in-china-one-of-them-has-over-100m-streams/. Accessed
November 16, 2022.
xxi Dmitri Vietze and Drew Silverstein, “What Makes Amper’s AI Music Tools
Different with CEO Drew Silverstein,” Music Tectonics (podcast), March 25,
2020, https://musictectonics.libsyn.com/what-makes-ampers-ai-music-
tools-different-with-ceo-drew-silverstein. Accessed June 2, 2022.
xxii Eric Harvey, “How Smart Speakers Are Changing the Way We Listen to
Music,” Pitchfork, June 29, 2018, https://pitchfork.com/features/article/how-
smart-speakers-are-changing-the-way-we-listen-to-music/. Accessed June
2, 2022.
xxiii The Infinite Dial 2021, Edison Research and Triton Digital, https://www.edi
sonresearch.com/the-infinite-dial-2021-2/. Accessed July 17, 2022.
xxiv The Infinite Dial 2021.
xxv Murray Stassen, “Warner Joins $14M Funding Round in AI Music Start-up
Lifescore,” Music Business Worldwide, March 7, 2022, https://www.musicbu
sinessworldwide.com/warner-joins-11m-funding-round-in-ai-music-star
tup-lifescore-founded-by-siri-co-inventor-tom-gruber1/. Accessed June
2, 2022.
xxvi Salvatore Raieli, “Generate a Piano Cover With AI,” Medium, November 10,
2022, https://medium.com/mlearning-ai/generate-a-piano-cover-with-ai-
f4178bc9cb30. Accessed November 16. 2022.
xxvii Phillip Tracy, “Apple’s Newest Acquisition Could Mean Changes for Apple
Music,” Gizmodo, February 7, 2022, https://gizmodo.com/apple-ai-music-
acquisition-1848496216. Accessed June 2, 2022.
xxviii Sophia Ciocca, “Spotify’s Discover Weekly: How Machine Learning Finds
Your New Music,” Hackernoon, October 10, 2017, https://hackernoon.com/
spotif ys-discover-weekly-how-machine-learning-finds-your-new-music-
19a41ab76efe. Accessed June 2, 2022.
xxix Sam Blake, “How LA-Based Super Hi-Fi Hopes to Change Streaming Audio
using AI,” Dot.LA, September 21, 2020, https://dot.la/super-hi-fi-2647727032.
html. Accessed June 2, 2022.
xxx Zoe Beery, “Your Most-Played Song of 2020 Is . . . White Noise?”, New York
Times, December 24, 2020, https://www.nytimes.com/2020/12/24/style/
white-noise-ambient-music-spotify.html. Accessed June 2, 2022.
xxxi Dan Jeffries, “The Musician in the Machine,” Google Magenta AI, https://mage
nta.tensorfl ow.org/musician-in-the-machine. Accessed June 2, 2022.
References 443
xxxii Alex Bainter, “How Mubert (Probably) Works; Generative Music at Scale,”
Medium, August 29, 2019, https://medium.com/@alexbainter/how-mubert-
probably-works-e44de23c45bd. Accessed June 7, 2022.
xxxiii Michael Tauberg, “Spotify Is Killing Song Titles,” Medium, March 23, 2018,
https://michaeltauberg.medium.com/spotify-is-killing-s ong-titles-5f48b
7827653. Accessed June 2, 2022.
xxxiv Rob Arcand, “The Artists Using Artificial Intelligence To Dream Up The
Future Of Music,” Spin, June 4, 2019, https://www.spin.com/2019/06/ai-
music-artificial-intelligence-feature-holly-herndon-yacht/. Accessed June
2, 2022.
xxxv T. O’Brien, “Listen to an AI Sing an Uncanny Rendition of ‘Jolene’,”
Engadget, November 2, 2022, https://www.engadget.com/holly-plus-
holly-herndon-dolly-parton-jolene-ai-cover-204750236.html. Accessed
November 3, 2022.
xxxvi Holly Herndon, “Holly +. Holly Herndon,” http://www.hollyherndon.com/
holly. Accessed June 2, 2022.
xxxvii Grimes & Endel, “Endel x Grimes AI Lullaby,” Endel.io. https://ailullaby.endel.
io/. Accessed June 2, 2022.
xxxviii “Deepfake Queen to Deliver Channel 4 Christmas Message,” BBC, December
23, 2020, https://www.bbc.com/news/technology-55424730. Accessed June
2, 2022.
xxxix Winston Cho, “Does Kendrick Lamar Run Afoul of Copyright Law by Using
Deepfakes in ‘The Heart Part 5’?”, Hollywood Reporter, May 12, 2022, https://
www.hollywoodreporter.com/business/digital/does-kendrick-lamar-run-
afoul-of-copyright-law-by-using-deepfa kes-in-the-heart-part-5-1235145
596/. Accessed April 5, 2023.
xl Jonathan and Ruoming Pang Shen, “Tacotron 2: Generating Human-like
Speech from Text,” Google AI Blog, December 19, 2017, https://ai.googleb
log.com/2017/12/tacotron-2-generating-human-like-speech.html. Accessed
June 2, 2022.
xli Tim Knowles, “To Be or Not To Be Me: Jay-Z Asks Big Question,” Times
(London), April 30, 2020, https://www.thetimes.co.uk/article/to-be-or-not-
to-be-me-jay-z-asks-big-question-hk6vffl0m. Accessed June 2, 2022.
xlii Dan Robitzski, “Mind-Melting AI Makes Frank Sinatra Sing ‘Toxic’ by Britney
Spears,” Futurism, May 20, 2020, https://futurism.com/mind-melting-ai-
frank-sinatra-toxic-britney-spears. Accessed June 2, 2022.
xliii “Behind the Doodle: Celebrating Johann Sebastian Bach,” Google, https://
www.youtube.com/watch?v=XBfYPp6KF2g&t=104s. Accessed June
2, 2022.
xliv Katherine B. Forrest, “Copyright Law and Artificial Intelligence: Emerging
Issues,” Journal of the Copyright Society of the USA, 65 (Fall 2018): 355–370.
xlv Susannah Cullinane, “Monkey Does Not Own Selfie Copyright, Appeals
Court Rules,” CNN, April 24, 2018, https://www.cnn.com/2018/04/24/us/mon
key-selfie-peta-appeal/index.html. Accessed June 2, 2022.
xlvi Jon Porter, “US Patent Office Rules That Artificial Intelligence Cannot be a
Legal inventor,” The Verge, April 29, 2020, https://www.theverge.com/2020/
4/29/21241251/artificial-intelligence-inventor-united-states-patent-tradem
ark-offi ce-intellectual-property. Accessed June 7, 2022.
444 References
xlvii Kris Holt, “You can’t copyright AI-created art, according to US officials,”
Engadget, February 21, 2022, https://www.engadget.com/us-copyright-offi ce-
art-ai-creativity-machine-190722809.html. Accessed June 7, 2022.
xlviii “Copyright Registration Guidance: Works Containing Material Generated by
Artificial Intelligence,” Federal Register 88, no. 51 (March 16, 2023): 16190–
16194, https://www.federalregister.gov/documents/2023/03/16/2023-05321/
copyright-registration-guidance-works-containing-material-generated-by-
artificial-intelligence. Accessed March 31, 2023.
xlix Dani Deahl, “The USPTO Wants to Know if Artificial Intelligence Can Own
the Content it Creates,” The Verge, November 13, 2019, https://www.theverge.
com/2019/11/13/20961788/us-government-ai-copyright-patent-trademark-
offi ce-notice-artificial-intelligence. Accessed June 2, 2022.
l Amper, “Amper Terms of Service,” Amper. https://www.ampermusic.com/
terms-2/. Accessed June 2, 2022.
li Stuart Dredge, “UK bodies protest against unlicensed AI data-mining of
music,” Music Ally, July 7, 2020, https://musically.com/2022/07/07/uk-bod
ies-protest-against-unlicensed-ai-data-mining-of-music/. Accessed July
8, 2022.
lii Bill Hochberg, YouTube Won’t Take Down A Deepfake Of Jay-Z Reading
Hamlet —To Sue, Or Not To Sue,” Forbes, May 18, 2020, https://www.forbes.
com/sites/williamhochberg/2020/05/18/to-sue-or-not-to-sue—-that-is-the-
jay-zs-deepfake-question. Accessed June 2, 2022.
liii Ashley King, “Virtual Characters Are Hitting the Spotify Charts,” Digital
Music News, July 21, 2021, https://www.digitalmusicnews.com/2021/07/21/
virtual-characters-metaverse-spotify-charts/. Accessed June 2, 2022.
liv Cecelia Hubbert, “Midler v. Ford Motor Company,” University of Denver
Sturm College of Law, https://www.law.du.edu/documents/sports-and-entert
ainment-law-journal/case-summaries/1988-midler-v-ford-motor-co.pdf.
Accessed June 7, 2022.
lv Cade Metz. “Lawsuit Takes Aim at the Way AI is Built,” New York Times,
November 23, 2022. https://www.nytimes.com/2022/11/23/technology/copi
lot-microsoft-ai-lawsuit.html. Accessed April 5, 2023.
lvi Beatrice Nolan. “AI art generators face separate copyright lawsuits from Getty
Images and a group of artists,” Business Insider, January 19, 2023, https://www.
businessinsider.com/ai-art-artists-getty-images-lawsuits-stable-diffusion-
2023-1. Accessed January 31, 2023.
lvii Michael Nash, “Something New: Artificial Intelligence and the Perils of
Plunder,” Music Business Worldwide, February 14, 2023, https://www.musicbu
sinessworldwide.com/michael-nash-universal-something-artificial-intellige
nce-and-the-perils-plunder/. Accessed February 15, 2023.
lviii “Smart Speakers Driving New Music Consumption Habits, Says New
AudienceNet Survey,” Music Business Association, October 10, 2018, https://
musicbiz.org/news/smart-speakers-driving-new-music-consumption-hab
its-says-new-audiencenet-study/. Accessed June 2, 2022.
lix Mark Sutherland, “Amazon Music Director Ryan Redington on Country’s
Streaming Revolution,” Music Week, March 12, 2020, https://www.musicw
eek.com/digital/read/amazon-music-director-r yan-redington-on-country-s-
streaming-revolution/079179. Accessed June 2, 2022.
References 445
lx Keith Caulfield, “‘Encanto’ & ‘We Don’t Talk About Bruno’ Rule Luminate’s
2022 Mid-Year Charts.” Billboard, July 14, 2022, https://billboard.com/music/
chart-beat/Encanto-we-don’t-talk-about-bruno-luminate-2022-midyear-cha
rts-1235114205/. Accessed April 5, 2023.
lxi Richard Lawler, “Kanye West’s $200 Stem Player will be the only way to get
his next album, Donda 2,” The Verge, February 18, 2022, https://www.theve
rge.com/2022/2/18/22940748/donda-2-stem-player-kanye-west-exclusive-
music. Accessed June 7, 2022.
Chapter 12
i “The Music You Want Wherever You Want It,” John Okolowicz collection of
publications and advertising on radio and consumer electronics (Accession
2014.277), Hagley Museum & Library, Wilmington, DE 19807.
ii Amy Scattergood, “Artisans of the Roast,” Los Angeles Times, October 25, 2006,
https://www.latimes.com/food/la-fo-roast102506-story.html. Accessed April
4, 2023.
iii This and all subsequent statistics in this section come from annual consumer
surveys by National Coffee Association of the USA, the American coffee in-
dustry trade association. Gourmet coffee includes espresso-based drinks.
iv Daniel McGinn, “The Buzz Machine,” Boston Globe Sunday Magazine, August
7, 2011, http://archive.boston.com/business/articles/2011/08/07/the_inside_
story_of_keurigs_rise_to_a_billion_dollar_coffee_empire/. Accessed April
4, 2023.
v Josh Dzieza, “Keurig’s Attempt to ‘DRM’ Its Coffee Cups Totally Backfired,”
The Verge, February 5, 2015, https://www.theverge.com/2015/2/5/7986327/
keurigs-attempt-to-drm-its-coff ee-cups-totally-backfired. Accessed June
2, 2022.
vi Sabrina Imbler, “The Best Keurig Machine (But We Really Don’t Recommend
It),” Wirecutter, September 10, 2021, https://www.nytimes.com/wirecutter/
reviews/b est-keurig-machine/#the-drawbacks-to-keurig-machines-and-
why-we-dont-recommend-them. Accessed June 2, 2022.
vii National Coffee Association of the USA.
viii Bill Rosenblatt, “Disruption Comes to Higher-Ed Publishing,” Publishers
Weekly, August 16, 2019, https://www.publishersweekly.com/pw/by-topic/
digital/content-and-e-books/article/80947-disruption-comes-to-higher-ed-
publishing.html. Accessed April 4, 2023.
ix W. H. Powell, “Top 10 Greatest BuzzFeed Lists Of All Time,” Medium, June
28, 2016, https://medium.com/@whpeezy/top-10-buzzfeed-lists-of-all-time-
that-will-tick le-your-list-b one-until-you-have-a-listgasm-f fb1ce7cd52c.
Accessed June 2, 2022.
x “Podtrac Industry Ranking: Top 20 Podcasts, April 2022,” Podtrac, http://
analytics.podtrac.com/podcast-rankings. Accessed June 20, 2022.
xi See generally Mike Isaac, Super Pumped: The Battle for Uber (New York: W. W.
Norton, 2019).
xii Dr. Todd Boyd, interview with the authors, November 1, 2022.
446 References
xiii Todd Boyd, Young, Black, Rich and Famous: The Rise of the NBA, the Hip Hop
Invasion and the Transformation of American Culture (New York: Doubleday,
2003), 150–61.
xiv Tatiana Cirisano, “Is TikTok Becoming a Record Label? The Question Misses
the Point,” MIDIA Research Blog, May 18, 2022, https://midiaresearch.com/
blog/is-t iktok-b ecoming-a-record-label-t he-question-miss es-t he-p oint.
Accessed July 14, 2022.
xv Joshua Klein, “Robert Christgau: Christgau’s Consumer Guide: Albums of the
‘90s,” AV Club, March 29, 2002, https://www.avclub.com/robert-christgau-
christgaus-consumer-guide-albums-of-1798193851. Accessed April 7, 2023.
The estimate of 35,000 albums per year released in the 1990s works out to just
under 1,000 tracks per days assuming 10 tracks per album.
xvi “Top Entertainment Trends for 2023: What the Data Says, as Presented at
SXSW 2023,” Luminate, Undated, https://luminatedata.com/reports/sxsw-
top-entertainment-trends-for-2023/. Accessed April 7, 2023.
xvii Dan Rys, “Lucian Grainge Calls for ‘Updated Model’ for Music Industry: Read
His Memo to UMG Staff,” Billboard, January 12, 2023, https://www.billbo
ard.com/pro/lucian-g rainge-umg-f ull-staff-memo-2023-read-mess age/.
Accessed April 7, 2023.
xviii Bill Rosenblatt, “UltraViolet, Hollywood’s Attempt To Control The Digital
Video Supply Chain, Will Shut Down,” Forbes, February 3, 2019, https://www.
forbes.com/sites/billrosenblatt/2019/02/03/ultraviolet-hollywoods-attempt-
to-control-the-digital-video-supply-chain-will-shut-down/. Accessed June
2, 2022.
xix “Hulu.com Proves Victorious over Skeptics,” SFGATE.com, December 23,
2008, https://www.sfgate.com/entertainment/article/Hulu-com-proves-vic
torious-over-skeptics-3257176.php. Accessed June 2, 2022.
xx Samuel Spencer, “How Many Subscribers Do Netflix, Disney+and the Rest of
the Streaming Services Have?” Newsweek, May 11, 2021, https://www.newsw
eek.com/netfl ix-amazon-hulu-disney-most-subscribers-streaming-service-
1590463. Accessed June 2, 2022.
xxi Erik Gruenwedel, “Streaming Fatigue: Too Many Choices, at Too High a
Cost?” MediaPlayNews, November 22, 2021, https://www.mediaplaynews.
com/streaming-fatigue/. Accessed July 19, 2022.
i Bill Rosenblatt, “The Future of Blockchain Technology in the Music Industry,”
Journal of the Copyright Society of the USA 66 (Spring 2019): 271–89.
ii “Ethereum Transactions Per Second Chart,” Blockchair, https://blockchair.
com/ethereum/charts/transactions-per-second. Accessed July 17, 2022.
iii MacKenzie Sigalos, “Ethereum Had a Rough September. Here’s Why and How
It’s Being Fixed,” CNBC.com, October 2, 2021, https://www.cnbc.com/2021/
10/02/ethereum-had-a-rough-september-heres-why-and-how-it-gets-fixed.
html. Accessed April 29, 2022.
iv William Entriken, Dieter Shirley, Jacob Evans, and Nastassia Sachs, “EIP-
721: Non- Fungible Token Standard, Ethereum Improvement Proposals,
no. 721,” Ethereum.org, https://eips.ethereum.org/EIPS/eip-721. Accessed
May 28, 2022.
v “Music/Web3 Dashboard,” Water & Music, https://www.waterandmusic.com/
data/music-web3-dashboard/. Accessed November 16, 2022; “Music NFT
References 447
Sales in 2021: What We Learned,” Water & Music, May 6, 2022, https://www.
waterandmusic.com/music-nft-sales-in-2021-what-we-learned/. Accessed
November 16, 2022.
vi Joshua P. Friedlander and Matthew Bass, “Year-End 2021 RIAA Revenue
Statistics,” Recording Industry Association of America, https://www.riaa.com/
wp-content/uploads/2022/03/2021-Year-End-Music-Industry-Revenue-Rep
ort.pdf. Accessed May 20, 2022.
vii “How Do You Buy an NFT?” Coinbase, Undated, https://www.coinbase.com/
learn/crypto-basics/how-to-buy-nft. Accessed May 28, 2022.
viii “Music/Web3 Dashboard.”
ix Jeff Ihaza, “Tory Lanez Released an ‘NFT Album,’ and Then Things Got Extra
Weird,” Rolling Stone, September 30, 2021, https://www.rollingstone.com/
music/music-features/tory-lanez-nft-legal-questions-1233890/. Accessed
April 29, 2022.
x London Jennn, “Tory Lanez Says His NFT Album Is Reselling on a Secondary
Market for $100k,” AllHipHop.com, August 25, 2021, https://allhiphop.com/
news/tory-lanez-says-his-nft-album-is-selling-for-100k/. Accessed April
29, 2022.
xi Samantha Hissong, “Kings of Leon Will Be the First Band to Release an Album
as an NFT.” Rolling Stone, March 3, 2021, https://www.rollingstone.com/pro/
news/kings-of-leon-when-you-see-yourself-album-nft-crypto-1135192/.
