Mobile TV Should Be More Than A Television: The Final Report of Podracing Project
Mobile TV Should Be More Than A Television: The Final Report of Podracing Project
Mobile TV should be more than a television. The final report of Podracing project
The primary study question was formulated as following: If a user had
the possibility to watch the latest television news from the mobile phone,
or listen to the news on the radio, or read text news with the mobile,
what would he or she choose?
Abstract
This publication contains the results of the project “Intuitive and parallel media service
platform to 3G, podcasting, and DVB-H” (Podracing). The aim of the project was to
compare three different media formats (text, audio and video), and three delivery
methods (broadcast, unicast and pre-download) from end user’s perspective.
The primary study question was formulated as following: If a user had the possibility to
watch the latest television news from the mobile phone, or listen to the news on the
radio, or read text news with the mobile, what would he or she choose?
As a result, mobile terminals were used in such a variety of situations that no single
media format and no delivery method was able to fulfil all needs.
Previous studies suggested that mobile television involves low commitment. This was
contrary to the findings of Podracing: The users did not usually do anything else while
watching. Furthermore, just surfing around or viewing an arbitrary program flow seems
not be enough. Even though one of key motives of mobile television use is to kill time,
people may want to kill boredom with special content, which is just suitable for that
moment.
3
Mobile TV should be more than a television. The final report of Podracing project [Mobiili-tv:n tulisi olla
muutakin kuin pelkkä pieni televisio]. Toim. Ville Ollikainen. Espoo 2008. VTT Tiedotteita – Research
Notes 2439. 71 s. + liitt. 4 s.
Avainsanat mobile TV, podcasting, DVB-H, media usability, business models
Tiivistelmä
Jos kuluttajalle annetaan mahdollisuus valita, katsooko hän kännykästään viimeisimmät
televisiouutiset, kuunteleeko uusimmat radiouutiset vai lukeeko tuoreimmat tekstimuotoiset
uutiset, minkä hän valitsee? Entä viihdesisällöstä? Tutkimus antoi vastauksia näihin
kysymyksiin kenttäkokeen avulla. Kenttäkokeessa tutkittiin rinnakkain sekä eri media-
muotoja että jakelutapoja, jotka olivat tasapuolisesti koekäyttäjien saatavilla. Myös
erilaisiin sisältöihin liittyvät lisäarvopalvelut kuuluivat tutkimuksen aihepiiriin.
Tuloksena voitiin todeta, ettei mikään mediamuoto tai jakelutapa yksinään vastaa mobiili-
käyttäjän tarpeisiin. Mobiilitelevision pitää olla enemmän kuin pelkkä televisio.
4
Preface
This publication contains the results of the project “Intuitive and parallel media service
platform to 3G, podcasting, and DVB-H” (Podracing). The two year project has been
funded by Tekes – the Finnish Funding Agency for Technology and Innovation, VTT
Technical Research Centre of Finland and several companies. The aim of the project
was to compare three different media formats (text, audio and video), and three delivery
methods (broadcast, unicast and pre-download) from end user’s perspective.
The project had previously conducted two field trials, which have been reported in two
publications. The first publication compiled the main findings of the first user trial
which concentrated on news content in different media formats. The second publication
compared different mobile TV technologies: DVB-H and 3G. In the third and last field
trial the three media formats and three delivery methods were available in one
integrated service.
In addition to Tekes and VTT, companies financing the project were represented in the
project management group as well as the research partners. At the end of the project the
group comprised chairman Vesa Erkkilä (Digita), Jonas Kronlund (Elisa), Juhani Reiman
(Lingsoft), Teemu Lehtonen (MTV3), Jani Hätönen (Radio Nova), Ari Pöyhtäri (Sofia
digital), Eskoensio Pipatti (SWelcom), Jouni Siren (YLE), Marko Heikkinen (Tekes) and
Carlos Herrero (TKK), Caj Södergård (VTT), Esa Reunanen (TAY), and Heli Paavola
(TAY). The group has made a great effort on behalf of the project.
The project was carried out by VTT Technical Research Centre (VTT), Helsinki
University of Technology (TKK) and the University of Tampere (TAY). The project
group consisted [contributions to chapters of this publication in brackets] of Ville
Ollikainen (VTT) [editor, 1, 4.2, 6], Elina Noppari [4.3, 4.4, 4.5, 4.6, 6], Maarit Mäkinen,
Esa Reunanen [6] and Jorma Riihikoski (TAY) [2.2, 5, 6], Carlos Herrero [1] and Pia
Ojanen (TKK) [3.1, 3.3], Tero Hannula, Tuomo Kivinen [2.1], Juha-Pekka Koivisto,
Timo Kinnunen [3.2], Virpi Oksman [2.1, 4.1, 4.3], Chengyuan Peng, Antti Tammela [1]
and Markus Ylikerälä (VTT).