Accessed April 29, 2022.
xii Dylan Smith, “Warner Music Group Inks ‘First-Ever Partnership’ With POAP
‘To Mint Shared Memories As NFTs,’” Digital Music News, April 14, 2022,
https://www.digitalmusicnews.com/2022/04/14/warner-music-group-poap-
partnership/. Accessed May 28, 2022.
xiii Samantha Hissong, “Shawn Mendes’ Manager Is Helping Launch a New NFT
Marketplace,” Rolling Stone, March 2, 2021, https://www.rollingstone.com/
pro/news/shawn-mendes-andrew-gertler-nft-marketplace-sturdy-exchange-
1135558/. Accessed April 29, 2022.
xiv Kristin Robinson, “HitPiece Wanted to Make an NFT for Every Song—Only
Its Founders Forgot to Ask Artists First,” Billboard, February 2, 2022, https://
www.billboard.com/pro/hitpiece-nft-every-song-artist-permission-found
ers/. Accessed June 5, 2022.
Bibliography
Arcand, Rob. “The Artists Using Artificial Intelligence To Dream Up The Future Of
Music.” Spin, June 4, 2019. https://www.spin.com/2019/06/ai-music-artificial-intellige
nce-feature-holly-herndon-yacht/. Accessed June 2, 2022.
“Are Songs Getting Even Shorter in 2021.” Hit Songs Deconstructed. https://www.hitson
gsdeconstructed.com/hsd_wire/the-short-game/. Accessed July 8, 2022.
Aspan, Maria. “For Shakira, First Came the Album, Then Came the Single.” New York
Times, June 12, 2006. https://www.nytimes.com/2006/06/12/business/media/12shak
ira.html. Accessed June 7, 2022.
Aswad, Jem. “Global Music Biz Revenue to Double to $131 Billion in 2030, Says Bullish
Goldman Sachs Report.” Variety, June 14, 2022. https://variety.com/2022/music/news/
global-music-revenue-double-goldman-sachs-report-1235293827/#!. Accessed June
28, 2022.
Atwood, Brett. “Capitol to Sell Digital Singles.” Billboard. September 13, 1997.
“Audio File Size Calculations.” AudioMountain.com. http://www.audiomountain.com/
tech/audio-file-size.html. Accessed May 23, 2022.
Austen, Jake. TV a Go Go: Rock on TV. Chicago: Chicago Review Press, 2005.
Bainter, Alex. “How Mubert (Probably) Works; Generative Music at Scale.” Medium,
August 29, 2019. https://medium.com/@alexbainter/how-mubert-probably-works-
e44de23c45bd. Accessed June 7, 2022.
Baio, Andy. “The Tangled Issue of Cover Song Copyright on YouTube.” Wired, March 5,
2012. https://www.wired.co.uk/article/cover-song-licensing-on-youtube. Accessed
May 25, 2022.
Banks, Jack. Monopoly: MTV’s Quest to Control the Music. Boulder, CO: Westview
Press, 1996.
“The Beatles.” EdSullivan.com. https://www.edsullivan.com/artists/the-beatles/. Accessed
May 4, 2022.
Beery, Zoë. “Your Most-Played Song of 2020 Is . . . White Noise?” New York Times,
December 24, 2020. https://www.nytimes.com/2020/12/24/style/white-noise-ambi
ent-music-spotify.html. Accessed June 2, 2022.
Berliner, Emile, and Frederic William Wile. Emile Berliner, Maker of the Microphone.
Indianapolis: Bobbs-Merrill, 1926.
Bernhard, Robert. “Consumer Electronics. Higher Fi by Digits: Digitally Encoded Audio
Disks and Associated Playback Gear Promise Breakthrough in High Fidelity, Low
Noise.” IEEE Spectrum 16 (1979): 28–32.
Birnbaum, Jane. “The Media Business: Without a Scratch, Used CD’s Rise Again.”
New York Times, September 6, 1993. https://www.nytimes.com/1993/09/06/business/
the-media-business-without-a-scratch-used-cds-rise-again.html.
Blake, Sam. “How LA-Based Super Hi-Fi Hopes to Change Streaming Audio using AI.”
Dot.LA, September 21, 2020. https://dot.la/super-hi-fi-2647727032.html. Accessed
June 2, 2022.
Blockchair. “Ethereum Transactions Per Second Chart,” https://blockchair.com/ether
eum/charts/transactions-per-second. Accessed July 17, 2022.
Blumenthal, Eli. “Hulu and Spotify Partner to Bundle TV and Music Streaming for $9.99
per Month.” USA Today, March 12, 2019. https://www.usatoday.com/story/tech/talk
ingtech/2019/03/12/apple-looming-hulu-and-spotify-partner-9-99-streaming-bun
dle/3138377002/. Accessed June 5, 2022.
Blumenthal, Ralph. “Goody Chain and Ex- Official Are Sentenced in Tapes Case.”
New York Times, November 6, 1982, 29.
Bibliography 451
Bodzsar, Tamas. “How Long Should a Chorus Be.” Songwriting Essentials. https://www.
howtowritebettersongs.com/how-long-should-a-chorus-be/. Accessed May 25, 2022.
Borland, John. “Judge Issues Injunction Against Napster.” CNET, January 2, 2002. https://
www.cnet.com/tech/services-and-software/judge-issues-injunction-against-napster/.
Accessed May 23, 2022.
Borland, John. “Legal P2P Opens for Business.” CNET, September 26, 2006. https://www.
cnet.com/tech/home-entertainment/legal-p2p-opens-for-business/. Accessed May
25, 2022.
Borland, John. “Metallica Fingers 335,435 Napster Users.” CNET, January 2, 2002. https://
www.cnet.com/tech/services-and-software/metallica-fingers-335435-napster-users/.
Accessed May 23, 2022.
Borland, John. “Napster Boots Dr. Dre Fans From Service.” CNET, January 2, 2002.
https://www.cnet.com/news/napster-boots-dr-dre-fans-from-service/. Accessed June
7, 2022.
Bossi, Andrea. “There Were More Than One Trillion Music Streams In U.S. During 2019.”
Forbes, January 10, 2020. https://www.forbes.com/sites/andreabossi/2020/01/10/over-
one-trillion-music-streams-in-us-during-2019-record-high. Accessed June 13, 2022.
Boyer, Peter. “CBS Assails NBC News over a Payola Report.” New York Times, April 3,
1986. https://www.nytimes.com/1986/04/03/arts/cbs-assails-nbc-news-over-a-pay
ola-report.html. Accessed July 12, 2022.
Boyd, Todd. Young, Black, Rich and Famous: The Rise of the NBA, the Hip Hop Invasion
and the Transformation of American Culture. New York: Doubleday, 2003.
Brockman, Daniel, and Jason W. Smith. “The Rise and Fall of the Columbia House Record
Club—And How We Learned to Steal Music.” Boston Phoenix, November 18, 2011.
https://thephoenix.com/Boston/music/129722-rise-and-fall-of-the-columbia-house-
record-clu/. Accessed May 25, 2022.
Bronson, Po, “Rebootlegger.” Wired, July 1, 1998. https://www.wired.com/1998/07/
newmedia-4/. Accessed May 23, 2022.
Browne, David, “How the 45 RPM Single Changed Music Forever.” Rolling Stone, March
15, 2019. https://www.rollingstone.com/music/music-features/45-vinyl-singles-hist
ory-806441/. Accessed July 13, 2022.
“B.T Interview: Why Network Radio Must Adapt or Die.” Broadcasting, Telecasting,
October 31, 1955, 35+.
Buchanan, Matt. “RIAA Tech Chief: DRM Not Dead, Will Become More Powerful than
You Can Possibly Imagine.” Gizmodo, May 8, 2008. https://gizmodo.com/riaa-tech-
chief-drm-not-dead-will-become-more-powerfu-388648. Accessed June 5, 2022.
Buxton, Madelien. “Spotify’s New Free Version Is So Good You Might Cancel Your
Subscription.” Refinery 29, April 24, 2018. https://www.refi nery29.com/en-us/2018/
04/197232/spotify-for-free-upgrade. Accessed June 5, 2022.
Calamar, Gary, and Phil Gallo. Record Store Days: From Vinyl to Digital and Back Again.
New York: Union Square & Co, 2012.
Carlsson, Sven, and Jonas Leijonhufvud. The Spotify Play: How CEO and Founder
Daniel Ek Beat Apple, Google, and Amazon in the Race for Audio Dominance.
New York: Diversion Books, 2021.
Carman, Ashley. “Spotify Acquires Another Podcast Network to Keep Building its
Original Show Catalog.” The Verge, March 26, 2019. https://www.theverge.com/2019/
3/26/18282301/spotify-podcasts-parcast-acquisition-gimlet-media-anchor. Accessed
June 5, 2022.
452 Bibliography
Carpenter, Shelby. “Inside Vevo’s Plan to Beat YouTube and Become The MTV of the
Digital Age.” Forbes, October 19, 2016. https://www.forbes.com/sites/shelbycarpenter/
2016/10/19/vevo-play-beat-youtube-mtv-digital-age/. Accessed May 25, 2022.
Caruso Jr., Enrico, Andrew Farkas, and William R. Moran. Enrico Caruso: My Father and
My Family. Portland, OR: Amadeus Press, 1997.
Caulfield, Keith. “1 of Every 25 Vinyl Albums Sold in U.S. in 2022 Was by Taylor Swift.”
Billboard, January 11, 2023. https://www.billboard.com/pro/taylor-swift-vinyl-alb
ums-1-of-every-25-sold/. Accessed January 31, 2023.
Caulfield, Keith. “’Encanto’ & ‘We Don’t Talk About Bruno’ Rule Luminate’s 2022 Mid-
Year Charts.” Billboard, July 14, 2022. https://www.billboard.com/music/chart-beat/
encanto-we-dont-talk-about-bruno-luminate-2022-midyear-charts-1235114205/.
Accessed April 5, 2023.
Chang, Jeff. Can’t Stop Won’t Stop: A History of the Hip- Hop Generation.
London: Picador, 2005.
Cho, Winston. “Does Kendrick Lamar Run Afoul of Copyright Law by Using Deepfakes
in ‘The Heart Part 5?,’” Hollywood Reporter, May 12, 2022. https://www.hollywoodr
eporter.com/business/digital/does-kendrick-lamar-run-afoul-of-copyright-law-by-
using-deepfakes-in-the-heart-part-5-1235145596/. Accessed April 5, 2023.
Christman, Ed. “Spotify and Publishing Group Reach $30 Million Settlement Agreement
Over Unpaid Royalties.” Billboard, March 17, 2016. https://www.billboard.com/music/
music-news/spotify-nmpa-publishing-30-million-settlement-unpaid-royalties-7263
747/. Accessed June 7, 2022.
Cilia, John. “The Search for the Oldest Record Store in the World.” Trackagescheme.com,
July 21, 2016. http://trackagescheme.com/the-search-for-the-oldest-record-store-in-
the-world/. Accessed May 9, 2022.
Ciocca, Sophia. “Spotify’s Discover Weekly: How Machine Learning Finds Your New Music.”
Hackernoon, October 10, 2017. https://hackernoon.com/spotifys-discover-weekly-how-
machine-learning-finds-your-new-music-19a41ab76efe. Accessed June 2, 2022.
Cirisano, Tatiana. “Is TikTok Becoming a Record Label? The Question Misses the Point.”
MIDIA Research Blog, May 18, 2022. https://midiaresearch.com/blog/is-tiktok-becom
ing-a-record-label-the-question-misses-the-point. Accessed July 14, 2022.
Cirisano, Tatiana. “Record Industry Groups Take Action Against Fake Music Streams in
Germany.” Billboard, August 24, 2020. https://www.billboard.com/pro/ifpi-bvmi-take-
action-against-fake-music-streams-germany/. Accessed June 7, 2022.
Clover, Julian. “Les Echos-Le Parisien Buys into Medici.TV.” Broadband TV News,
February 25, 2020. https://www.broadbandtvnews.com/2020/02/25/les-echos-le-paris
ien-buys-into-medici-tv/. Accessed May 25, 2022.
Coase, Ronald H. “Payola in Radio and Television Broadcasting.” Journal of Law and
Economics 22, no. 2 (October 1979): 269–328.
Cohen, Simon. “Amazon Music Unlimited Is Getting More Expensive in 2023.” Digital
Trends, January 20, 2023. https://www.digitaltrends.com/home-theater/amazon-
music-unlimited-february-2023-price-hike/. Accessed January 31, 2023.
Coinbase. “How Do You Buy an NFT?” https://www.coinbase.com/learn/crypto-basics/
how-to-buy-nft. Accessed May 28, 2022.
Cook, Emory. “Binaural Disks.” High Fidelity, November–December 1952, 33–35.
Copeland, Peter. “Manual of Analogue Sound Restoration Techniques.” The British
Library. https://web.archive.org/web/20110409161239/http:/www.bl.uk/reshelp/find
helprestype/sound/anaudio/analoguesoundrestoration.pdf. Accessed April 29, 2022.
Bibliography 453
“Copyright Registration Guidance: Works Containing Material Generated by Artificial
Intelligence.” Federal Register 88, no. 51 (March 16, 2023): 16190–16194, https://www.
federalregister.gov/documents/2023/03/16/2023-05321/copyright-registration-guida
nce-works-containing-material-generated-by-artificial-intelligence. Accessed March
31, 2023.
Corbett, Ian. “What Data Compression Does To Your Music.” Sound On Sound, April
2012. https://www.soundonsound.com/techniques/what-data-compression-does-
your-music. Accessed June 7, 2022.
Cosgrove-Mather, Bootie. “Poll: Young Say File-Sharing OK.” CBS News, September 18,
2003. https://www.cbsnews.com/news/poll-young-say-file-sharing-ok/. Accessed
June 7, 2022.
Costello, Sam, “History of the iPod: From the First iPod to the iPod Classic.” Lifewire,
January 12, 2020. https://www.lifewire.com/history-ipod-classic-original-2000732.
Accessed May 23, 2022.
Cousino Inc. “Audio Vendor Brochure and Manual.” http://lcweb2.loc.gov/master/mbrs/
recording_preser vation/manuals/Cousino%20Universal%20Audio%20Vendor.pdf.
Accessed May 15, 2022.
Covert, Adrian. “A Decade of iTunes Singles Killed the Music Industry.” CNN Business,
April 25, 2013. https://money.cnn.com/2013/04/25/technology/itunes-music-decline/
index.html. Accessed June 7, 2022.
Cullinane, Susannah. “Monkey Does Not Own Selfie Copyright, Appeals Court Rules.”
CNN, April 24, 2018. https://www.cnn.com/2018/04/24/us/monkey-selfie-peta-app
eal/index.html. Accessed June 2, 2022.
Daniel, Eric D., C. Dennis Mee, and Mark H. Clark. Magnetic Recording: The First 100
Years. Piscataway, NJ: IEEE Press, 1999.
Dannen, Fredric. Hit Men: Power Brokers and Fast Money Inside the Music Business.
New York: Vintage, 1991.
Deahl, Dani. “The Music Modernization Act Has Been Signed Into Law.” The Verge,
October 11, 2018. https://www.theverge.com/2018/10/11/17963804/music-modern
ization-act-mma-copyright-law-bill-labels-congress. Accessed June 7, 2022.
Deahl, Dani. “The USPTO Wants to Know If Artificial Intelligence Can Own the Content
It Creates.” The Verge, November 13, 2019. https://www.theverge.com/2019/11/13/
20961788/us-government-ai-copyright-patent-trademark-office-notice-artificial-
intelligence. Accessed June 2, 2022.
“Deepfake Queen to Deliver Channel 4 Christmas Message.” BBC, December 23, 2020.
https://www.bbc.com/news/technology-55424730. Accessed June 2, 2022.
Denisoff, R. Serge. Inside MTV. Milton Park: Routledge, 1988.
Denisoff, R. Serge. Tarnished Gold: The Record Industry Revisited. Milton
Park: Routledge, 1986.
Denisoff, R. Serge. Solid Gold: The Popular Record Industry. Piscataway, NJ: Transaction
Publishers, 1975.
Denton, Jack. “A Requiem for the Microsoft Zune.” Fast Company. June 15, 2022. https://
www.fastcompany.com/90761088/a-requiem-for-the-microsoft-zune-the-little-gad
get-that-couldnt. Accessed April 3, 2023.