Different mobile television delivery methods are a new and very fast developing
technology. Hopefully, this publication will give the readers some new information and
guidance in this exciting field.
Ville Ollikainen
Senior Research Scientist
VTT Technical Research Centre of Finland
5
Contents
Abstract..............................................................................................................................3
Tiivistelmä .........................................................................................................................4
Preface ...............................................................................................................................5
1. Introduction..................................................................................................................9
2. Background for the research: The use of mobile media services .............................12
2.1 Previous Podracing trials...................................................................................12
2.1.1 First Podracing trial: Text vs. audio vs. video in news content .............12
2.1.2 Second Podracing trial: DVB-H vs. 3G streaming ................................13
2.2 Background of mobile TV business ..................................................................15
6
3.3.1.11 Prices...........................................................................................33
6. Conclusions................................................................................................................66
Acknowledgements .........................................................................................................68
References .......................................................................................................................69
Appendices
7
List of acronyms and symbols
3G The third generation of mobile phone standards and technology
IP Internet Protocol
SW Software
UI User Interface
8
1. Introduction
Ville Ollikainen, Antti Tammela, Carlos Herrero
The project studied these three delivery methods. Let’s now have an overview on the
topic.
Watching television on a small screen has been possible for decades, as already in
1960’s portable televisions became available for consumers. Until these days, however,
the delivery technology as well as content and programme schedules have been targeted
to normal television, not at all for mobile use. More recent mobile TV technologies have
potential to enable true mobile television services.
For broadcasters, mobile TV offers new audiences and new prime times, while mobile
operators aim for business opportunities and reducing churn. As there is also business
potential for terminal manufacturers, technology providers, content producers and
advertisement agencies, just to name few, only one major question seems to remain:
how will the consumer accept the idea of mobile television?
From technical point of view, mobile TV networks are considerably different from
traditional TV. There are currently two approaches in delivering mobile TV: The first is
via a 3G cellular network. The second is through a broadcast network – in Europe DVB-
H, T-DMB or DAB-IP. Existing 3G infrastructures provide the easiest launch for
delivering content to mobiles: For 3G consumers there is no need to get new handsets,
since in practise all 3G phones are video enabled. However, when the number of users
9
increases, 3G streaming fails to scale up to a mass market of mobile TV services.
Within a couple of years, 3G networks are expected to get a broadcast upgrade, MBMS
(Mobile Broadcast for Multimedia Services) [Hartung et al. 2007]. In the future, IP-
based wireless technologies like WiMax may play a substantial role as a competitor in
mobile streaming.
Downloading content to terminal memory has also been enabled by rapid development
in non-volatile memory technologies. Increased production volume, mainly caused by
digital photography, has slashed the price of one gigabyte to a fraction within just a
couple of years. A reasonably priced 2GB memory chip can store 15 hours of 300 Kbps
video. Consequently, music players, such as the above-mentioned iPod, have turned into
general media players, premium mobile phones such as the Nokia N93 have quality
video record and playback capabilities, and portable game consoles such as Playstation
Portable (PSP) can store hours of mobile quality video.
We should keep in mind that mobile TV handset is capable of reproducing not only
video, but also audio. Larger screens make even text and images more applicable.
10
The study aims of the whole Podracing project emerged from the integration of different
media formats and delivery technologies (see Figure 1). Our primary study question has
been formulated as following:
If a user had the possibility to watch the latest television news from the mobile phone,
or listen to the news on the radio, or read text news with the mobile, what would he or
she choose?
Video on-demand
Audio download
11
2. Background for the research:
The use of mobile media services
2.1 Previous Podracing trials
Previously two field trials were conducted in Podracing project. These trials have been
reported separately, but a short summary is provided below.
2.1.1 First Podracing trial: Text vs. audio vs. video in news content
The first trial had a focus on different media format: video, audio and text. News
content in these three formats was provided in an equal manner in a unified user
interface.
The first field trial started in March 2006 with 10 users who used the service with 3G
phones (Nokia 6630 and N70) for one month with an Elisa subscription. Before the test
period, the users were interviewed and they received information concerning the test.
Demographic data and media user profiles were gathered from the informants. The ages
of the users ranged from 23 to 56. All of them worked at least part time and had used
mobile services before. During the test period, the users reported their user experiences
in a test diary. After the test, they were asked to fill in a usability evaluation form and
they were interviewed again.
Mobility and real time effect were considered the most important characteristics in the
first Podracing service. Being combined, these properties challenge all other media
technologies. In the mobile news delivery, the users appreciated both continuously
updated information and media formats carrying large amount of information. Having
everything in real time was considered important, and as a consequence only the latest
news had a high demand. From usability point of view, ease of use and quick launch
and response times were also appreciated. Compared to earlier studies on mobile video
content [Repo et al. 2003] there has been some changes in technology platform:
Especially the use of earpieces with the mobile phone has became common since. This
makes it more convenient to consume media content in privacy without disturbing
others, for instance in public places.