Digital Media Association. “Streaming Forward 2020 Report.” https://dima.org/stream
ing-forward-report/. Accessed June 5, 2022.
“Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April
2019 on Copyright and Related Rights in the Digital Single Market and Amending
454 Bibliography
Directives 96/9/EC and 2001/29/EC.” Official Journal of the European Union, May 17,
2019. https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32019L0
790&rid=1. Accessed May 25, 2022.
Donnelly, Bob, Danielle Aguirre, Kris Muñoz, and Andrew Sparkler. “Flipping Publishing
Micro-Pennies into a Bucket.” SXSW 2014 CLE. https://foxrothschild.gjassets.com/
content/uploads/2017/03/Flipping-Publishing-Mirco-Pennies-into-a-Bucket.pdf.
Accessed May 4, 2022.
Dormehl, Luke. “38 Years Ago, CDs Rewrote Our Relationship With Music and Primed
Us For 2020.” Digital Trends, October 12, 2020. https://www.digitaltrends.com/featu
res/how-cds-prepared-us-for-the-future/. Accessed May 23, 2022.
Dowling, Stephen. “The Most Iconic Long Intros.” BBC, April 8, 2019. https://www.bbc.
com/c ulture/article/20190407-t he-most-iconic-long-s ong-intros-in-p op-music.
Accessed June 7, 2022.
Dredge, Stuart. “African Streaming Service Boomplay Now Has 75m Users.” Music Ally,
June 3, 2020. https://musically.com/2020/06/03/african-streaming-service-boomplay-
now-has-75m-users/. Accessed June 5, 2022.
Dredge, Stuart. “How Much Do Musicians Really Make from Spotify, iTunes and
YouTube?” Guardian, April 3, 2015. https://www.theguardian.com/technology/2015/
apr/03/how-much-musicians-make-spotify-itunes-youtube. Accessed April 3, 2023.
Dredge, Stuart. “The Future of Smart Speakers: How Voice Tech Is Impacting the Music
Business.” Midem: Music Industry Insights, October 10, 2019. https://blog.midem.com/
2019/10/the-future-of-smart-speakers/. Accessed June 2, 2022.
Dredge, Stuart. “Thom Yorke Calls Spotify ‘The Last Desperate Fart of a Dying Corpse.’”
Guardian, October 7, 2013. https://www.theguardian.com/technology/2013/oct/07/
spotify-thom-yorke-dying-corpse. Accessed June 7, 2022.
Dredge, Stuart. “Spotify CEO Talks Covid-19, Artist Incomes and Podcasting.” Music
Ally, July 30, 2020. https://musically.com/2020/07/30/spotify-ceo-talks-covid-19-art
ist-incomes-and-podcasting-interview/. Accessed June 7, 2022.
Dredge, Stuart. “UK Bodies Protest against Unlicensed AI Data-Mining of Music.” Music
Ally, July 7, 2020. https://musically.com/2022/07/07/uk-bodies-protest-against-unl
icensed-ai-data-mining-of-music/. Accessed July 8, 2022.
Duggan, Daniel. “Investor Buys Former Handleman Co. HQ in Troy for $3 Million.”
Crain’s Detroit Business, February 24, 2010. https://www.crainsdetroit.com/article/
20100224/DM01/302249988/investor-buys-former-handleman-co-hq-in-troy-for-3-
million. Accessed May 25, 2022.
Dzieza, Josh. “Keurig’s Attempt to ‘DRM’ Its Coffee Cups Totally Backfired.” The Verge,
February 5, 2015. https://www.theverge.com/2015/2/5/7986327/keurigs-attempt-to-
drm-its-coff ee-cups-totally-backfired. Accessed June 2, 2022.
Edison Research and Triton Digital. “The Infinite Dial 2015.” https://www.edisonresea
rch.com/the-infinite-dial-2015/. Accessed May 25, 2022.
Edison Research and Triton Digital. “The Infinite Dial 2021.” https://www.edisonresea
rch.com/the-infinite-dial-2021-2/. Accessed July 17, 2022.
Edison Research and Triton Digital. “The Infinite Dial 2022.” http://www.edisonresearch.
com/wp-content/uploads/2022/03/Infinite-Dial-2022-Webinar-revised.pdf. Accessed
May 25, 2022.
EMI. “EMI Music, Google and YouTube Strike Milestone Partnership.” EMI press release,
May 31, 2007. https://web.archive.org/web/20070602142911/http:/www.emigroup.
com/Press/2007/press38.htm. Accessed May 25, 2022.
Bibliography 455
“Ends Price-Fixing By ‘License Plan’: Supreme Court Gives Decision for Macy’s Against
the Victor Company.” New York Times, April 10, 1917. https://www.nytimes.com/1917/
04/10/archives/ends-pricefixing-by-license-plan-supreme-court-gives-decision-for.
html. Accessed May 9, 2022.
Engel, Friedrich, Peter Hammar, and Richard L. Hess. “A Selected History of Magnetic
Recording.” https://www.richardhess.com/tape/history/Engel_Hammar--Magnetic_T
ape_History.pdf. Accessed April 29, 2022.
Entriken, William, Dieter Shirley, Jacob Evans, and Nastassia Sachs. “EIP-721: Non-
Fungible Token Standard, Ethereum Improvement Proposals, No. 721.” Ethereum.org.
https://eips.ethereum.org/EIPS/eip-721. Accessed May 28, 2022.
Espe, Eric. “Music Industry Sees Threat in CD-Copying Technology.” Silicon Valley
Business Journal, August 30, 1998. https://www.bizjournals.com/sanjose/stories/1998/
08/31/story4.html. Accessed May 23, 2022.
“Essential YouTube stats for 2022.” Datareportal. https://datareportal.com/essential-yout
ube-stats. Accessed May 25, 2022.
Everard, Andrew. “Never Mind the Lack of MP3 Quality, Feel the Lack of MP3 Breadth.”
What Hi-Fi?, April 1, 2009. https://www.whathifi.com/news/usa-never-mind-lack-
mp3-quality-feel-lack-mp3-breadth. Accessed June 7, 2022.
Farrell, David. “Stingray Buys Qello Concerts. Now No. 1 SVOD Supplier.” FYI Music
News, January 4, 2018. https://www.fyimusicnews.ca/articles/2018/01/04/stingray-
buys-qello-concerts-now-no-1-svod-supplier. Accessed May 25, 2022.
Farrell, Maureen, and Anne Steele. “Tencent Music Files for U.S. IPO.” Wall Street Journal,
October 2, 2018. https://www.wsj.com/articles/tencent-music-files-for-u-s-ipo-153
8496869. Accessed May 25, 2022.
Farrell, Maureen. “Tencent Music Rises in Trading Debut.” Wall Street Journal, December
12, 2018. https://www.wsj.com/articles/tencent-music-opens-higher-in-trading-
debut-11544632668. Accessed June 5, 2022.
Feigenbaum, Edward, Avron Barr, and Paul Cohen. The Handbook of Artificial Intelligence,
Vols. 1–3. Los Altos, CA: William Kaufmann, 1982.
Ferchow, Davey. “How the DIY Cassette Movement of the 1970s and ‘80s Changed Music
Forever.” Discogs Blog, October 14, 2021. https://blog.discogs.com/en/how-the-diy-
cassette-movement-of-1970s-1980s-changed-music-forever/. Accessed April 29, 2022.
“Finley to Use 4 and 8 Track Units.” Billboard, July 31, 1965, 3.
“First Recording of Computer-Generated Music—Created by Alan Turing—Restored.”
Guardian, September 26, 2016. https://www.theguardian.com/science/2016/sep/26/
first-recording-computer-generated-music-created-alan-turing-restored-enigma-
code. Accessed June 2, 2022.
“The First Sony Walkman Goes on Sale.” A&E Television Networks. https://www.hist
ory.com/this-day-in-history/the-first-sony-walkman-goes-on-sale. Accessed April
29, 2022.
Fisher, Marc. Something in the Air: Radio, Rock, and the Revolution That Shaped a
Generation. New York: Random House, 2007.
Fitch, Charles. “How FM Stereo Came to Life.” Radio World, January 27, 2016. https://
www.radioworld.com/columns-and-views/roots-of-radio/how-fm-stereo-came-to-
life. Accessed August 3, 2022.
Flint, Matt. “A File-Sharing Timeline: From the Creation of MP3 to the Trial of The Pirate Bay.”
The Takeaway, April 17, 2009. https://www.wnycstudios.org/podcasts/takeaway/articles/
10560-file-sharing-timeline-creation-mp3-trial-pirate-bay. Accessed October 15, 2022.
456 Bibliography
Fontenot, Robert. “The History of American Bandstand.” Live About, June 12, 2019.
https://www.liveabout.com/american-bandstand-important-events-timeline-2523
794. Accessed May 4, 2022.
Forde, Eammon. “Spotify’s Discover Weekly Has Driven 2.3bn Hours of Listening.” Music
Ally, July 10, 2020. https://musically.com/2020/07/10/spotifys-discover-weekly-has-
driven-2-3bn-hours-of-listening/. Accessed June 5, 2022.
Fornatale, Peter, and Joshua Mills. Radio in the Television Age. New York: Overlook
Press, 1980.
Forrest, Katherine B. “Copyright Law and Artificial Intelligence: Emerging Issues.”
Journal of the Copyright Society of the USA 65 (Fall 2018): 355–70.
Fox Kraut, Barbara. “The Audio Home Recording Act.” DePaul Journal of Art, Technology,
and Intellectual Property Law 2, no. 2 (Spring 1992): 38–40. https://via.library.depaul.
edu/cgi/viewcontent.cgi?article=1495&context=jatip. Accessed May 23, 2022.
Fox, Barry. “Out of the Bell Tower.” New Scientist, May 6, 1995. https://www.newscientist.
com/article/mg14619764-400-out-of-the-bell-tower/. Accessed May 23, 2022.
Frank, Jay. Futurehit:DNA. Nashville: Futurehit, Inc., 2009.
Freeeman, Kim. “Classic Rock Thrives In 18 Months.” Billboard, October 25, 1986, 10+.
Friedlander, Joshua P., and Matthew Bass. “Year-End 2021 RIAA Revenue Statistics.”
Recording Industry Association of America. https://www.riaa.com/wp-content/uplo
ads/2022/03/2021-Year-End-Music-Industry-Revenue-Report.pdf. Accessed May
20, 2022.
Gapper, John. “Taylor Swift Is Perfect at the Three-Minute Spotify Song.” Financial Times.
November 25, 2022. https://www.ft.com/content/afdd5bfe-c5ae-411c-900f-8c5eab846
944. Accessed April 4, 2023.
Gartner. “Gartner Hype Cycle.” https://www.gartner.com/en/research/methodologies/
gartner-hype-cycle. Accessed November 18, 2022.
Gelatt, Roland. The Fabulous Phonograph 1877– 1977. Revised 2nd Edition.
New York: MacMillan Publishing, 1977.
Genzlinger, Neil. “Lou Ottens, Father of Countless Mixtapes, Is Dead at 94.” New York
Times, March 11, 2021. https://www.nytimes.com/2021/03/11/arts/music/lou-ottens-
dead.html. Accessed April 3, 2023.
Goldstein, Paul. Copyright’s Highway: From the Printing Press to the Cloud. 2nd Edition.
Redwood City, CA: Stanford University Press, 2003.
Goodman, Fred. “MP3 Technology Poised to Redefine Music Industry.” Rolling Stone,
March 9, 1999. https://www.rollingstone.com/music/music-news/mp3-technology-
poised-to-redefi ne-music-industry-97515/. Accessed May 23, 2022.
Goodwin, Andrew. Dancing in the Distraction Factory. Minneapolis: University of
Minnesota Press, 1992.
Google. “Behind the Doodle: Celebrating Johann Sebastian Bach.” https://www.youtube.
com/watch?v=XBfYPp6KF2g&t=104s. Accessed June 2, 2022.
Google Magenta AI. “Making Music and Art Using Machine Learning.” https://magenta.
tensorfl ow.org/. Accessed June 2, 2022.
Green, Stanley. “Jukebox Piracy.” The Atlantic Monthly, April 1, 1962, 136.
Grimes & Endel. “Endel x Grimes AI Lullaby.” Endel.io. https://ailullaby.endel.io/.
Accessed June 2, 2022.
Grimmelmann, James, Yan Ji, and Tyler Kell. “The Tangled Truth about NFTs and
Copyright.” The Verge, June 8, 2022. https://www.theverge.com/23139793/nft-crypto-
copyright-ownership-primer-cornell-ic3. Accessed June 9, 2022.
Grohl, Dave. “A Statement from the Desk of Record Store Day 2015 Ambassador Dave Grohl.”
Record Store Day. https://recordstoreday.com/CustomPage/3882. Accessed May 25, 2022.
Bibliography 457
Groves, Landon. “Hate Your Favorite Band’s Latest Song? On Spotify, It Might Be a Fake.”
Input Mag, July 17, 2020. https://www.inputmag.com/culture/spotify-is-letting-
scammers-rip-artists-off-in-plain-sight. Accessed June 7, 2022.
Gruenwedel, Erik. “Streaming Fatigue: Too Many Choices, at Too High a Cost?”
MediaPlayNews, November 22, 2021. https://www.mediaplaynews.com/streaming-fati
gue/. Accessed July 19, 2022.
Guiness, Harry. “Spotify Free vs. Premium: Is it Worth Upgrading?” How to Geek, August
4, 2022. https://www.howtogeek.com/350288/spotify-free-vs-premium-is-it-worth-
upgrading/. Accessed April 3, 2023.
Haliday, Josh. “Limewire Shut Down by Federal Court.” Guardian, October 27, 2010,
https:// w ww.theg u ard i an.com/ t ec h nol o gy/ 2 010/ o ct/ 2 7/ l imew i re- s hut- d own.
Accessed June 7, 2022.
Hansell, Saul. “Amazon Launches a Music Store, Not a Service.” New York Times Bits Blog,
September 25, 2007. https://bits.blogs.nytimes.com/2007/09/25/amazon-launches-a-
music-store-not-a-service. Accessed June 7, 2022.
Hardisty, Brad. “The Reel History of Analog Tape Recording.” Performer, May 4, 2015.
https://performermag.com/home-recording/the-reel-history-of-analog-tape-record
ing/. Accessed April 29, 2022.
Harrison Tape Guides. New York: Weiss Publishing Corp., 1956–1976.
Harrington, Richard. “The Record Industry Goes To War On Home Taping.” The
Washington Post, June 15, 1980. https://www.washingtonpost.com/archive/lifestyle/
1980/06/15/the-reocrd-industry-goes-to-war-on-home-taping/80be4100-3fa2-4f73-
8999-1efb6dd282d9/. Accessed April 3, 2023.
Harvey, Eric. “How Smart Speakers Are Changing the Way We Listen to Music.” Pitchfork,
June 29, 2018. https.//pitchfork.com/features/article/how-smart-speakers-are-chang
ing-the-way-we-listen-to-music/. Accessed June 2, 2022.
Harvilla, Bob. “How Soundscan Changed Everything We Knew About Popular Music.”
The Ringer, May 25, 2021. https://www.theringer.com/music/2021/5/25/22452539/
soundscan-billboard-charts-streaming-numbers. Accessed November 16, 2022.
Havens, Lyndsey. “Why Did Vinyl Subscriptions Spike This Year—And Will It Continue?”
Billboard, December 18, 2020. https://www.billboard.com/articles/business/9501250/
vinyl-subscriptions-spike-2020-bright-eyes/. Accessed April 3, 2023.
Hawtin, Steve. “Songs from the 1900s.” The World’s Music Charts. https://tsort.info/music/
ds1900.htm. Accessed May 9, 2022.
Heisler, Y. “Steve Jobs Was Never a Fan of Subscription Music.” Engadget, May 19, 2014.
https://www.engadget.com/2014-05-19-steve-jobs-was-never-a-fan-of-subscription-
music.html. Accessed June 13, 2022.
Herndon, Holly. “Holly + .” Holly Herndon. http://www.hollyherndon.com/holly.
Accessed June 2, 2022.
Heylin, Clinton. Bootleg: The Secret History of the Other Recording Industry. New York: St.
Martin’s Griffin, 1994.
Hiatt, Brian. “Napster Offers Labels $1 Billion in Licensing Fees.” MTV News, February
20, 2001. https://www.mtv.com/news/5r4iyc/napster-offers-labels-1-billion-in-licens
ing-fees /. Accessed April 3, 2023.
Hiltzik, Michael A. “Viacom to Buy MTV and Showtime in Deal Worth $667.5 Million.”
Los Angeles Times, August 27, 1985. https://www.latimes.com/archives/la-xpm-1985-
08-27-fi-25404-story.html. Accessed April 3, 2023.
Hirsch, Julian. “Advent 201 Cassette Tape Deck.” Stereo Review, October 1, 1971, 40–42.
Hirsch, Julian. “Julian Hirsch Hits the Cassette Decks.” Stereo Review, November 1,
1970, 56–69.
458 Bibliography
Hissong, Samantha. “Kings of Leon Will Be the First Band to Release an Album as an
NFT.” Rolling Stone, March 3, 2021. https://www.rollingstone.com/pro/news/kings-of-
leon-when-you-see-yourself-album-nft-crypto-1135192/. Accessed April 29, 2022.
Hissong, Samantha. “Shawn Mendes’ Manager Is Helping Launch a New NFT
Marketplace.” Rolling Stone, March 2, 2021. https://www.rollingstone.com/pro/
news/shawn-mendes-andrew-gertler-nft-marketplace-sturdy-exchange-1135558/.