When the users were able to receive news in different media formats in their mobile
phones, text format were used most frequently (see Table 1). Typically, users
considered text news format the most convenient for various kinds of situations. Text
12
was especially suitable for quick news headlines updates “on the go”. Text based news
format was also found less vulnerable for problems caused by disturbances in 3G
reception. However, regarding the total amount of time, news in video format got
longest viewing sessions. It seems that watching video news took place less frequently
than reading text news, but at once, when the reception was good, people were watching
on-demand news longer than having just at a glance.
Table 1. Text format was the most often used media type, but total usage time was
longest for video.
Text was considered the most reliable media form, sometimes associated with
newspapers. On the other hand, when time was limited, video was considered being able
to explain things in an understandable way. When watching mobile video, users were
focused on it: One or both hands were needed to hold the mobile device, and users
concentrated on watching the small screen and listening to the audio. Even at home
users did not leave the mobile video on for “background noise” – the sound levels were
considered very good, but users still held the phone in their hands and most often did
not do anything else while watching. This finding was quite contrary to the low
commitment expectation suggested by Knoche and McCarthy [2005].
The number of sessions when listening audio news was relatively low, but the average
duration of the sessions was of the same magnitude as in watching video. This was a bit
surprising, since the mobile phone in use (Nokia N70) had radio implemented in the
phone, and the users preferred listening to it in real time instead of selecting news
articles from the service. Monitoring radio usage and disabling radio from the phone
had both technical obstacles, but the use of radio was taken into account in interviews.
13
The field test started in October 2006 with ten families who used 3G TV and DVB-H with
Nokia N92 phone for a period of one month with Elisa subscription. Before the test period,
the users were interviewed and they received information concerning the test. Demographic
data and media user profiles were gathered from the informants. The ages of the informants
ranged from 12 to 56. During the test, the main informants carried the testing phone as
their primary mobile, using it for both professional and personal communication. The
testers used their company phone subscribtions for using the service. During the test period,
the informants reported their user experiences in a test diary. After the test, users were
asked to fill in a usability evaluation form and they were interviewed again. The
informants had different kinds of hobbies, lifestyles, media profiles and interests. The
tested mobile TV services consisted of a wide range of different kinds of contents: from
main TV channels to sports news channels and from fashion TV to user generated contents.
It appears that the quality of reception is still one of the major technical problems
perceived in mobile TV use.
The empirical research suggested that the users perceived a significant difference
between different delivery methods, e.g. 3G and DVB-H and pre-downloading, and the
difference is likely to affect on the length on watching sessions of mobile TV. During
the test period especially 3G networks had some reliability and coverage problems.
However, when technology gives the users more bandwidth and larger coverage areas,
also the 3G TV viewing experience is likely to become better. Podcasting to the mobile
phone was quite a new concept for the test users; however they thought that this
delivery method would be useful for instance on long journeys, and in general in places
with no 3G or DVB-H coverage. Obviously the advantage of podcasting is that
watching the podcasted contents does not depend on the network connections – and thus
can be done any time later, regardless of the location.
The findings of the test confirm many previous studies concerning the contexts of using
mobile TV. Mobile television is mostly used in public sphere. The test users picked
mainly the contents and channels which were already familiar. However since the test
period was quite short, we were not able to see if the situations would have changed
after a longer period of time. The interactive services were found interesting, and we
could expect more innovative services developing in that area. Mobile TV alone caused
a surprisingly high increase in estimated media consumption. It would be interesting to
know if there would be a decrease in the time spent for reading free papers, since some
contexts for consuming these media are similar (e.g. in commuting).
It was found that the price of using the mobile TV services, when they are finally
finished products on the market, is also a significant factor that will affect the use. In the
third trial of this project, reported in this document, the expectations on pricing models
of mobile TV services are studied in detail.
14
2.2 Background of mobile TV business
Jorma Riihikoski
15
• The default channel is “La 3 Live”, which contains low-cost programming (production
costs 200 €/hour).
• Only one Samsung and one LG handset are supported (because of early implementation,
the service is not fully compliant to DVB-H standards).
• Among pricing options for the TV service are e.g. daily charge of 3 € and monthly
charge of 29 €.
Compared to Italy, Korea case sets a contrasting example in Korea. Case Korea: both
networks are said to make big losses, the satellite-DMB service has 600,000 subscribers
while it would need 5.5 million to reach break-even [Dermot Nolan, TBS].
Several estimations of mobile TV growth have been made. One of them, made by
Informa Telecoms and Media, is presented in Table 2.
In the early years of the forecast period, the most advanced networks will be S-DMB
and T-DMB services, dominating broadcast TV handset sales worldwide from their
stronghold of South Korea, and also from the early, less successful launches in
Germany. However, their combined total is due to be overtaken by the Japanese ISDB-
T standard in 2007. By 2012, there will be 23.6 million T-DMB subscribers and 10.14
million S-DMB users worldwide, as the former extends into the Middle Eastern markets
and China (T-MMB). [Burk & McQueen 2006.]
There is also more optimistic estimation of faster growth of mobile users. According
Gartner amount of users will reach 500 millions in four years time. Although in this
amount are included all the people who are watching some kind of TV broadcast in their
mobile devices, only 30% of users will order the services and the rest will get them with
operator’s package deal. Mobile networks TV turnover will grow from 100 million to
16
15 billion US dollars in 2010. Broadcast (meaning DVB-H-, MediaFLO- or DMB)
turnover grows from 200 million to 10.8 billion USD.
Mobile TV appears to be expanding compared to traditional TV, but for the moment it
is a marginal phenomenon. International examples portray business opportunities
through expanding subscribers’ case, but also losses and failures.
17
3. The Podracing prototype for third trial
3.1 System description
Pia Ojanen
The general aim of the Podracing project was to compare different media formats (e.g.
text, audio and video), delivery methods, and mobile networks. The first trial
concentrated on news service. The first prototype used on-demand delivery and 3G
network. The second trial concentrated on comparing 3G and DVB-H, and gave some
user opinions on podcasting. Third trial, presented in this publication, had podcasting
integrated to the service. The prototype was evaluated in a field trial.
What a user actually sees is an xml-based user interface with links to video, audio and
text files. Content is arranged by category and content provider. Special browser has to
be installed to the phone in order to be able to access this service. Content is delivered
via IP based 3G and WLAN using different delivery methods. Browser includes a built
in tool for controlling podcasting specific downloads.
Application can be viewed by opening the browser application in the Mobile phone,
which fulfils some minimum requirements. Having either WLAN or 3G connection
available is essential, since the service cannot be viewed off-line.
18
User Interface consists of navigation and content areas (Figure 3). Graphical navigation
elements were included, as the minimum requirements for the application already
required graphical capabilities from the phone and adequate transmission speed. Thus
also download times should be tolerable.
Automated content provisioning process was created for producing suitable material for
mobile devices: After broadcasting a television program or publishing text based news,
these items are sent to or retrieved by the Content Server, which modifies them to
correct size and format. Content and available metadata is saved to the database. User
interface presents available media content, podcast subscription status.
Most handheld devices with network capabilities are Java-enabled and support some
version of the Mobile Information Device Profile (MIDP). Java applications, called
midlets, developed for a certain MIDP version, should be compatible with on all devices
supporting that profile. Even though MIDP only supports only a few video and audio
coding formats, it would be possible to develop a mobile video news service completely
in Java code [Lee & Jayant 2006]. However, due to the MIDP constraints of playing
audio and video files, it is recommended to use the native media player of the device.
Handheld devices share common characteristics, e.g., both memory and processor
capacity are very limited, and J2ME applications are widely supported. However,
different devices can provide different media frameworks and network capabilities.
Hence, for the project J2ME-based SW development is preferred, while run-time
memory should be kept minimal.
In the beginning, Podracing project had a plan to exclusively use Nokia N92 phone with
DVB-H capability. The phone supports MIDP 2.0 standard, WLAN and 3G technologies
can be used for data transfer. Display resolution is 240 x 320, which is adequate for
presenting graphical user interface and video content; and the device supports memory
19
cards of up to 2 GBytes capacity. Models with quite similar capabilities, but without
DVB-H, became available (Nokia N93 and Nokia N95) during the project. All these
models were supported (Figure 4). DVB-H properties became later on less important, as
it turned out that there were no programming interfaces available for integrating
broadcast transmissions to the Podracing service.
The System architecture is based on the client/server model (Figure 5). In the third field
test there were two servers running the service:
• Content Server
• Streaming Server.
Streaming
Client UI Scheduler FTP Media Server
Podcast Sync
Podracing DB
Server
RTP
HTTP
Mobile Client
20
3.1.4 Server software
The Content Server software consists of three components: Content Fetcher, user and
object databases and User Interface (UI) generator. The Content Fetcher receives and
retrieves media files automatically from different sources, e.g., Finnish TV broadcasters,
and adds the media files and corresponding metadata as Java objects into a Structured
Query Language (SQL) based database system, the object database. In addition to that, a
separate database, the user database, stores information about users’ preferences and
podcast subscriptions.
The UI generator was implemented using Java Server Pages (JSP) [http://www.jsp.com]
and Spring Framework [http://www.springframework.org] to create the user interface for a
declarative environment. The markup language interpreted by the browser was a reduced
version of XHTML Mobile Profile, created by Sofia Digital.
Content Server contains most of application logic in the system. It contains following
software components:
• Servlet Container
o Hosts web applications
o Technology: Apache Tomcat (http://tomcat.apache.org/)
• Scheduler
o Triggers as scheduled tasks on timely basis
• Podracing Database
o Provides persistence mechanism
o Technology choices: MySQL (http://www.mysql.com/)
• Client UI
o User Interface for browsing content at the server
o Technology choices: JSP, Spring MVC
• Podcast Synchronization Server
o Interface for retrieving podcasts from the server.