Accessed April 29, 2022.
“The History of 78 RPM Recordings.” Yale University Library. https://web.library.yale.
edu/cataloging/music/historyof78rpms. Accessed May 9, 2022.
Hochberg, Bill. “YouTube Won’t Take Down A Deepfake Of Jay-Z Reading Hamlet—To
Sue, Or Not To Sue.” Forbes, May 18, 2020. https://www.forbes.com/sites/williamh
ochberg/2020/05/18/to-sue-or-not-to-sue---that-is-the-jay-zs-deepfake-quest ion.
Accessed June 2, 2022.
Hogan, Marc. “Universal Settles Influential Eminem Digital-Revenues Lawsuit.” Spin,
October 31, 2012. https://www.spin.com/2012/10/universal-settles-influential-emi
nem-digital-revenue-lawsuit/. Accessed June 17, 2022.
Holland, Bill. “Audio- Only Home Taping Bill Readied on Senate Side.” Billboard,
September 7, 1985, 1+.
Holt, Kris. “You Can’t Copyright AI-Created Art, According to US Officials.” Engadget,
February 21, 2022. https://www.engadget.com/us-copyright-offi ce-art-ai-creativity-
machine-190722809.html. Accessed June 7, 2022.
Homewood, Ben. “UMG and Tidal Plot New Streaming Model to Benefit Artists and Fans.”
Music Week, January 31, 2023. https://www.musicweek.com/digital/read/umg-and-
tidal-plot-new-streaming-model-to-benefit-artists-and-fans. Accessed January 31, 2023.
Horrigan, John. “Home Broadband Adoption 2006: Part 1. Broadband Adoption in the
United States.” Pew Research Center, May 28, 2006. https://www.pewresearch.org/internet/
2006/05/28/part-1-broadband-adoption-in-the-united-states/. Accessed May 25, 2022.
Houghton, Bruce. “Spotify Cuts Off Playlist Payola Service SpotLister.” Hypebot, March
19, 2018. https://www.hypebot.com/hypebot/2018/03/spotify-cuts-off-playlist-pay
ola-service-spotlister.html. Accessed June 7, 2022.
Hu, Jim. “About Royalty Rates.” The Trichordist. https://thetrichordist.com/category/roya
lty-rates-2/. Accessed May 25, 2022.
Hu, Jim. “Pressplay Comes to Life after Long Wait.” CNET, April 14, 2002. https://www.
cnet.com/tech/s ervi ces- and- s oftware/pressp l ay- comes-to-l ife-after-l ong-wait/.
Accessed June 7, 2022.
Hubbert, Cecelia. “Midler v. Ford Motor Co.” University of Denver Sturm College of Law.
https://w ww.law.du.edu/documents/spor ts-and-entertainment-law-journal/case-
summaries/1988-midler-v-ford-motor-co.pdf. Accessed June 7, 2022.
“Hulu.com Proves Victorious over Skeptics.” SFGATE.com, December 23, 2008. https://
www.sfgate.com/entertainment/article/Hulu-com-proves-victorious-over-skeptics-
3257176.php. Accessed June 2, 2022.
IBM. “What Is Machine Learning?” IBM Cloud Education. https://www.ibm.com/cloud/
learn/machine-learning. Accessed June 2, 2022.
“IBM PC AT.” Old-Computers.com. http://www.old-computers.com/museum/computer.
asp?st=1&c=185. Accessed May 23, 2022.
Ihaza, Jeff. “Tory Lanez Released an ‘NFT Album,’ and Then Things Got Extra Weird.”
Rolling Stone, September 30, 2021. https://www.rollingstone.com/music/music-featu
res/tory-lanez-nft-legal-questions-1233890/. Accessed April 29, 2022.
Bibliography 459
Imbler, Sabrina. “The Best Keurig Machine (But We Really Don’t Recommend It).”
Wirecutter, September 10, 2021. https://www.nytimes.com/wirecutter/reviews/best-
keurig-machine/#the-drawbacks-to-keurig-machines-and-why-we-dont-recomm
end-them. Accessed June 2, 2022.
Immink, Kees A. “The Compact Disc Story.” Journal of the Audio Engineering Society 46,
no. 5 (May 1998): 458–65.
Immink, Kees A. “Shannon, Beethoven and the Compact Disc.” IEEE Information Theory
Society Newsletter, December 2007, 42–46.
Impoco, Jim. “The Beatles Suck. Yeah, We Said That.” Newsweek, February 3, 2014. https://
www.newsweek.com/beatles-suck-yeah-we-said-227748. Accessed May 4, 2022.
Ingham, Tim. “‘Fake Artists’ Have Billions of Streams on Spotify. Is Sony Now Playing the
Service at Its Own Game?” Rolling Stone, May 15, 2019. https://www.rollingstone.com/
pro/features/fake-artists-have-billions-of-streams-on-spotify-is-sony-now-playing-
the-service-at-its-own-game-834746/. Accessed June 13, 2022.
Ingham, Tim. “Daniel Ek Says Spotify’s Subscription Growth is Increasing But That’s Not
True in Monetary Terms.” Music Business Worldwide, April 29, 2020. https://www.mus
icbusinessworldwide.com/daniel-ek-says-spotif ys-subscription-growth-is-increas
ing-but-thats-not-true-in-monetary-terms/. Accessed June 13, 2022.
Ingham, Tim. “It’s Happened: 100,000 Tracks Are Now Being Uploaded to Streaming
Services Like Spotify Each Day.” Music Business Worldwide, October 6, 2022. https://
www.musicbusinessworldwide.com/its-happened-100000-tracks-are-now-being-
uploaded/. Accessed December 4, 2022.
Ingham, Tim. “Over 66% of All Music Listening in the US Is Now of Catalog Records
Rather Than New Releases.” Music Business Worldwide, July 13, 2021. https://www.mus
icbusinessworldwide.com/over-66-of-all-music-listening-in-the-us-is-now-of-cata
log-records-rather-than-new-releases/. Accessed June 7, 2022.
Ingham, Tim. “So . . . How Much Did Tiktok Actually Pay the Music Industry from Its
$4bn in Revenues Last Year?” Music Business Worldwide, July 14, 2022. https://www.
musicbusinessworldwide.com/so-how-much-did-tiktok-actually-pay-the-music-
industry-from-its-4bn-in-revenues-last-year/. Accessed July 25, 2022.
Ingham, Tim. “The Great Big Spotify Scam: Did a Bulgarian Playlister Swindle Their Way
to a Fortune on Streaming Service?” Music Business Worldwide, February 20, 2018.
https://www.musicbusinessworldwide.com/great-big-spotify-scam-bulgarian-playlis
ter-swindle-way-fortune-streaming-service/. Accessed June 7, 2022.
Ingham, Tim. “Yes, Lucian Grainge Has Banned All Exclusives. No It’s Not All About
Frank Ocean.” Music Business Worldwide, August 25, 2016. https://www.musicbusine
ssworldwide.com/yes-lucian-grainge-banned-streaming-exclusives-umg-no-not-
frank-ocean/. Accessed June 5, 2022.
Ingraham, Nathan. “iTunes Store at 10: How Apple Built a Digital Media Juggernaut.” The
Verge, April 26, 2013. https://www.theverge.com/2013/4/26/4265172/itunes-store-at-
10-how-apple-built-a-digital-media-juggernaut. Accessed June 7, 2022.
Interactive Advertising Bureau. “IAB U.S. Podcast Advertising Revenue Study: Full-Year
2021 Results & 2022–2024 Growth Projections.” https://www.iab.com/wp-content/
uploads/2022/05/IAB-FY-2021-Podcast-Ad-Revenue-and-2022-2024-Growth-Projec
tions_FINAL.pdf. Accessed November 29, 2022.
International Federation of the Phonographic Industry. “IFPI Releases Engaging with
Music 2021.” October 21, 2021. https://www.ifpi.org/ifpi-releases-engaging-with-
music-2021/. Accessed June 5, 2022.
460 Bibliography
International Federation of the Phonographic Industry. “Engaging with Music 2022.”
https://www.ifpi.org/wp-content/uploads/2022/11/Engaging-with-Music-2022_full-
report-1.pdf. Accessed November 28, 2022.
International Federation of the Phonographic Industry. “Rewarding Creativity: Fixing
the Value Gap (excerpt from Global Music Report 2017).” https://web.archive.org/web/
20190221033533/https:/www.ifpi.org/downloads/GMR2017_ValueGap.pdf. Accessed
May 25, 2022.
Isaac, Mike. Super Pumped: The Battle for Uber. New York: W.W. Norton, 2019.
“ITA Takes Giant Steps to Push Quality Tape.” Billboard, May 20, 1972, 37.
“iTunes Store and DRM-Free Music: What You Need to Know.” Macworld, January 7,
2009. https://www.macworld.com/article/1138000/drm-faq.html. Accessed June
7, 2022.
Jackson, John. A House on Fire: The Rise and Fall of Philadelphia Soul. Oxford: Oxford
University Press, 2004.
Jacobs Media. “Techsurvey 2022 Results.” https://jacobsmedia.com/techsurvey-2022-
results/. Accessed November 22, 2022.
Jafee, Larry. Record Store Day: The Most Improbable Comeback of the 21st Century. Los
Angeles: Rare Bird, 2022.
Jeffries, Dan. “The Musician in the Machine.” Google Magenta AI. https://magenta.ten
sorfl ow.org/musician-in-the-machine. Accessed June 2, 2022.
Jennn, London. “Tory Lanez Says His NFT Album Is Reselling on a Secondary Market for
$100k.” AllHipHop.com, August 25, 2021. https://allhiphop.com/news/tory-lanez-says-
his-nft-album-is-selling-for-100k/. Accessed April 29, 2022.
Jeong, Sarah. “A $1.6 Billion Spotify Lawsuit Is Based on a Law Made for Player Pianos.”
The Verge, March 14, 2018. https://www.theverge.com/2018/3/14/17117160/spotify-
mechanical-license-copyright-wixen-explainer. Accessed June 7, 2022.
Jobs, Steve. “Thoughts on Music.” Apple, February 6, 2007. https://web.archive.org/web/
20071223160841/https://www.apple.com/hotnews/thoughtsonmusic/. Accessed June
7, 2022.
Juang, B. H., and Lawrence Rabiner. Fundamentals of Speech Recognition. London: Pearson
College Division, 1993.
Jukebox Licensing Office. “History of the Jukebox License Office.” http://www.jukeboxlice
nse.org/history.htm. Accessed April 29, 2022.
Kafka, Peter. “Grooveshark, the Free Music Service That Used to Scare the Labels Gives
Up.” Vox, April 30, 2015. https://www.vox.com/2015/4/30/11562064/grooveshark-the-
free-music-service-that-used-to-scare-the-big-labels. Accessed June 7, 2022.
Kahney, Keander. “Straight Dope on the iPod’s Birth.” Wired, October 17, 2006. https://
www.wired.com/2006/10/straight-dope-on-the-ipods-birth/. Accessed May 23, 2022.
Katz, Mark. Capturing Sound: How Technology Has Changed Music. Berkeley: University
of California Press, 2004.
Kaufman, Gil. “File- Sharing Networks Can Be Liable for Copyright Infringement,
Supreme Court Rules.” MTV News, June 27, 2005. https://www.mtv.com/news/n3n
kvo/f ile-sharing-networks-can-b e-liable-for-copyrig ht-infringements-supreme-
court-rules/. Accessed April 3, 2023.
Kawaida, Michael. “Mixtapes: A Brief History Of Hip-Hop’s Ever Evolving Tool.” HNHH
(HotNewHipHop), February 25, 2010. https://www.hotnewhiphop.com/mixtapes-
a-brief-history-of-hip-hops-ever-evolving-tool-news.103882.html. Accessed April
29, 2022.
Bibliography 461
Keyes, Daniel. “Amazon Has Surpassed 150 Million Prime Subscribers Globally.” Business
Insider, February 3, 2020. https://www.businessinsider.com/amazon-surpasses-150-
million-prime-subscribers-2020-2. Accessed June 5, 2022.
King, Ashley. “German Pop Artist Mike Singer Uses ‘Stream-to-Unlock’ Trick to Boost
Streaming Algorithms.” Digital Music News, July 14, 2020. https://www.digitalmu
sicnews.com/2020/07/14/mike-singer-stream-to-unlock-campaign/. Accessed June
7, 2022.
King, Ashley. “For the First Time Ever, Annual Music Streams Top 1 Trillion—In the
US Alone.” Digital Music News, November 30, 2022. https://www.digitalmusicnews.
com/2022/11/30/annual-music-streams-top-one-trillion-us-alone. Accessed January
31, 2023.
King, Ashley. “Virtual Characters Are Hitting the Spotify Charts.” Digital Music News,
July 21, 2021. https://www.digitalmusicnews.com/2021/07/21/virtual-characters-
metaverse-spotify-charts/. Accessed June 2, 2022.
King, Ashley. “YouTube Music Exec Calls TikTok Short Videos ‘Junk Food’—‘It Has to
Lead You to Long-Form Content So It’s Not Empty Calories.’” Digital Music News, July
14, 2022. https://www.digitalmusicnews.com/2022/07/08/youtube-music-exec-tik
tok-junk-food/. Accessed July 14, 2022.
Klein, Joshua. “Robert Christgau: Christgau's Consumer Guide: Albums Of The '90s.”
AV Club, March 29, 2002. https://www.avclub.com/robert-christgau-christgaus-consu
mer-guide-albums-of-1798193851. Accessed April 7, 2023.
Knight, Shawn. “Spotify Hits 205 Million Premium Subscribers, but Financial Losses
Swell.” Techspot. January 31, 2023. https://www.techspot.com/news/97444-spotify-
hits-205-million-premium-subscribers-but-financial.html. Accessed January 31, 2023.
Knoedelsleder, William. “$1 Million in Suspected ‘New Payola’ Is Probed.” Los Angeles
Times, December 22, 1987. https://www.latimes.com/archives/la-xpm-1987-12-22-
mn-30675-story.html. Accessed April 3, 2023.
Knopper, Steve. Appetite for Self- Destruction. Seattle: CreateSpace Independent
Publishing Platform, 2017.
Knopper, Steve. “iTunes’ 10th Anniversary: How Steve Jobs Turned the Industry Upside
Down.” Rolling Stone, April 26, 2013. https://www.rollingstone.com/culture/culture-
news/itunes-10th-anniversary-how-steve-jobs-turned-the-industry-upside-down-
68985/. Accessed June 7, 2022.
Knowles, Tim. “To Be or Not To Be Me: Jay-Z Asks Big Question.” Times (London), April
30, 2020. https://www.thetimes.co.uk/article/to-be-or-not-to-be-me-jay-z-asks-big-
question-hk6vffl0m. Accessed June 2, 2022.
Komurki, John. Cassette Cultures: Past and Present of a Musical Icon.
Salenstein: Benteli, 2019.
Labaton, Stephen. “5 Music Companies Settle Federal Case On CD Price-Fixing.”
New York Times. May 11, 2000. https://www.nytimes.com/2000/05/11/business/5-
music-companies-settle-federal-case-on-cd-price-fixing.html. Accessed April 4, 2023.
Lafayette Radio Electronics Catalogs. “Lafayette Radio Electronics.” https://worldradio
history.com/Lafayette_Catalogs.htm. Accessed April 29, 2022.
Lawler, Richard. “Kanye West’s $200 Stem Player Will Be the Only Way to Get His Next
Album, Donda 2.” The Verge, February 18, 2022. https://www.theverge.com/2022/
2/18/22940748/donda-2-stem-player-kanye-west-exclusive-music. Accessed June
7, 2022.
LeBel, C. J. “Tape or Disc?” High Fidelity, October 1, 1959, 56+.
462 Bibliography
Leeds, Jeff. “EMI Agrees to Fine to Resolve Payola Case.” New York Times. June 16, 2006,
https://www.nytimes.com/2006/06/16/business/worldbusiness/emi-agrees-to-fine-
to-resolve-payola-case.html. Accessed April 4, 2023.
Leeds, Jeff. “Radio Broadcasters Agree to Fine over Payola.” New York Times. March 6,
2007, https://www.nytimes.com/2007/03/06/technology/06iht-payola.4812569.html.
Accessed April 4, 2023.
Leight, Elias. “Fake Streams Could Be Costing Artists $300 Million a Year.” Rolling Stone,
June 18, 2019. https://www.rollingstone.com/pro/features/fake-streams-indie-labels-
spotify-tidal-846641/. Accessed June 7, 2022.
Leight, Elias. “Sped-Up Songs Are Taking Over TikTok and Driving Songs Up the Charts.”
Billboard, November 10, 2022. https://www.billboard.com/pro/sped-up-songs-tiktok-
streaming-charts/. Accessed November 16, 2022.
Leight, Elias. “Think You Have a Hit? Make Sure It’s the First Song on Your Album.”
Rolling Stone, March 7, 2019. https://www.rollingstone.com/music/music-features/
why-the-first-song-on-the-album-is-the-best-803283/. Accessed June 5, 2022.
Leight, Elias. “TikTok Pays Artists ‘Almost Nothing’ in Music Royalties—And the
Industry Is Losing Patience.” Billboard, November 7, 2022. https://www.billboard.com/
pro/tiktok-pays-artists-little-music-royalties-insiders-fed-up/. Accessed November
16, 2022.