Client software consists of browser and synchronization application, which have been
combined to one J2ME MIDP application. The Browser is Sofia Backstage® Mobile
Digitext browser [http://www.sofiadigital.com]. The Podcast synchronization application,
named Podsync, is used for retrieving Podcast files from the object database, requesting
the list of new files to be downloaded from the user database. The Podsync application
21
can also delete files from the memory card and inform Content Server, through user
database, about the files that are stored on local memory card at any moment. Sofia
Backstage® Mobile Digitext Browser has been customized for Podracing UI, and
Podsync was integrated into it.
Sofia Digital Backstage browser can serve as an example as XML based Client, which
is not a normal web browser. It has quite similar features with Mobile web browsers and
couple of extra features provided only by this browser. One main advantage was
possibility to close co-operation with Sofia digital, in order to implement project
specific extra features to the browser.
A special limited version of XHTML Mobile Profile was used as a markup language for
second version of the browser. One of the major improvements in the second version is
the absence of transcoding, which reduces response times significantly. Fonts and
pictures are clearer, because the view is not being scaled from TV resolution to fit the
screen. Figure 6 illustrates some sample screen shots.
22
3.2 Podracing content
Timo Kinnunen
Three types of content, text, audio and video, from three broadcasting companies,
Nelonen, MTV and YLE, is fetched, processed and stored by Podracing servers. The
text material is basically news articles in tagged XML-files, which can be parsed with
Java Xerces XML-parser. The XML also includes some metadata used by the project. A
new fresh set of news articles is fetched once every hour. The audio material is a static
set of clips, fetched and transcoded into 3GPP audio-only clips. There has been no
update for audio material. The selected video material is automatically transferred every
day from broadcasters.
In video processing, Windows Media Format (MTV3 and Nelonen) is first resized with
AVISynth and then encoded into 3GPPv5 (MPEG-4 SP) format with Helix Mobile
Producer Standard. A speech-to-text analysis tool from Lingsoft is used to extract
keywords from MTV3 and Nelonen tv-news broadcasts. YLE provides video files in
3GPP format, so no encoding is needed in YLE content. Finally Darwin Streaming
Server is used to stream 3GPP files to Real Player clients in mobile devices. Figure 7
below illustrates content acquisition processes used in the project.
text
Series Episodes
video files (3gp)
speech
video files (3gp) to text
Content Server
video files
Tv-news broadcasts
- VIDEO files from Nelonen FTP (wmv) ENC video files (3gp) audio files
(wav)
(Tv-news and series)
23
3.2.2 On-demand content
On-demand content contains tv-news broadcasts from YLE, MTV3 and Nelonen,
Rimpauttaja audio clips from Radio Nova and text-news articles from YLE and MTV3.
Podcasted content contains series episodes from MTV3 (“Salatut elämät”) and Nelonen
(“Bella”, “Inno”, “Ota tai jätä”, “Pop-ikoni”, “Retro” and “Start!”).
Five 24/7 broadcast simulation channels are generated as five different Darwin Streaming
Server Playlists: Nelonen series, Nelonen news, MTV3 series, MTV3 news and YLE
news. Using predefined playlist filenames it was possible to update playlist dynamically
and automatically as new material arrived to our server. All news channels rotated latest
news broadcasts from one day. MTV3 series channel rotated six latest episodes of one
series (“Salatut elämät”) whereas Nelonen series channel rotated six different series
(“Bella”, “Inno”, “Ota tai jätä”, “Pop-ikoni”, “Retro” and “Start!”) using one episode of
each, until all episodes were played, each series then started again from the first episode.
Pia Ojanen
Brand new version of the Sofia Browser was available for the third trial. The XHTML
style markup language used for the browser is targeted to low end devices and is thus
considerably limited. The lack of up-to-date documentation, as well as continuous
development gave a challenge for the implementation. On the other hand the browser
had gone through major enhancements after trial 2, so for instance the performance had
improved significantly.
24
3.3.1.2 Programming languages
The server-side of the system was implemented using Java programming language. User
Interface was build using Thin Client Approach. This approach saves development
costs, since no application specific custom client is needed.
Real Player is used for playing all video and audio files (Figure 8). It is a default
application in most of Series 60 Nokia Mobile Phones and as such a familiar application
to more advanced users.
Current version of Real Player is quite limited when it comes to controlling videos and
video streams. User is not able to stop the clip and then start over again, or jump from
one part of the video to another, but it has to be watched from the beginning to end.
More advance versions became available during the project.
Figure 8. On the left on-demand video, Nelonen news, opened with Real Player. On the
right Podcasted video, episode of the series “Mothers and daughters”, opened from
phone memory with Real Player.
The user navigates through the pages and links by using the directional pad of the
mobile phone. The two menu buttons of the phone (located under the screen or next to it
depending on the screen orientation) are also used to open Podracing menu and as a
back-button.
25
In studies related to mobile browsing, text based navigation is often recommended.
Nevertheless, images were selected as navigation elements in this project, mainly
because they are well supported in second generation phones, such as the models in
question. Display is considerably larger than in the previous models and faster network
connection is available for downloading and streaming videos. It is thus reasonable to
expect that device is able to render small images at reasonable speed. This kind of
navigation model is often used in PC world.
On the other hand it was not considered during the design that some of the users would
be using only text based parts of the service (news feed). Browsing only text material
would benefit from the lighter text based navigation: The user might want to access
news content and already downloaded podcasting content also when available network
connection is very limited.
Browser supports paging model, so content was divided in three separate “frames”
(Figure 9), even though frame does not have the same meaning in this context as in
normal web development. It can be considered as one page element instead.
26
Frame 1 Logo of the service is located on the left side. It could be easily changed, if the
service model would be different.
Brand colours and logo are displayed on the left side. Brand logo will function also
as a link to brand main page.
Available content categories belonging to the selected brand, current category is
highlighted. This way user is always aware of her location on the service.
Frame 2 Navigation elements are located in this frame. Depending of the level it will be
either brand logos or text. Brand logos are already familiar concepts to users.
Frame 3 Content is always in this frame. Only videos will open in separate window.
If content does not fit into one view if will be divided to separate pages. Paging
model is already familiar from text TV. Arrow keys will move user forward or
backward.
Graphical icons for media types were added, so that user would always be aware which
media type, as illustrated in Table 3.
First entity in the content view was news. First view displayed 8 most recent articles
from the channel. User was able to select All News, if he was interested in a wider
selection.
News feed was automatically generated. The content came from the broadcasting
companies, which also provided metadata about news articles and video clips. In this
way news were categorized and sorted using different criteria.
27
At this point all news items were either text articles or video clips. They could as well
be audio clips, if they would have been available. Icon on the left indicated file type of
an item (Figure 10).
Figure 10. Main view of the service (left). Most recent news (right).
News News main view contains 8 most recent news article or video clip from the
main selected brand. Link to ALL NEWS is available in the bottom and it leads to
view news category view, if available.
Category Left side of the view contains navigation, containing pre-defined news
view categories. Every link will open news set of this particular category to right
side of the view:
Main
Local
Foreign
Economy
Media and IT
Sports
Entertainment
Others
Right side of the view contains list of articles, which are part of the selected
category.
Text articles had their own template, which contained article name, publication date,
timestamp and actual text. Version in the second trial had a possibility to display a
picture, if available, but it was removed from version 3, because image scaling was not
supported in new version of the browser. Related articles were displayed at the end of
each article. (Figure 11.) They were automatically fetched from the database using
keywords in the metadata of the article.
28
Figure 11. Individual News article (left) and Related articles (right).
Related A list of articles which have same keywords than the available article itself.
articles Links will lead to similar article pages with an article and related articles.
Series view contained list of available series. On the left side user was able to see the
state of the series. (Figure 12.) There were two possible options for the state: series has
been already ordered, or user is able to order the series.
When the user selected a series and clicked the link, the series navigation view was
displayed. On the left side all available episodes were listed.
If user ordered the series, only three most recent episodes were downloaded. It was
found out before the trial that if all episodes were available when user starts to use the
service it would take too much time to download them at once.
29
Figure 12. List of available series and status if they are ordered or not (left) and
Episode list and information for series. Episode is not yet downloaded (right).
Series Left side of the content area contains state of the order, options [order] and
[ordered]. User can change the state by clicking the link.
Right side of the content area contains names of the series, which are also links
to episode view.
Episode Left side contains a list of available series. It contains name of the series and
navigation, which contains all available episodes of this series. User can browse
all episodes using this navigation.
Right side Contains title of the episode, release time and status, whether the
episode has been ordered or not and is it available in the device or if it still needs
to be downloaded.
In the second trial Podcasting downloads were triggered by separate application, called
Podsync. The user had to open a standalone Java application in order to update new
series to his phone. The implementation was separated from the User Interface. If the
user ordered a series he was supposed to open Podsync application and launch the
downloading process. After all downloads were completed the user was able to see the
episodes from the User Interface.
In the third trial user was able to view status of all downloads active during the session.
This functionality was ported from the Podsync to the browser application.
After the episode was downloaded to the terminal, a link “[view]” appeared to the
screen. (Figure 13.) This link opened the video, now existing in the memory of the
phone.
30
Figure 13. Progress bar for downloading series to the phone memory (left) and Episode
is available for watching behind the link (right).
Podsync All downloads are visible in the Podsync status view. User can access this view
view from phone menu. This functionality is integrated to the browser.
Episode After episode has been downloaded by Podsync it will be available for viewing in
view the User Interface. [View] -link will open the episode in Real Player.