Leight, Elias. “Want to Get on the Radio? Have $50,000?” Rolling Stone, August 6, 2019.
https://www.rollingstone.com/pro/features/radio-stations-hit-pay-for-play-867825/.
Accessed April 4, 2023.
Leslie, John, and Ross Snyder. “History of The Early Days of Ampex Corporation.”
AES Historical Committee, December 17, 2010. https://www.aes.org/aeshc/docs/
company.histories/ampex/leslie_snyder_early-days-of-ampex.pdf. Accessed April
4, 2023.
Lessig, Lawrence. Code: And Other Laws of Cyberspace. New York: Basic Books, 1999.
Lessig, Lawrence. “Laws that Choke Creativity.” TED Talks, https://www.ted.com/talks/
lawrence_lessig_laws_that_choke_creativity. Accessed April 29, 2022.
Leswing, Kif. “Apple Music Has Reportedly Passed Spotify in Paid Subscribers in the US.”
CNBC, April 5, 2019. https://www.cnbc.com/2019/04/05/apple-music-has-reportedly-
passed-spotify-in-paid-subscribers-in-the-us.html. Accessed June 5, 2022.
Leval, Pierre N. “Toward a Fair Use Standard.” Harvard Law Review 103, no. 5
(1990): 1105–36. https://doi.org/10.2307/1341457. Accessed April 4, 2023.
Levine, Robert. “For the Record: How Vinyl Got Its Groove Back—To the Tune of a Billion
Dollars.” Billboard, May 25, 2022. https://www.billboard.com/pro/vinyl-boom-analy
sis-for-the-record. Accessed July 21, 2022.
Lindvall, Helienne. “Behind the Music: Is the Long Tail a Myth.” Guardian, January 8,
2009. https://www.theguardian.com/music/musicblog/2009/jan/08/long-tail-myth-
download. Accessed June 7, 2022.
Lipshutz, Jason. “RIP Lead Singles: Why Hip-Hop Titans Are Dropping Full Albums
All at Once,” Billboard, June 21, 2022, https://www.billboard.com/pro/drake-future-
album-release-strategy-hip-hop-singles/. Accessed April 4, 2023.
“List of Most Streamed YouTube Videos.” Wikipedia. https://en.wikipedia.org/wiki/
List_of_most-viewed_YouTube_videos. Accessed June 7, 2022.
Lomas, Natasha. “Apple Pay and iOS App Store Under Formal Antitrust Probe in Europe.”
TechCrunch, June 16, 2020. https://techcrunch.com/2020/06/16/apple-pay-and-ios-
app-store-under-formal-antitrust-probe-in-europe/. Accessed June 7, 2022.
Bibliography 463
Longdon, Victoria. “Why Is a CD 74 Minutes Long? It’s because of Beethoven.” Classicfm.
com, May 3, 2019. https://www.classicfm.com/discover-music/why-is-a-cd-74-minu
tes/. Accessed May 9, 2022.
Lopez, Mark Hugo, and Daniel Dockterman. “U.S. Hispanic Country-of-Origin Counts
for Nation, Top 30 Metropolitan Areas.” Pew Research Center. May 26, 2011. https://
assets.pewresearch.org/wp-content/uploads/sites/7/reports/142.pdf. Accessed April
5, 2023.
“The Loudness Wars: Why Music Sounds Worse.” NPR, December 31, 2009. https://
www.npr.org/2009/12/31/122114058/the-loudness-wars-why-music-sounds-worse.
Accessed May 23, 2022.
Luminate. “Top Entertainment Trends for 2023: What the Data Says, As Presented at
SXSW 2023.” https://luminatedata.com/reports/sxsw-top-entertainment-trends-for-
2023/. Accessed April 7, 2023.
Luminate. “U.S. Year-End Music Report for 2022.” https://luminatedata.com/reports/
luminate-2022-u-s-year-end-report/. Accessed April 5, 2023.
“Lyrics of ‘If it Wasn’t for the Irish and the Jews.’” The WashingtonSquare Harp and
Shamrock Orchestra. https://wshso.wordpress.com/tunes/songs/irish-jews-lyrics/.
Accessed May 9, 2022.
Mack, Zachary. “How Streaming Affects the Lengths of Songs.” The Verge, May 28, 2019.
https://www.theverge.com/2019/5/28/18642978/music-streaming-spotify-song-len
gth-distribution-production-switched-on-pop-vergecast-interview. Accessed June
5, 2022.
“Madonna Swears at Music Pirates.” BBC News, April 22, 2003. http://news.bbc.co.uk/2/
hi/technology/2962475.stm. Accessed June 7, 2022.
Mann, Charles. “The Hot New Bad Idea.” Inside. December 2000.
“March 1982: ‘I Want My MTV’ Campaign Launched.” Totally80s.com. https://www.tot
ally80s.com/article/march-1982-i-want-my-mtv-campaign-launched. Accessed May
4, 2022.
Marcus, Greil. Like a Rolling Stone: Bob Dylan at the Crossroads. New York:
PublicAffairs, 2006.
Mardesich, Jodi. “How the Internet Hits Big Music.” CNN Money, May 10, 1999. https://
money.cnn.com/s/fortune/fortune_archive/1999/05/10/259548/index.html. Accessed
June 7, 2022.
Margolis, Dan. “Music to a Customers Ears: CD Buyers Can Hear Them First in Stores.”
Los Angeles Times, June 14, 1995. https://www.latimes.com/archives/la-xpm-1995-06-
14-fi-13105-story.html. Accessed April 4, 2023.
Marketing Charts. “What Do Ad Loads Look Like On Internet Radio?” https://www.mark
etingcharts.com/digital-66030. Accessed April 29, 2022.
Marks, Craig, and Rob Tannenbaum. I Want My MTV. New York: Dutton, 2011.
Marr, Bernard. “The Key Definitions of Artificial Intelligence That Explains Its
Importance.” Forbes, February 14, 2018. https://www.forbes.com/sites/bernardmarr/
2018/02/14/the-key-definitions-of-artificial-intelligence-ai-that-explain-its-importa
nce/. Accessed June 7, 2022.
Matchett, Karl. “Barcelona Agree £235m Camp Nou and Kit Sponsorship Deal with
Spotify.” Yahoo News, March 15, 2022. https://news.yahoo.com/barcelona-agree-
235m-camp-nou-204133143.html. Accessed June 7, 2022.
McCormick, Rich. “The Awesome ‘Guardians of the Galaxy’ Mixtape Will Be Released on
Cassette.” The Verge, October 21, 2014. https://www.theverge.com/2014/10/21/7026
464 Bibliography
095/guardians-of-the-galaxy-mixtape-will-be-released-on-cassette-tape. Accessed
April 29, 2022.
McDermott, Jim. “5 Reasons the Major Labels Didn’t Really Blow it With Napster.”
Hypebot. https://www.hypebot.com/hypebot/2015/05/five-reasons-the-major-labels-
didnt-blow-it-with-napster.html. Accessed May 23, 2022.
McGinn, Daniel. “The Buzz Machine.” Boston Globe Sunday, August 7, 2011, http://arch
ive.boston.com/business/articles/2011/08/07/the_inside_story_of_keurigs_rise_to_
a_billion_dollar_coffee_empire/. Accessed April 4, 2023.
McIntyre, Hugh. “Adele’s ‘30’ Was The Bestselling Vinyl Album Of 2021, Taylor Swift
Rules With Several Bestsellers.” Forbes, January 6, 2022. https://www.forbes.com/sites/
hughmcintyre/2022/01/06/adeles-30-was-the-bestselling-vinyl-album-of-2021-tay
lor-swift-rules-with-several-bestsellers/. Accessed May 25, 2022.
Mechanical Licensing Collective. “About the MLC.” https://www.themlc.com/our-story.
Accessed June 7, 2022.
Mechanical Licensing Collective. “How Can YOU Play Your Part™?” https://www.themlc.
com/play-your-part. Accessed July 17, 2022.
Melvin, Chuck. “AT&T Creates ‘Ultimate Jukebox’ for Rock Hall.” Cleveland Plain Dealer,
April 1, 1998.
Michaels, Sean. “Most Music Didn’t Sell a Single Copy in 2008.” Guardian, December 23, 2008.
https://www.theguardian.com/music/2008/dec/23/music-sell-sales. Accessed April 4, 2023.
Metz, Cade. “Lawsuit Takes Aim at the Way AI is Built.” New York Times, November 23,
2022. https://www.nytimes.com/2022/11/23/technology/copilot-microsoft-ai-lawsuit.
html Accessed April 5, 2023.
Miles, Barry. Zappa: A Biography. New York: Grove Press, 2004.
Millard, Andre. America on Record: A History of Recorded Sound. Cambridge: Cambridge
University Press, 2005.
Miller, Larry. “Sync or Swim—Licensing Music for Podcasts.” Musonomics (podcast),
September 28, 2020. http://musonomics.org/sync-or-swim-licensing-music-for-
podcasts/. Accessed April 29, 2022.
Milner, Greg. Perfecting Sound Forever: An Aural History of Recorded Music.
New York: Farrar, Straus and Giroux, 2010.
Mishra, James. “The Dark History of the Jukebox: How the Mafia Used Murder to Build
Music Machine Empires.” Click Track, May 15, 2020. https://www.clicktrack.fm/p/the-
dark-history-of-the-jukebox-how. Accessed November 12, 2022.
MIREX. “MIREX2020 Results.” https://www.music-ir.org/mirex/wiki/2020:MIREX2
020_Results. Accessed August 5, 2020.
Mnookin, Seth. “Universal’s CEO Once Called iPod Users Thieves. Now He’s Giving
Songs Away.” Wired, November 27, 2007. https://www.wired.com/2007/11/mf-morris/
. Accessed June 7, 2022.
Molanphy, Chris. “A Deal with the TV Gods.” Hit Parade (podcast), June 18, 2022. https://
slate.com/podcasts/hit-parade/2022/06/television-is-a-hitmaking-jukebox. Accessed
July 5, 2022.
Molanphy, Chris. “Feat. Don’t Fail Me Now.” Slate, July 31, 2015. https://slate.com/cult
ure/2015/07/the-history-of-featured-rapp ers-and-other-featured-artists-in-p op-
songs.html. Accessed May 25, 2022.
Molanphy, Chris. “The War Against the Single.” Hit Parade (podcast), September 29,
2017. https://chris.molanphy.com/hit-parade-the-great-war-against-the-single-edit
ion/. Accessed May 23, 2022.
Bibliography 465
Moore, Geoffrey. Crossing the Chasm. New York: Harper Business, 2014.
“More to Love: Target and Michael Bublé Reveal Exclusive Bonus Tracks on His New
Album ‘To Be Loved.’ ” CNN.com, April 24, 2018. https://www.cnn.com/2018/04/24/
us/monkey-selfie-peta-appeal/index.html. Accessed June 5, 2022.
Morton, David L. “The History of Magnetic Recording in the United States, 1888–1978.”
PhD diss., Georgia Institute of Technology, 1995.
Morton, David L. Sound Recording: The Life Story of a Technology. Baltimore: Johns
Hopkins University Press, 2004.
Mountain, Toby. “The Birth of Stereo Recording.” Northeastern Digital. https://www.nort
heasterndigital.com/post/the-birth-of-stereo-recording. Accessed April 29, 2022.
“MTV News Interviews Sean Parker and Shawn Fanning,” https://www.facebook.com/
watch/?v=447955682687893. Accessed June 7, 2022.
Mulligan, Mark. “Did July 1st 2019 Mark the End of Spotify’s Music Creator Dream?”
Midia Music Industry Blog, October 8, 2021. https://www.midiaresearch.com/blog/did-
july-1st-2019-mark-the-end-of-spotifys-music-creator-dream. Accessed June 7, 2022.
Mulligan, Mark. “How the DNA of a Hit Has Changed Over 20 Years.” Midia Research,
July 13, 2020. https://www.midiaresearch.com/blog/how-the-dna-of-a-hit-has-chan
ged-over-20-years. Accessed June 13, 2022.
Music Business Association. “Smart Speakers Driving New Music Consumption Habits,
Says New AudienceNet Survey.” Music Business Association, October 10, 2018. https://
musicbiz.org/news/smart-speakers-driving-new-music-consumption-habits-says-
new-audiencenet-study/. Accessed June 2, 2022.
“The Music You Want Wherever You Want It,” John Okolowicz collection of publications
and advertising on radio and consumer electronics (Accession 2014.277), Hagley
Museum & Library, Wilmington, DE 19807.
“MusicNet Launches Battle for Fans.” BBC, December 4, 2001. http://news.bbc.co.uk/2/
hi/entertainment/1691108.stm. Accessed June 7, 2022.
“Nakamichi’s $1,100 Cassette Deck.” High Fidelity. August 1, 1973, 33–35.
“Napster Settlement Offer Rejected.” CBS News, February 11, 2001. https://www.cbsnews.
com/news/napster-settlement-offer-rejected/. Accessed June 2, 2022.
Nash, Michael. “Something New: Artificial Intelligence and the Perils of Plunder.” Music
Business Worldwide, February 14, 2023. https://www.musicbusinessworldwide.com/
michael-nash-universal-something-artificial-intelligence-and-the-perils-plunder/.
Accessed February 15, 2023.
Neer, Richard. FM: The Rise and Fall of Rock Radio. New York: Villard, 2001.
Nelson, Chris. “Recordable CDs Undercut Traditional Bootleg Market.” MTV.com News,
May 21, 1998. https://www.mtv.com/news/dsuz61/recordable-cds-undercut-traditio
nal-bootleg-market/. Accessed April 5, 2023.
Nelson, Chris. “Music Piracy Shifts to Recordable CDs, Record Industry Says.” MTV.com
News, August 19, 1999. https://www.mtv.com/news/g5btet/music-piracy-shifts-to-rec
ordable-cds-record-industry-says/. Accessed May 23, 2022.
“Network Radio Averaged 16 Minutes of Spots per Hour, as Rates Crept Up.” InsideRadio,
November 29, 2018. https://www.insideradio.com/free/network-radio-averaged-16-
minutes-of-spots-per-hour-as-rates-crept-up/article_97ff6ae0-f3a8-11e8-b201-8770a
1fdf2ed.html. Accessed April 29, 2022.
“New Music Machine Thrills All Hearers at First Test Here.” New York Times, October 7,
1925, https://www.nytimes.com/1925/10/07/archives/new-music-machine-thrills-all-
hearers-at-first-test-here-researches.html. Accessed April 4, 2023.
466 Bibliography
Nicas, Jack, and David McCabe. “Their Businesses Went Virtual. Then Apple Wanted a
Cut.” New York Times. July 28, 2020. https://www.nytimes.com/2020/07/28/technol
ogy/apple-app-store-airbnb-classpass.html. Accessed April 4, 2023.
“Nielsen Report Puts Radio at the Center of the Audio Universe.” InsideRadio, June 30,
2022. https://www.insideradio.com/nielsen-report-puts-radio-at-the-center-of-the-
audio-universe/article_a3188d52-f7f1-11ec-8617-6b7586059cd3.html. Accessed
January 30, 2023.
“Nikki Sixx Quotes.” Inspirational Stories. https://www.inspirationalstories.com/quotes/t/
nikki-sixx/. Accessed June 28, 2022.
Niasse, Amina. “Vinyl Record Sales Climb Just 1% after Years of Rapid Growth.”
Bloomberg, July 14, 2022. https://www.bloomberg.com/news/articles/2022-07-14/
vinyl-record-sales-climb-just-1-after-years-of-rapid-growth. Accessed July 21, 2022.
Nicas, Jack. “YouTube Tops 1 Billion Hours of Video a Day, on Pace to Eclipse TV.” Wall
Street Journal, February 27, 2017. https://www.wsj.com/articles/youtube-tops-1-bill
ion-hours-of-video-a-day-on-pace-to-eclipse-tv-1488220851. Accessed April 4, 2023.
Nolan, Beatrice. “AI Art Generators Face Separate Copyright Lawsuits from Getty Images
and a Group of Artists.” Business Insider, January 19, 2023. https://www.businessinsi
der.com/ai-art-artists-getty-images-lawsuits-stable-diffusion-2023-1. Accessed
January 31, 2023.
Nolasco, Stephanie. “Elvis Presley’s First Appearance on the Ed Sullivan Show
Remembered 64 Years Later.” Foxnews.com, September 9, 2020. https://www.foxn
ews.com/entertainment/elvis-presley-ed-sullivan-show-anniversary. Accessed May
4, 2022.
NPR and Edison Research. “Smart Audio Report.” June 17, 2022. https://www.edisonr
esearch.com/smart-audio-report-2022-from-npr-and-edison-research/. Accessed
April 4, 2023.
O’Brien, T. “Listen to an AI Sing an Uncanny Rendition of ‘Jolene.’ ” Engadget, November
2, 2022. https://www.engadget.com/holly-plus-holly-herndon-dolly-parton-jolene-ai-
cover-204750236.html. Accessed November 3, 2022.
O’Malley, Chris. “A New Spin.” Time, August 24, 1998. https://content.time.com/time/sub
scriber/article/0,33009,988955,00.html.
“Official Charts Company Shakes Up Rules to Stop the Ed Sheeran Effect.” Guardian, June
27, 2017. https://www.theguardian.com/music/2017/jun/27/official-charts-company-
shakes-up-rules-support-new-talent-ed-sheeran-effect. Accessed April 4, 2023.
Ogunnaike, Lola. “Record Labels Must Pay Shortchanged Performers.” New York Times,
May 5, 2004. https://www.nytimes.com/2004/05/05/arts/record-labels-must-pay-
shortchanged-performers.html. Accessed April 4, 2023.
Oman, Ralph (Register of Copyrights). Report on Copyright Implications of Digital Audio
Transmission Services. Washington, DC: US Copyright Office, 1991).
Open AI. “Open AI Jukebox.” https://openai.com/blog/jukebox/. Accessed June 2, 2022.
Ottens, L. F. “The Compact Cassette for Audio Tape Recorders.” Journal of the Audio
Engineering Society 15, no. 1 (1967): 26–28.
Owens, Buck, and Randy Poe. Buck ‘Em!: The Autobiography of Buck Owens. Lanman,
MD: Backbeat, 2013.
Owsinski, Bobby. “The Music Long Tail Might Not Be That Long After All.” Music 3.0
Blog, November 13, 2019. https://music3point0.com/2019/11/13/music-long-tail-
rebuff ed/#ixzz6QtTtdmzJ. Accessed June 5, 2022.
Bibliography 467
Owsinski, Bobby. “The Streaming Performance Royalty Explained.” Music 3.0 Blog, May
19, 2016. https://music3point0.com/2016/05/19/streaming-performance-royalty/
#ixzz6souR0TTQ. Accessed May 4, 2022.
Pagano, Penny and William Knoedelsleder Jr. “Senate Plans Record Industry Payola
Probe.” Los Angeles Times, April 3, 1986, https://www.latimes.com/archives/la-xpm-
1986-04-03-mn-2589-story.html. Accessed April 4, 2023.
Paine, Andre. “Tencent Music Paying Users up 52% in Q2.” Music Week, August 11, 2020.
https://www.musicweek.com/digital/read/tencent-music-paying-users-up-52-in-q2/
080734. Accessed May 25, 2022.
Pallante, Maria (Register of Copyrights). Copyright and the Music Marketplace.
Washington, DC: US Copyright Office, February 2015.
Pan, Joann. “Grooveshark Circumvents Mobile Bans by Launching an HTML5 Player.”
Mashable, September 5, 2012. https://mashable.com/archive/grooveshark-html5-pla
yer. Accessed June 7, 2022.
Pareles, Jon. “Issue and Debate; Royalties on Recorders and Blank Audio Tapes.” New York
Times, November 21, 1985, https://www.nytimes.com/1985/11/21/arts/issue-and-deb
ate-royalties-on-recorders-and-blank-audio-tapes.html. Accessed April 4, 2023.
Pareles, Jon. “Now on CDs, First 4 Beatles Albums.” New York Times, February 25, 1987,
https://www.nytimes.com/1987/02/25/arts/now-on-cd-s-first-4-beatles-albums.html.
Accessed April 4, 2023.
Pareles, Jon. “Record-It-Yourself Music on Cassette.” New York Times, May 11, 1987,
https://w ww.nytimes.com/1987/05/11/arts/record-it-yourself-music-on-cassette.
html. Accessed April 4, 2023.
Pareles, Jon. “Trying to Get in Tune With the Digital Age; Recording Industry Seeks
a Standard For Distributing Music on the Web.” New York Times, February 1, 1999,
https://www.nytimes.com/1999/02/01/business/trying-get-tune-with-digital-age-
recording-industry-seeks-standard-for.html. Accessed April 4, 2023.
Pareles, Jon. “With a Click, A New Era of Music Dawns.” New York Times, November 15,
1998, https://www.nytimes.com/1998/11/15/arts/music-with-a-click-a-new-era-of-
music-dawns.html. Accessed April 4, 2023.
Parks, Kevin. Music & Copyright in America: Toward the Celestial Jukebox.
Chicago: American Bar Association, 2014.
Passman, Donald. All You Need to Know About the Music Business, 10th Edition.
New York: Simon & Schuster, 2019.
Passy, Jacob. “How Spotify Influences What Songs become Popular (or Not).” Market
Watch, June 18, 2018. https://www.marketwatch.com/story/how-spotify-infl uences-
what-songs-become-popular-or-not-2018-06-18. Accessed June 5, 2022.
Peek, J. B. H. Origins and Successors of the Compact Disc: Contributions of Philips to Optical
Storage. New York: Springer, 2009.
Peek, J. B. H. “The Emergence of the Compact Disk.” IEEE Communications 48, no. 1
(January 2010): 10–17.
Peltz, James F. “Cetec Gauss Bets Standard Tape Can Survive Digital Threat.” Los Angeles
Times, August 11, 1987, https://www.latimes.com/archives/la-xpm-1987-09-01-fi-
5454-story.html. Accessed April 4, 2023.
Peoples, Glenn. “1 in Every 60 U.S. Streams Last Week Was Drake.” Billboard, June 28,
2022, https://www.billboard.com/pro/drake-honestly-nevermind-streams-marketsh
are/. Accessed April 4, 2023.
468 Bibliography
Peoples, Glenn. “Cloud Village Music Subscribers Grew 51%, Revenues Up 38.6% in Q1.”
Billboard, May 24, 2022, https://www.billboard.com/pro/cloud-village-q1-2022-earni
ngs/. Accessed April 4, 2023.
Perzanowski, Aaron, and Jason Schultz. The End of Ownership: Personal Property in the
Digital Economy. Cambridge, MA: MIT Press, 2016.
Peters, Marybeth. “Statement to Internet Subcommittee of the Judiciary Committee.” U.S.
Copyright Office. https://www.copyright.gov/docs/regstat031104.html. Accessed May
9, 2022.
Pex. “Music Became Even More Valuable on YouTube in 2019.” https://pex.com/blog/
state-of-youtube-2019-music-more-valuable/. Accessed May 25, 2022.
Phirtiyal, Sankalp. “Google Expands Jio Partnership with Indian Smartphone, Cloud Tie-
Ups.” Reuters, June 24, 2021. https://www.reuters.com/technology/google-says-cloud-
partnership-with-indias-jio-boost-5g-plans-2021-06-24/. Accessed June 13, 2022.
“Play/Pause: A Look at 30 Years of the Compact Disc.” Business Week, October 10, 2012.
https://www.bloomberg.com/news/photo-essays/2012-10-01/play-pause-a-look-at-
30-years-of-the-compact-disc. Accessed May 23, 2002.
Podtrac. “Podtrac Industry Ranking: Top 20 Podcasts, April 2022.” http://analytics.podt
rac.com/podcast-rankings. Accessed June 20, 2022.
Pope, Adrian. “Clamp Down on Free Spotify? Now Is Not the Time for Knee-Jerk
Decisions.” PIAS: The Independent Blog. https://www.piasgroup.net/blog/clamp-down-
on-free-spotify-now-is-not-the-time-for-knee-jerk-decisions/. Accessed June 5, 2022.
“Popular Music Is More Collaborative Than Ever.” Economist, February 2, 2018, https://
www.economist.com/graphic-detail/2018/02/02/popular-music-is-more-collaborat
ive-than-ever. Accessed April 4, 2023.
“Portable Music Players Enter the Spin Zone.” Wired, September 30, 2009. https://www.
wired.com/2009/09/1001first-cd-players/. Accessed May 23, 2022.
Porter, Jon. “US Patent Office Rules that Artificial Intelligence Cannot Be a Legal
Inventor.” The Verge, April 29, 2020. https://www.theverge.com/2020/4/29/21241251/
artificial-intelligence-inventor-united-states-patent-trademark-offi ce-intellectual-
property. Accessed June 7, 2022.
Powell, W. H. “Top 10 Greatest BuzzFeed Lists Of All Time.” Medium, June 28, 2016.
https://medium.com/@whpeezy/top-10-buzzfeed-lists-of-all-time-that-will-tickle-
your-list-bone-until-you-have-a-listgasm-ffb1ce7cd52c. Accessed June 2, 2022.
Prato, Greg. “David Coverdale Tells the Story Behind Whitesnake’s Iconic “Here I Go
Again” Video.” Consequenceofsound.com, March 13, 2019. https://consequence.net/
2019/03/david-coverdale-whitesnake-here-i-go-again-video/. Accessed May 4, 2022.
Prendiville, Kieran. “Tomorrow’s World: The Compact Disc.” BBC, February 24, 2015.
https://youtu.be/bMp1pSVxoqw. Accessed May 23, 2002.
Pteridis, Alexis. “Mark Ronson: ‘I Was Floundering. I Was Drinking too Much and Giving
Orders.’” Guardian, February 22, 2019, https://www.theguardian.com/music/2019/
feb/22/mark-ronson-i-was-floundering-i-was-drinking-too-much-and-giving-ord
ers. Accessed April 4, 2023.
Public Knowledge. “Public Knowledge Responds to RIAA’s Plans to Sue P2P File Traders.”
Public Knowledge press release, September 7, 2004. https://www.publicknowledge.
org/press-release/public-knowledge-responds-to-riaas-plans-to-sue-p2p-file-traders/
. Accessed June 7, 2022.
Radio and Records Music Airplay Charts. Radio and Records, https://worldradiohistory.
com/Archive-All-Music/Radio_and_Records.htm. Accessed July 13, 2022.
Bibliography 469
Radio Shack Catalogs. Radio Shack. https://www.radioshackcatalogs.com/. Accessed
April 29, 2022.
Raieli, Salvatore. “Generate a Piano Cover With AI.” Medium, November 10, 2022. https://
medium.com/mlearning-ai/generate-a-piano-cover-with-ai-f4178bc9cb30. Accessed
November 16. 2022.
Rasen, Edward. “Compact Discs: Sound of the Future.” Spin, May 1985. https://www.spin.
com/2021/09/compact-discs-sound-of-the-future-2/.
Recording Industry Association of America. “About Piracy.” https://www.riaa.com/resour
ces-learning/about-piracy/. Accessed May 23, 2022.
Recording Industry Association of America. “U.S. Sales Database.” https://www.riaa.
com/u-s-sales-database/. Accessed June 7, 2022.
Resnikoff, Paul. “A Proposal to Build Spotify (from 1982).” Digital Music News, June 21,
2012. https://www.digitalmusicnews.com/2012/06/21/proposes. Accessed November
26, 2022.
Resnikoff, Paul. “Spotify Executive Calls Artist ‘Entitled’ for Requesting Payment of One
Penny Per Stream.” Digital Music News, June 29, 2021. https://www.digitalmusicnews.
com/2021/06/29/spotify-executive-entitled-pay-penny-per-stream/. Accessed June
13, 2022.
Resnikoff, Paul. “What Succeeds on Spotify’s Rap Caviar—A Statistical Analysis.” Digital
Music News, August 18, 2020. https://www.digitalmusicnews.com/2020/08/18/spot
ify-rapcaviar-analysis. Accessed June 13, 2022.
Resnikoff, Paul. “YouTube Music Says It Pays the Same Royalty Rate as Spotify—At Least
on Its Subscription Streams.” Digital Music News, October 9, 2019. https://www.digit
almusicnews.com/2019/10/09/youtube-music-premium-subscription/. Accessed May
25, 2022.
Reynolds, Simon. Rip It Up and Start Again: Postpunk 1978–1984. London: Penguin, 2006.
Richtel, Matt. “File Sharing Sites Found Not Liable For Infringement.” New York Times,
August 20, 2008, https://www.nytimes.com/2004/08/20/business/technology-file-shar
ing-sites-found-not-liable-for-infringement.html. Accessed April 4, 2023.
Richtel, Matt. “Plans to Sell Music on the Internet Raise Antitrust Concerns.” New York
Times, August 7, 2001, https://www.nytimes.com/2001/08/07/business/technology-
plans-to-sell-music-on-the-internet-raise-antitrust-concerns.html/. Accessed April
4, 2023.
Richtel, Matt. “The Napster Decision: Appellate Judges Back Limitations on Copying
Music.” New York Times. February 13, 2001, https://www.nytimes.com/2001/02/13/
business/napster-decision-overview-appellate-judges-back-limitations-copying-
music.html. Accessed April 4, 2023.
Ricker, Thomas. “First Click: Remember When Steve Jobs Said Even Jesus Couldn’t Sell
Music Subscriptions?” The Verge, June 8, 2015. https://www.theverge.com/2015/6/
8/8744963/steve-jobs-jesus-people-dont-want-music-subscriptions. Accessed June
28, 2022.
Ringer, Barbara. The Unauthorized Duplication of Sound Recordings (Copyright Law
Revision Study No. 26). Washington, DC: US Copyright Office, 1961.
“Rio PMP300.” Wikipedia. https://en.wikipedia.org/wiki/Rio_PMP300. Accessed June
17, 2022.
Robertson, Adi. “The US Copyright Office Says an AI Can’t Copyright Its Art.” The Verge,
February 21, 2022. https://www.theverge.com/2022/2/21/22944335/us-copyright-off
ice-reject-ai-generated-art-recent-entrance-to-paradise. Accessed June 2, 2022.
470 Bibliography
Robinson, Kristin. “HitPiece Wanted to Make an NFT for Every Song—Only Its Founders
Forgot to Ask Artists First.” Billboard, February 2, 2022. https://www.billboard.com/
pro/hitpiece-nft-every-song-artist-permission-founders/. Accessed June 5, 2022.
Robitzski, Dan. “Mind-Melting AI Makes Frank Sinatra Sing “Toxic” by Britney Spears.”
Futurism, May 20, 2020. https://futurism.com/mind-melting-ai-frank-sinatra-toxic-
britney-spears. Accessed June 2, 2022.
Roedy, Bill, and David Fisher. What Makes Business Rock. New York: Wiley, 2011.
Rogers, Jude. “Total Rewind: 10 Key Moments in the Life of the Cassette.” Guardian,
August 30, 2013, https://www.theguardian.com/music/2013/aug/30/cassette-store-
day-music-tapes. Accessed April 4, 2023.
Rosen, Gary. Adventures of a Jazz Age Lawyer. Oakland: University of California
Press, 2020.
Rosenberg, Rutger, and Jason Joven. “How To Understand and Use Music Data Analytics
Tools.” Hypebot. https://www.hypebot.com/hypebot/2020/06/how-to-understand-
and-use-music-data-analytics-tools.html. Accessed June 7, 2022.
Rosenblatt, Bill. “Copyright Office’s Section 512 Report Finds the Balance Askew.”
Copyright and Technology, May 26, 2020. https://copyrightandtechnology.com/2020/
05/26/copyright-offi ce-releases-section-512-report/. Accessed May 25, 2022.
Rosenblatt, Bill. “Disruption Comes to Higher-Ed Publishing.” Publishers Weekly, August
19, 2019, https://www.publishersweekly.com/pw/by-topic/digital/content-and-e-books/
article/80947-disruption-comes-to-higher-ed-publishing.html. Accessed April 4, 2023.
Rosenblatt, Bill. “EU Article 13 (Now Article 17) Passes After More Changes, Making
Copyright Filtering More Likely.” Forbes, April 1, 2019. https://copyrightandtechnol
ogy.com/2019/04/01/eu-article-13-now-article-17-passes-after-more-changes-mak
ing-copyright-filtering-more-likely/. Accessed May 25, 2022.
Rosenblatt, Bill. “The Future of Blockchain Technology in the Music Industry.” Journal of
the Copyright Society of the USA 66 (Spring 2019): 271–89.
Rosenblatt, Bill. “The Short, Unhappy Life of Music Downloads.” Forbes, May 7, 2018.
https://www.forbes.com/sites/billrosenblatt/2018/05/07/the-short-unhappy-life-of-
music-downloads/. Accessed June 5, 2022.
Rosenblatt, Bill. “UltraViolet, Hollywood’s Attempt To Control The Digital Video Supply
Chain, Will Shut Down.” Forbes, February 3, 2019. https://www.forbes.com/sites/bil
lrosenblatt/2019/02/03/ultraviolet-hollywoods-attempt-to-control-the-digital-video-
supply-chain-will-shut-down/. Accessed June 2, 2022.
Rosenblatt, Bill. “Vinyl Is Bigger Than We Thought. Much Bigger.” Forbes, September 18,
2018. https://www.forbes.com/sites/billrosenblatt/2018/09/18/vinyl-is-bigger-than-
we-thought-much-bigger/. Accessed April 29, 2022.
Rosenblatt, Bill, Bill Trippe, and Stephen Mooney. Digital Rights Management: Business
and Technology. New York: John Wiley & Sons, 2001.
Røssel, Thomas, and Camilla Louise Brandt. “Polaris Nordic: Digital Music in the
Nordics.” Polaris. https://polarismusichub.com/wp-content/uploads/2020/05/Full_Re
port_Polaris_Nordic_Digital-Music-in-the-Nordics-2020.pdf. Accessed June 5, 2022.
Roth, Maxine. “The Story of Music on Tape.” Tape Recording, January 1, 1968, 44–48.
Ruhlmann, William. Breaking Records: 100 Years of Hits. London: Routledge, 2004.
Ruhlmann, William. “Your Hit Parade: 1957 Review.” All Music Guide. https://www.allmu
sic.com/album/your-hit-parade-1957-mw0000884963. Accessed May 4, 2022.
Russell, Jon. “Spotify Says It Paid $340M to Buy Gimlet and Anchor.” TechCrunch,
February 14, 2019. https://techcrunch.com/2019/02/14/spotify-gimlet-anchor-340-
million/. Accessed June 5, 2022.
Bibliography 471
Rys, Dan. “Lucian Grainge Calls For ‘Updated Model’ For Music Industry: Read His
Memo to UMG Staff.” Billboard, January 12, 2023. https://www.billboard.com/pro/luc
ian-grainge-umg-full-staff-memo-2023-read-message/. Accessed April 7, 2023.