Broadcasting channels were available as links in their own view. Channel name and
short description were available as static content. The content was the same as offered
for Podcasting. DVB-H was not available, so broadcast was emulated with 3G
streaming. Content quality was a bit lower in streaming, but otherwise the service was
the same.
When the user selects a link to the content, Real Player becomes launched. (Figure 14.)
In the Content server there was a pre-defined set of programs which loop over and over.
When administrator wanted to change the episodes it was done in the server and the
change became effective immediately.
This was the most unreliable of the three different delivery methods. The phones used it
was not possible to use any other connection for video streaming, except 3G. Thus, even
if the user would have had a WLAN connection available and would have used it for
browsing and downloading podcasted series, as well as text based news, the streaming
with Real Player was operational over a 3G connection only.
31
Figure 14. Broadcasting view, available broadcasts and short descriptions (left). News
broadcast is playing after user has selected the link (right).
Broadcasting Most recent news clips and entertainment clips are available from
view Broadcasting view.
Broadcast Loop consists of most recent news or entertainment clip from the brand. It
loop will be updated frequently, as new content is published. Loop will be opened
in Real Player, using 3G network connection.
The concept of Theme Channels was introduced in the third trial. It was a concept
combining content from all Channels/Brands, even though this was not completely in
line with the original brand based approach. In the actual implementation there were
pre-defined theme categories and themes, which were always related to a category
(Table 4).
32
“Themes” were current topics, which were selected so that they would produce enough
hits in the database. Themes were kept constant throughout the field trials.
Lingsoft speech recognizer software, which was used for keyword retrieval from video
and audio files was not optimized for analyzing whole sentences, but rather it should
recognize individual words, for example dictations that have been defined in advance.
As the project had also engaged with this product the pre-defined theme set was used.
3.3.1.11 Prices
Price lists were added to the User Interface in the third trial of the project. There was no
actual functionality on these views, they were like reminders to users, giving an idea
how much the service would cost in real life. (Figure 15.)
33
4. Field test results
4.1 Podracing study aims and research methods
Virpi Oksman
The purpose of the field study in the third trial was to explore users’ mobile TV choices
in different everyday situations. Qualitative and quantitative methods were combined to
make sure that adequate data was collected. Semi-structured interviews, questionnaires
and media diaries helped to disclose users’ media habits and how they present their
expectations and preferences. We also asked the users to take some photos with the
camera phone about the situations in which they might use the mobile TV service, and
of other places, things or contexts that are important for them. This helped us gain an
understanding about the role of media in the users’ everyday lives.
The field test started in May 2007 with eleven households who used Podracing service
with Nokia N92, N93 or N95 phone for a period of three months with Elisa
subscription. Before the test period, the users were interviewed and they received
information and instructions concerning the test. Demographic data and media user
profiles were gathered from the informants. The ages of the informants ranged from 24
to 56. During the test, the main informants carried the testing phone as their primary
mobile, using it for both professional on personal communication. During the test
period, the informants reported their user experiences in a test diary. After the test, users
were asked to fill in an evaluation form and they were interviewed again. The
informants had different kinds of hobbies, lifestyles, media profiles and interests.
The test period was overlapping with summer holidays of most of the test users. A
typical summer holiday in Finland is four weeks.
Ville Ollikainen
Total of ten users were logged in the trial. Five of them had the service available about
three months, two approximately 2.5 months, one about 2 months, one 1 month, and one
3 weeks. One user took part in interviews after using the service.
An average user was accessing the Podracing service during 22 days within the trial
period, which means that in general the service was not used on daily basis.
Additionally, a holiday season affected the usage, which made it less feasible to analyse
34
how the usage changed over the test period. Furthermore, any analysis comparing
weekdays would also have been affected by the holiday season.
Keeping these restrictions in mind, Figure 16 presents the usage over the entire test
period: The first users got the service in mid May and the second wave in June. The end
of the trial was not fixed, and the number of users got lower towards the end. The figure
indicates that the holiday season (typically in July, but may vary) did not have and
significant quantitative influence on the usage.
Activity
A typical user was trying out different functionalities in the beginning of the test period,
and the use of the service became more consistent after the first week. This is why the
first week is excluded from all analysis unless otherwise mentioned, and discussed
separately.
There were two different log files: log file of the Podracing service and log file of the
streaming server. The streaming server logged all activities concerning streaming on-
demand video services and broadcast emulating video loop services. On-demand events
were seen in both logs, whereas statistics of loop services were available only in
streaming log.
In the streaming log the users were not identified. Most of the on-demand requests were
found from streaming log by comparing time stamps between the logs. Information
about the success of on-demand content requests was available in the streaming log:
35
According to log files six users were able to view streaming on-demand content
successfully. Two users had attempts, but without success, and for two users there was
not a single attempt for on-demand video in the service log.