Sanjek, Russell and David. American Popular Music in the 20th Century. Oxford: Oxford
University Press, 1991.
Sarser, David. “Tapes, Disks, and Coexistence.” High Fidelity, March 1, 1955, 44+.
Scattergood, Amy. “Artisans of the Roast.” Los Angeles Times, October 25, 2006. https://
www.latimes.com/food/la-fo-roast102506-story.html. Accessed April 4, 2023.
Schatz, Lake. “Spotify CEO To Artists: ‘You Can’t Record Music Once Every Three To
Four Years And Think That’s Going To Be Enough.’” Consequence of Sound, August
1, 2020. https://consequence.net/2020/08/spotify-daniel-ek-artist-recording-comme
nts/. Accessed June 13, 2022.
Scheirer, Eric. “The End of SDMI.” MP3.com, October 15, 1999. https://web.archive.org/
web/20000229055832/http://www.mp3.com/news/394.html. Accessed June 7, 2022.
Schiff, Devon. “GO AEROSMITH: How ‘Head First’ Became the First Digitally
Downloadable Song 20 Years Ago Today.” Vice.com, June 27, 2014. https://www.vice.
com/en/article/6vapxr/go-aerosmith-how-head-first-became-the-first-song-availa
ble-for-digital-download-20-years-ago-today. Accessed May 23, 2022.
Schumacher-Rasmussen, Eric. “Record Industry Sues Morpheus and Other Decentralized
File Sharing Services.” MTV News, October 3, 2001. https://www.mtv.com/news/gpu
n33/record-industry-sues-morpheus-and-other-decentralized-file-sharing-services/.
Accessed June 7, 2022.
Schultz, Rod. “The Many Facades of DRM.” 2012. https://web.archive.org/
web/20150105090945/http://fortunedotcom.files.wordpress.com/2014/12/2012_
misc_drm.pdf. Accessed May 18, 2023.
Schwann Record and Tape Guides. Schwann. Boston: Schwann Publications, 1971–1978.
Sewell, Amanda. “How Copyright Affected the Musical Style and Critical Reception
of Sample-Based Hip-Hop.” Journal of Popular Music Studies 26, nos. 2– 3
(2014): 295–320.
Shah, Neil. “Didn’t Like That New Album? Another One Is Coming Before You Know
It.” Wall Street Journal, March 26, 2018, https://www.wsj.com/articles/didnt-like-that-
new-album-another-one-is-coming-before-you-know-it-1522076320. Accessed April
4, 2023.
Shah, Neil. “A Chart-Topping Album Sparks Debate Over Length.” Wall Street Journal,
February 6, 2018, https://www.wsj.com/articles/migos-album-culture-ii-tops-charts-
but-sparks-debate-over-its-length-1517848564. Accessed April 4, 2023.
Shannon, Daniel. “Sales of Stereos Are Less Sound.” New York Times, May 23, 1982,
https://www.nytimes.com/1982/05/23/business/sales-of-stereos-are-less-sound-as-
audio-industry-s-markets-slip.html. Accessed April 4, 2023.
Shapiro, Ariel. “Warner Music Adopts SoundCloud’s User-centric Revenue Model.” The
Verge, July 21, 2022. https://www.theverge.com/2022/7/21/23272548/warner-music-
soundcloud-user-centric-model-spotify. Accessed July 22, 2022.
Shelton, Robert. “Happy Tunes on Cash Registers.” New York Times, March 16, 1958,
https://www.nytimes.com/1958/03/16/archives/happy-tunes-on-cash-registers-rec
ord-industry-sees-tape-and-stereo.html. Accessed April 4, 2023.
Shen, Jonathan, and Ruoming Pang. “Tacotron 2: Generating Human-like Speech from
Text.” Google AI Blog, December 19, 2017. https://ai.googleblog.com/2017/12/tacot
ron-2-generating-human-like-speech.html. Accessed June 2, 2022.
472 Bibliography
“A Short History of How Jukeboxes Changed the World.” Rock-Ola. https://www.rock-
ola.com/blogs/news/a-short-history-of-how-jukeboxes-changed-the-world. Accessed
May 25, 2022.
Shotwell, James. “36% of Americans Now Own Smart Speakers: Here’s What That Means
for Musicians.” Haulix, February 26, 2019. https://haulixdaily.com/2019/02/smart-
speakers-music-business. Accessed June 2, 2022.
Sigalos, MacKenzie. “Ethereum Had a Rough September. Here’s Why and How It’s Being
Fixed.” CNBC.com, October 2, 2021. https://www.cnbc.com/2021/10/02/ethereum-
had-a-rough-september-heres-why-and-how-it-gets-fixed.html. Accessed April
29, 2022.
Singh, Manish. “Google invests $4.5 billion in India’s Reliance Jio Platforms.” TechCrunch,
July 15, 2020. https://techcrunch.com/2020/07/15/google-invests-4-5-billion-in-ind
ias-reliance-jio-platforms/. Accessed May 23, 2022.
Sisario, Ben. “A New Spotify Initiative Makes the Big Record Labels Nervous.” New York
Times, September 6, 2018, https://www.nytimes.com/2018/09/06/business/media/spot
ify-music-industry-record-labels.html. Accessed April 4, 2023.
Slattery, Peter. “Scammers Are Gaming Spotify by Faking Collaborations with Famous
Artists.” OneZero, June 1, 2020. https://onezero.medium.com/scammers-are-gaming-
spotify-by-faking-collaborations-with-famous-artists-42d127e370dc. Accessed June
7, 2022.
“SME and WMG the biggest market share winners in 2021.” Music and Copyright Blog,
April 5, 2022. https://musicandcopyright.wordpress.com/2022/04/05/sme-and-wmg-
the-biggest-market-share-winners-in-2021/. Accessed May 25, 2022.
Smith, Dylan. “Tencent Music Subscribers Grow 50% to 42.7 Million—Wall Street Isn’t
Impressed.” Digital Music News, May 12, 2020. https://www.digitalmusicnews.com/
2020/05/12/tencent-music-subscribers-growth/. Accessed June 5, 2022.
Smith, Dylan. “Warner Music Group Inks ‘First-Ever Partnership’ With POAP ‘To Mint
Shared Memories As NFTs.’” Digital Music News, April 14, 2022. https://www.digit
almusicnews.com/2022/04/14/warner-music-group-poap-partnership/. Accessed
May 28, 2022.
Smith, Dylan. “Spotify CEO Daniel Ek Confirms 2023 Price Increase Plans Following
Apple Music Raises.” Digital Music News, October 27, 2022. https://w ww.digit
almusicnews.com/2022/10/27/spotify-price-increase-2023/. Accessed October
27, 2022.
Smith, Ethan, and Jessica E. Vascellaro. “Warner Pulls Music From YouTube.” Wall Street
Journal, December 22, 2008, https://www.wsj.com/articles/SB122980193788724073.
Accessed April 4, 2023.
Smith, Ethan. “MTV Overtakes Vevo as Top Online Music Destination.” Wall Street
Journal. September 8, 2010, https://www.wsj.com/articles/BL-SEB-46396. Accessed
April 4, 2023.
Smith, Michael, and Rahul Telang. “Assessing the Academic Literature Regarding the
Impact of Media Piracy on Sales.” Carnegie Mellon University, August 19, 2012, https://
papers.ssrn.com/sol3/papers.cfm?abstract_id=2132153. Accessed April 4, 2023.
Sony Music. “Sony BMG Music Entertainment Signs Content License Agreement with
YouTube.” Sony Music press release, October 8, 2006. https://www.sonymusic.com/
sonymusic/sony-bmg-music-entertainment-signs-content-license-agreement-with-
youtube/. Accessed May 25, 2022.
Bibliography 473
Sorkin, Andrew Ross, and Jeff Leeds. “Music Companies Grab a Share of the YouTube
Sale.” New York Times, October 19, 2006, https://www.nytimes.com/2006/10/19/tec
hnology/music-companies-grab-a-share-of-the-youtube-sale.html. Accessed April
4, 2023.
“Sorry (Justin Bieber Song).” Wikipedia. https://en.wikipedia.org/wiki/Sorry_(Jus
tin_Bieber_song). Accessed June 5, 2022.
Soundcloud. “Mastering on SoundCloud Powered by Dolby.” https://community.soundcl
oud.com/mastering-on-soundcloud. Accessed June 2, 2022.
Spangler, T. S. “Morpheus, Other File-Sharing Services Sued by RIAA.” ExtremeTech,
October 3, 2001. https://www.extremetech.com/extreme/58270-morpheus-other-file
sharing-services-sued-by-riaa. Accessed June 17, 2022.
Spencer, Samuel. “How Many Subscribers Do Netflix, Disney+and the Rest of the
Streaming Services Have?” Newsweek, May 11, 2021. https://www.newsweek.com/
netfl ix-amazon-hulu-disney-most-subscribers-streaming-service-1590463. Accessed
June 2, 2022.
Spotify. “About Spotify.” https://newsroom.spotify.com/company-info/. Accessed May
25, 2022.
Spotify. “F-1 Registration.” Spotify. March 31, 2018, https://www.sec.gov/Archives/edgar/
data/1639920/000119312518063434/d494294df1.htm#rom494294_7.
Spotify. “Loud & Clear.” https://loudandclear.byspotify.com/. Accessed June 13, 2022.
Spotify. “Spotify Flagship Playlist ¡Viva Latino! Hits 10 Million Followers.” Spotify: For
the Record, July 26, 2019. https://newsroom.spotify.com/2019-07-26/spotify-flagship-
playlist-viva-latino-hits-10-million-followers/. Accessed June 5, 2022.
Spotify. “Where Is Spotify Available.” https://support.spotify.com/us/article/where-spot
ify-is-available/. Accessed June 5, 2022.
“‘Start Me Up’: the $3 Million Anthem That Launched Microsoft’s Windows 95.” 4As.
https://www.aaaa.org/timeline-event/start-3-million-anthem-launched-microsofts-
windows-95/?cn-reloaded=1. Accessed May 4, 2022.
Stassen, Murray. “BTS Set Spotify Record with 20.9M Day One Plays, but 10M Nearly of
Them Are Discounted on Platform’s Chart.” Music Business Worldwide, May 24, 2021.
https://www.musicbusinessworldwide.com/bts-set-new-spotify-streaming-record-
with-20-9m-day-one-plays-but-nearly-10m-of-them-are-discounted-on-platforms-
chart/. Accessed June 13, 2022.
Stassen, Murray. “JioSaavn’s Paying Subcriber Base Has Quadrupled Since Early 2019.”
Music Business Worldwide, May 28, 2020. https://www.musicbusinessworldwide.com/
jiosaavns-paying-subscriber-base-has-quadrupled-since-early-2019/. Accessed June
5, 2022.
Stassen, Murray. “Warner Is Signing Double the Number of Artists via AI Driven Tool
Sodatone.” Music Business Worldwide, November 24, 2020. https://www.musicbusine
ssworldwide.com/warner-is-signing-double-the-number-of-artists-via-ai-driven-ar-
tool-sodatone-than-it-did-last-year-now-its-hired-a-global-head-of-data-science/.
Accessed June 7, 2022.
Stassen, Murray. “Warner Joins $14M Funding Round in AI Music Start-up Lifescore.”
Music Business Worldwide, March 7, 2022. https://www.musicbusinessworldwide.com/
warner-joins-11m-funding-round-in-ai-music-startup-lifescore-founded-by-siri-co-
inventor-tom-gruber1/. Accessed June 2, 2022.
Stassen, Murray. “Apple Music Just Raised Its Subscription Price to $10.99 in the U.S. Will
Spotify be Next?” Music Business Worldwide, October 24, 2022. https://www.musicbu
474 Bibliography
sinessworldwide.com/apple-music-just-raised-its-subscription-price-to-10-99-in-
the-us-will-spotify-be-next1/. Accessed October 25, 2022.
Stassen, Murray. “Over 1,000 Songs with Human Mimicking AI Vocals Have Been
released by Tencent Music in China. One of Them Has 100M Streams.” Digital Music
News, November 15, 2022. https://www.musicbusinessworldwide.com/over-1000-
songs-human-mimicking-ai-vocals-have-been-released-by-tencent-music-in-china-
one-of-them-has-over-100m-streams/. Accessed November 16, 2022.
Stassen, Murray. “Still Convinced that TikTok Isn’t Turning into a Record Company?”
Music Business Worldwide, October 3, 2022. https://www.musicbusinessworldwide.
com/still-convinced-that-tiktok-and-bytedance-arent/. Accessed October 5, 2022.
Statista. “Smart Speaker Use Case Frequency.” https://www.statista.com/statistics/994
696/united-states-smart-speaker-use-case-frequency/. Accessed June 2, 2022.
“The Steady Reach of Radio: Winning Consumer Attention.” Nielsen, June 2019. https://
www.nielsen.com/us/en/insights/article/2019/the-steady-reach-of-radio-winning-
consumers-attention/. Accessed May 25, 2022.
Steffen, David J. From Edison to Marconi: The First Thirty Years of Recorded Music.
Jefferson, NC: McFarland, 2005.
Stereo Review Stereo Buyers Guide, 1975–95. Stereo Review. https://worldradiohistory.
com/Archive-All-Audio/HiFI-Stereo-Review.htm. Accessed April 29, 2022.
Stereo Review Tape Recorder Annual, 1968–79. Stereo Review. https://worldradiohistory.
com/Archive-All-Audio/HiFI-Stereo-Review.htm. Accessed April 29, 2022.
Sterling, Christopher H., and John Michael Kittross. Stay Tuned: A History of American
Broadcasting. Milton Park: Routledge, 2001.
Sterne, Jonathan. MP3: The Meaning of a Format. Durham, NC: Duke University
Press, 2012.
Stichting de Thuiskopie and World Intellectual Property Organization. International
Survey on Private Copying, WIPO Publication No. 1037E/16, 2015. https://www.wipo.
int/edocs/pubdocs/en/wipo_pub_1037_2016.pdf. Accessed May 27, 2022.
Strong, Maria (Acting Register of Copyrights). Section 512 of Title 17. Washington,
DC: United States Copyright Office, May 2020.
Stross, Randall. The Wizard of Menlo Park. New York: Broadway Books, 2007.
Stutz, Colin. “Spotify CEO Daniel Ek Says Direct Licensing With Artists ‘Doesn’t Make
Us a Label.’” Billboard, July 26, 2018. https://www.billboard.com/music/music-news/
spotify-daniel-ek-direct-licensing-artists-not-label-earnings-call-8467283/. Accessed
June 13, 2022.
Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the
Committee on the Judiciary, House of Representatives, 95th Congress. Performance
Rights in Sound Recordings. Washington, DC: U.S. Government Printing Office, 1978.
Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the
Committee on the Judiciary, House of Representatives, 95th Congress. Copyright
Law Revision: Hearings on H.R. 2223. Washington, DC: U.S. Government Printing
Office, 1976.
Sudario, Erielle. “Spotify Changes Playlist Rules Post Pocket Gods’ Protest Album.” We
Got This Covered, March 3, 2022. https://wegotthiscovered.com/music/spotify-chan
ges-playlist-rules-post-pocket-gods-protest-album/. Accessed June 7, 2022.
Sugar, Alan. What You See Is What You Get: My Autobiography. London: Macmillan, 2010.
Suisman, David. Selling Sounds: The Commercial Revolution in American Music.
Cambridge, MA: Harvard University Press, 2012.
Bibliography 475
Sullivan, Jennifer. “More Popular Than Sex.” Wired, October 14, 1999, https://www.wired.
com/1999/10/more-popular-than-sex/. Accessed April 5, 2023.
Sullivan, Jennifer. “The Sound of Busic (sic).” Wired, February 26,1999. https://www.
wired.com/1999/02/the-sound-of-busic/. Accessed May 23, 2022.
Sutherland, Mark. “Amazon Music Director Ryan Redington on Country’s Streaming
Revolution.” Music Week, March 12, 2020. https://www.musicweek.com/digital/read/
amazon-music-director-r yan-redington-on-country-s-streaming-revolution/079179.
Accessed June 2, 2022.
Swisher, Kara. “Circuit City Plugs into Visions of an Electronics Dream World.”
Washington Post, February 1, 1993. https://www.washingtonpost.com/archive/busin
ess/1993/02/01/circuit-city-plugs-in-to-visions-of-an-electronics-dream-world/
2f67d9fd-c21c-4931-99d3-1c3a13c8db43/. Accessed May 23, 2022.
Tarantino, Bob. “The Ninth Circuit’s Eminem License vs Sale Decision.” Entertainment &
Media Law Signal, September 9, 2010. http://www.entertainmentmedialawsignal.com/
the-ninth-circuits-eminem-license-vs-sale-decision/. Accessed June 7, 2022.
Tauberg, Michael. “Music Is Getting Shorter.” Medium, April 27, 2018. https://michael
taub erg.medium.com/music-and-our-attention-spans-are-getting-shor ter-8be37
b5c2d67. Accessed June 7, 2022.
Tauberg, Michael. “Spotify Is Killing Song Titles.” Medium, March 23, 2018. https://michael
tauberg.medium.com/spotify-is-killing-song-titles-5f48b7827653. Accessed June 2, 2022.
Taylor, Iain. “Audio Cassettes: Despite Being ‘A Bit Rubbish’, Sales Have Doubled during
the Pandemic—Here’s Why.” The Conversation, March 19, 2021. https://theconversat
ion.com/audio-cassettes-despite-being-a-bit-rubbish-sales-have-doubled-during-
the-pandemic-heres-why-157097. Accessed April 29, 2022.
Teather, David. “Napster Wins New Friend.” Guardian, November 1, 2000. https://www.theg
uardian.com/technology/2000/nov/01/internetnews.business. Accessed June 7, 2022.
“Thanks a Billion!” TikTok, September 27, 2021. https://newsroom.tiktok.com/en-us/1-
billion-people-on-tiktok. Accessed May 25, 2022.
“The Inevitable Choice: The Victrola.” 1919. David Sarnoff Library digital archive,
(Accession AVD.2464.001), Hagley Museum & Library, Wilmington, DE 19807.
Thill, Scott. “June 21, 1948: Columbia’s Microgroove LP Makes Albums Sound Good.”
Wired, June 21, 2010. https://www.wired.com/2010/06/0621first-lp-released/.
Accessed May 22, 2022.
Thompson, Clive. “How the Phonograph Changed Music Forever.” Smithsonian, January
2016. https://www.smithsonianmag.com/arts-culture/phonograph-changed-music-
forever-180957677/. Accessed April 4, 2023.
“TikTok Is Hitting Revenue Milestones at Lightning Speed.” Slow Reveal Graphs,
November 13, 2022. https://slowrevealgraphs.com/2022/11/13/tiktok-is-hitting-reve
nue-milestones-at-lightning-speed/. Accessed April 3, 2023.
“TikTok Users Worldwide (2020– 2025).” Insider Intelligence, June 1, 2022. https://
www.insiderintelligence.com/charts/global-tiktok-user-stats/. Accessed November
16, 2022.
“Today’s Top Hits Is the World’s Destination for the Very Best in Music.” Spotify: For the
Record, June 24, 2021. https://newsroom.spotify.com/2021-06-24/todays-top-hits-is-
the-worlds-destination-for-the-very-best-in-music/. Accessed June 13, 2022.
Toonkel, Jessica, Anne Steele, and Salvador Rodriguez. “TikTok Parent ByteDance
PlansMusic-Streaming Expansion.” Wall Street Journal, October 12, 2022, https://www.
476 Bibliography
wsj.com/articles/tiktok-parent-bytedance-plans-music-streaming-expansion-1166
5602937. Accessed November 16, 2022.
Tracy, Phillip. “Apple’s Newest Acquisition Could Mean Changes for Apple Music.”
Gizmodo, February 7, 2022. https://gizmodo.com/apple-ai-music-acquisition-184
8496216. Accessed June 2, 2022.
Tynan, Dan. “The 25 Worst Tech Products of All-Time.” PC World, May 26, 2006. https://
www.pcworld.com/article/535838/worst_products_ever.html. Accessed June 7, 2022.
Uy, Melanie. “3G vs. 4G Technology.” Lifewire, June 10, 2020. https://www.lifewire.com/
how-fast-are-4g-and-3g-internet-speeds-3974470. Accessed June 5, 2022.
Van Etten, David. “American Jukebox History.” https://www.jukeboxhistory.info/.
Accessed May 25, 2022.
“VE-10- 50 X.” The Victor Victrola Page. http://www.victor-victrola.com/index.html.
Accessed May 9, 2022.
Verna, Paul. “ITA at 25.” Billboard, March 11, 1995, 73.
Vevo. “VEVO Was Most Trafficked U.S. Entertainment-Music Web Network in December
2009.” Vevo press release, January 13, 2010. https://www.prnewswire.com/news-relea
ses/vevo-was-most-trafficked-us-entertainment-music-web-network-in-december-
2009-81347087.html. Accessed May 25, 2022.
Vevo. “The World’s Leading Music Video Network.” https://www.hq.vevo.com/. Accessed
July 25, 2022.
Vezina, Briggitte and Diane Peters, “Why We’re Advocating for a Cautious Approach to
Copyright and Artificial Intelligence,” Creative Commons, February 20, 2020, https://
creativecommons.org/2020/02/20/cautious-approach-to-copyright-and-artificial-
intelligence/ Accessed June 2, 2022.
“Victor, The World’s Greatest Musical Instrument.” 1903. David Sarnoff Library digital ar-
chive (Accession AVD.2464.001). Wilmington, DE: Hagley Museum & Library, 19807.
Vietze, Dmitri, and Drew Silverstein. “What Makes Amper’s AI Music Tools Different
with CEO Drew Silverstein.” Music Tectonics (podcast), March 25, 2020. https://mus
ictectonics.libsyn.com/what-makes-ampers-ai-music-tools-different-with-ceo-drew-
silverstein. Accessed June 2, 2022.
“Vinyl Music Gives Record Stores a Boost in a Digital World.” ABC News, April 19,
2017. https://web.archive.org/web/20180207210919/https://abcnews.go.com/amp/
Entertainment/wireStory/vinyl-music-record-stores-boost-digital-world-46887652.
Accessed April 29, 2022.
Vinylmint. “History of the Record Industry 1877–1920s.” Medium, June 7, 2014. https://
medium.com/@Vinylmint/history-of-the-record-industry-1877-1920s-48deacb4c
4c3. Accessed May 9, 2022.
Vinylmint. “History of the Record Industry, 1920–1950s.” Medium, June 8, 2014. https://
medium.com/@Vinylmint/history-of-the-record-industry-1920-1950s-6d491d7cb
606. Accessed May 9, 2022.
Waniata, Ryan. “The Life and Times of the Late, Great CD.” Digital Trends, February 7,
2018. https://www.digitaltrends.com/music/the-history-of-the-cds-rise-and-fall/.
Accessed May 23, 2022.
Ward, Ed. The History of Rock & Roll, Volume II, 1964–1977. New York: Flatiron
Books, 2019.
Warren, Lydia. “Steve Jobs Listened to Vinyl at Home . . . Because It Sounded Better Than
His iPod.” Daily Mail, January 31, 2012. https://www.dailymail.co.uk/news/article-
2094590/Steve-Jobs-listened-vinyl-home-iPod.html. Accessed June 17, 2022.
Bibliography 477
Water & Music. “Music/ Web3 Dashboard.” https://www.waterandmusic.com/data/
music-web3-dashboard/. Accessed May 28, 2022.
Water & Music. “Music NFT Sales in 2021: What We Learned,” Water & Music, May 6,
2022. https://www.waterandmusic.com/music-nft-sales-in-2021-what-we-learned/.
Accessed November 16, 2022.
“Web Music Sharers Under Attack.” CBS News, July 14, 2003. https://www.cbsnews.com/
news/web-music-sharers-under-attack/. Accessed June 7, 2022.
Wesch, Michael. “YouTube Statistics.” Digital Ethnography @ Kansas State University.
https://web.archive.org/web/20140723075938/http://mediatedcultures.net/thoughts/
youtube-statistics/. Accessed May 25, 2022.
Whately, Dan. “How TikTok Is Changing the Music Industry.” Business Insider, July 7,
2022. https://www.businessinsider.com/how-tiktok-is-changing-the-music-industry-
marketing-discovery-2021-7. Accessed July 14, 2022.
“Where Do Music Collections Come From?” American Assembly, October 15, 2012.
http:// pir a cy.ameri c ana s sem bly.org/ w here- d o- music- c oll e cti ons- c ome- f rom/ .
Accessed June 7, 2022.
Wiki Boombox. Analog Alley. https://www.wikiboombox.com/tiki-index.php. Accessed
May 3, 2022.
Wilentz, Sean. 360 Sound: The Columbia Records Story. San Francisco: Chronicle
Books, 2012.
Willman, Chris. “Adele’s ‘30’ Has the Biggest Bow of 2021 With 839,000 Album Units.”
Variety, November 28, 2021. https://variety.com/2021/music/news/adele-30-biggest-
number-one-album-year-1235120896/. Accessed April 4, 2023.
Witt, Stephen. How Music Got Free: A Story of Obsession and Invention. London: Penguin
Books, 2016.
Wu, Adrian. “Open Reel Tape—The Ultimate Analog Source?” Copper. https://www.psau
dio.com/copper/article/open-reel-tape-the-ultimate-analog-source/. Accessed April
29, 2022.
Young, Neil. “The CD and the Damage Done.” Harper’s, July 1992, 23–24.
Zaleski, Annie. “35 Years Ago: The U.K. Launches the ‘Home Taping Is Killing Music’
Campaign.” Diffuser.fm, October 25, 2016. https://diffuser.fm/home-taping-is-killing-
music-uk/. Accessed April 29, 2022.
Zuckerman, Arthur J. “Copy-Cat Tape Decks.” Popular Mechanics, October 1, 1983, 94+.
Index
For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on
occasion, appear on only one of those pages.
Figures are indicated by f following the page number
6C Framework, 10, 10f, 13, 14, 15, 353–82 All Directions (Temptations), 91
defined, 10 Amazon Echo, 9, 331–32, 339
8-track tape format, 2, 5–6, 145, 355 Amazon.com, 2
cars and, 5–6, 133f, 134, 149 digital music distribution and, 242–43,
limitations of, 134–35 246, 266, 268, 270, 271, 339–40, 377
revenues from, 153, 153f, 154 Amberol cylinder, 26–28
45-rpm record format, 2, 4–5, 97, 99–100, Amberola phonograph, 26–28
100f, 114, 354–55, 362 America Online (AOL), 7–8, 62–63
jukeboxes and, 109–11, 110f American Bandstand, 159
78-rpm record format, 4, 32, 38, 97, 98, 99, American Federation of Musicians
362, 376 (AFM), 59–60, 77–78, 123
origin of, 30 American Mobile Satellite Corporation, 67
1000 x 30: Nobody Makes Money Anymore American Music Fairness Act, 82
(Pocket of God), 280–81 American Record Manufacturers
and Distributors Association
a2b music, 218, 219, 221, 222f, 236, 237, (ARMADA), 125
256–58, 355–56 American Society of Composers, Authors,
Abbey Road (Beatles), 206 and Publishers (ASCAP), 12, 53–54,
Abrams, Lee, 70, 71 59–60, 78, 84–85, 88, 123
ABS Entertainment, 80 Amper Music, 338–39, 346–47
acoustic fingerprinting, 87–88, 232–33, Ampex, 130, 137
234–35, 297–99, 303–4, 334, 348 Amstrad TS55 dubbing cassette deck,
Audible Magic, 298, 299, 302, 303 141–42, 143
Content ID, 303, 304, 334 Anderson, Chris, 248
YouTube and, 298–99, 302, 303 Android Auto, 93–94
acoustic phonograph “Annie Get Your Gun,” 4
defined, 28 AOL Radio, 66
electrical phonograph compared to, Apodaca, Nathan, 318
30–31 Apollo Company, 52–53
reliability of recording with, 32–35 Apple CarPlay, 93–94
Adele, 154, 278 Apple Inc., 1, 2, 8, 66, 239–41, 377
Advent cassette decks, 138–39 Apple Music, 1, 95, 269, 271–72, 304, 310,
Advent Corporation, 138 311, 340
Aeolian Organ and Music Company, 22, origins of, 275–76
52–53, 54, 55–56, 57 “per play” royalty rate of, 264, 314
Afrika Bambaataa, 148 Apple Music TV, 311
“Alexander’s Ragtime Band,” 37–38 Apple Tree, The (stage play), 117–18
480 Index
copyright, 1, 2, 12–14, 51–57, 120, 366–68 Digital Millennium Copyright Act (DMCA)
artificial intelligence (AI) and, 345–49 (1998), 12–13, 305, 313, 367–68
cable television and, 125–26 origins of, 231
compulsory license and, 56 “safe harbor” and, 231–32
“cover” versions and, 56–57 streaming and, 284, 285, 301–2
first sale doctrine and, 208–9 Digital Performance Right in Sound
gaps in, 12–13, 77–83 Recordings Act (DPRA), 79–80
Internet video and, 301–7 Digital Rights Management (DRM), 116,
mechanical reproduction rights and, 218, 237–38, 240–41, 246, 252, 259,
53–57, 287–91 261–62, 355–56
music video and, 182–83 digital sampling, 3, 11
nonfungible tokens (NFTs) and, 392–93 copyright concerns and, 148–49
performance rights and, 77–79, 81, 82 Dingell, John, 86
performance rights organizations Directive on Copyright in the Digital
(PROs) and, 78–79, 87–89 Single Market (2019), 305
player piano rolls and, 51–53 disco, 75
podcasts and, 82–83 Discovery Music Network (DMN), 180,
radio and, 77–83 184–85
sound recordings and, 122–25 distributed ledger. See blockchain
streaming licenses and, 287–91 DJ Jazzy Jeff & The Fresh Prince, 206–7
synch licensing and, 182–83, 306–7 DMX, 3
Copyright Act of 1909, 122, 287–88 Do the Right Thing, 151
Copyright Act of 1976, 125–26, 208 Dolby, Ray, 137
Copyright Royalty Tribunal, 143 Dolby, Thomas, 173
Cornelius, Don, 161 Doll & Toy Phonograph Company, 21
Cousino, Bernard, 132 Don Kirshner’s Rock Concert, 161
Crosby, Bing, 4, 38, 44–45, 130 Donnelley, Bob, 86–87
Currier, Frank, 56 downloadable music, 1, 215–54, 359, 377
“cutouts,” 107 artist reaction to, 248–50
copyright battles and, 227–35, 250
Da Capo (Love), 118–19 customer reaction to, 252–54
DailyMotion, 299, 315 digital distribution and, 236–43, 248
Dalhart, Vernon, 50 “long tail” theory and, 248
Davis, Clive, 85 marketing of, 245–46
Davis, Miles, 118 metadata and, 219–21
Dean, Roger, 120 MP3 and, 216–18, 219–21
Decca Records, 44–45 Napster and, 223–26, 228–33, 244,
Deezer, 1, 265, 271–72, 314 248–49
Def Leppard, 89, 91 origins of, 215–17
“Despacito,” 9, 283, 321–22 peer-to-peer file sharing and, 8, 223–24
Devo (group), 172–73 revenues and, 5f, 243–45, 246–47, 248,
Dewey, John, 41 257f
Diamond, Dave, 90 software security and, 218
Diamond Disc record player, 28, 40 sound quality and, 250–51, 252
Diamond Rio PMP-300 MP3 player, 221– downloadable video, 379–80
23, 227–28 Hulu and, 380–81
Dictaphone, 20, 21 MovieLink and, 379
Digital Audio Tape (DAT), 211–12, 227 Ultra Violet and, 379–80
Index 483
Public Enemy (group), 148 station ownership caps and, 72, 74,
Pulse Code Modulation (PCM), 193 81–82
Pure Digital Technologies, 295 streaming audio and, 62–67
Putnam, Herbert, 53 Superstars FM, 70, 71
Top 40, 68, 70, 71, 89, 106
quadraphonic (quad) sound, 104, 105f, 132 transistor, 4–5, 6, 60–61, 68, 89, 114
Radio & Records (R&R), 70, 72–73, 73f,
R. Stevie Moore Cassette Club, 146–47 74–76, 373
“race” records, 49–50, 75, 372–73 Radio Music Licensing Committee
“rack jobbers,” 105–7 (RMLC), 87, 126
radio, 2, 3–4, 58, 59–96, 366–67 Radio y Musica, 75
advertising and, 4, 76, 95–96 ragtime, 37–38, 48–49
airplay “burn factor” and, 71 Real Guide, 63–64
Album-Oriented Rock (AOR), 70, RealAudio, 62–63
71–72 RealNetworks, 8, 62–64, 238
AM broadcasts and, 61, 69 RealOne, 238
artificial intelligence (AI) and, 341 Record Rental Amendment of 1984, 208–9
automobiles and, 61, 62f, 67, 92, 93–94, Recording Industry Association of
94f America (RIAA), 5, 5f, 81, 82, 86, 88,
Black, 75 101–2, 111, 121, 125, 143–44, 153,
college, 73–74 154, 210, 227–28, 233–35
copyright and, 77–83, 366–67 CD copying and, 212–13
country music, 68 SoundExchange and, 88
FM broadcasts and, 61, 69 records. See also 45-rpm record format;
FM rock, 69–70, 71, 72 78-rpm record format; long-playing
impact on record sales and, 44–45, 68, (LP) record format
70 origin of album and, 32, 362
in-band-on-channel (IBOC) production cost of, 42–43
transmission and, 67 royalties and, 43–44
industry growth of, 29 sales revenues and, 44–45
Internet databases and, 63–64 Red Hot Chili Peppers, 89–90
metadata and, 64–66 reel-to-reel tape, 5–6, 355
music formats and, 68, 69, 373 KLH Model Forty deck and, 138
New Rock, 74 Regency TR-1, 60, 68
online brands and, 66 R. H. Macy & Company v. Victor Talking
origins of recorded music and, 59–60 Machine Company (1917), 58
“payola” and, 83–87, 106 “Respect,” 91
performance rights and, 12–13, 77–79, Resso, 305
81, 82 “Return of the Son of Monster Magnet,
phase-locked loop (PLL) tuning and, 62 The,” 118–19
prerelease records and, 71–72 “Revelation,” 118–19
Progressive FM, 69–70 Rhapsody, 258–60, 288, 298, 311, 377
reach of, 4, 93f, 93 Rhodes, Nick, 205–6, 251
royalties and, 12–13, 78–79, 81, 83, RIAA equalization, 101–2
87–89, 367 RightsFlow, 288
satellite, 67–68, 76–77, 79–80 Rio MP3 player, 221
song length and, 90–91 Robertson, Michael, 254
Spanish-language, 75–76 Rockola, David, 109
Index 489