Because the loops used the same technical implementation we can assume the same six
users were able to access loop content, whereas four did not.
Eight users were successfully viewing downloaded Podcast content through the service.
It should be noted that easy of download was depending on the individual configuration:
In most cases there was Wi-Fi available for quick download, but 3G made the download
sluggish.
All ten users were able to browse text articles. When comparing media formats, all
comparisons, if not otherwise mentioned, were scaled to the amount of usage per user:
The number of users was 10 for text based services, 8 for downloaded content and 6 for
on-demand and video loop services.
Compared to other media formats, the amount of audio content was minimal, and it did
not change over time. Only four users were listening to audio and the number of audio
accesses was only 8, so audio is excluded from further analysis.
In some cases it was difficult to determine, if there had been success in watching an on-
demand streaming video. Consequently there was a small difference between service
log and streaming log results, which is corrected to the figures.
Since loop video accesses were not logged to the service log, no individual analysis for
the loops can be made. However, since we were able to calculate success rate of on-
demand services and we were able to see the amount of loop service accesses, we can
estimate the number of attempts to access loop services in a comparable manner.
36
Table 5. Comparing media formats.
Loops Other
Text On-demand Podcasting
(est.) Podracing use
Percentage of
29% 21% 7% 35% -
sessions
Average duration
2’02” 3’02” 12’57” 1’56” -
of a session
Average total
duration of 12’50” 14’10” 19’25” 14’57” 1 h 08’
media consumed
Broadcast emulating loops got highest click rate, but the duration of a session was
shortest. Downloaded content got fewest clicks, but when it was used, the sessions had
by far longest durations. Average amount how much each one of these media formats
was used, was surprisingly equal.
As we see from the figures, the use of Podracing service alone, without accessing any
specific media content seemed to be most popular.
Figure 17 illustrates session durations when browsing text based content. An average
text article was fairly quick to read through, average 1’18”, but typically several text
articles were read one after another. Sometimes the end of the last article was not found
from the log, since opening a text article may had been the last action from the user in
that session. For these cases the duration of reading last article is considered being the
average duration.
37
Text sessions
20
18
16
Number of sessions
14
12
10
8
6
4
2
0
<30" 30"-1' 1'- 1'30"- 2'- 2'30"- 3'- 3'30"- 4'- 4'30"- >5'
1'30" 2' 2'30" 3' 3'30" 4' 4'30" 5'
session duration
No on-demand streaming content was watched more than 10 minutes in a row. Short
durations dominate the usage, as illustrated in Figure 18.
On-demand sessions
12
10
Number of sessions
0
<1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10
session duration (minutes)
38
Downloaded podcast videos, on the other hand were watched either as a quick preview
or as the whole episode from the beginning to the end (Figure 19).
Downloaded video
5
Number of sessions
0
<3 3-6 6-9 9-12 12-15 15-18 18-21 21-24 24-27 27-30 30-60
session duration (minutes)
According to the log data people were using the service mainly in the early morning and
late night (Figure 20). Despite of higher activity, a typical night time session was shorter
than average.
18 % 4:19:12
16 % 3:50:24
14 % 3:21:36
12 % 2:52:48
10 % 2:24:00 N_session
8% 1:55:12 Time used
6% 1:26:24
4% 0:57:36
2% 0:28:48
0% 0:00:00
00:00- 02:00- 04:00- 06:00- 08:00- 10:00- 12:00- 14:00- 16:00- 18:00- 20:00- 22:00-
02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00
Figure 20. Prime time of Podracing service. Bars present percentage of sessions in two-
hour slots, while line shows total time of usage.
39
As mentioned, the holiday season affected the results, so the analysis over weekdays did
not show significant differences. Some of test users estimated that they might have used
the service in later hours because of the holiday season.
It can be reasoned, that while the first week represents time of trying our different
functionalities provided by Podracing service, it also represents time of using the
service as it is assumed to be. In this view the first week may show, what would be the
wishes from the user, whereas the rest of the period shows the use in everyday life. The
first week means the first week of each individual test user.
First of all, let’s have a look at media format comparison (Table 6).
Table 6. Comparing media formats, first week only (rest if the trial in parenthesis).
Loops Other
Text On-demand Podcasting
(est.) Podracing use
Percentage of 24% 27% 6% 45%
-
sessions (29%) (21%) (7%) (35%)
Average duration 2’04” 3’37” 30’37” 2’19”
-
of a session (2’02”) (3’02”) (12’57”) (1’56”)
Average total
duration of 3’19” 6’38” 11’29” 7’01” 44’
media consumed
There were no significant changes in the proportion of each media format between the
first week and the rest of the trial. Downloaded podcast episodes were viewed
considerably longer time. The short duration of using on-demand services can be
explained by technical difficulties in the beginning.
Prime time during the first week (Figure 21) shows most activity taking place during
late afternoon and early evening. This is in contrast to the rest of the trial, in which early
morning and late evening were dominating.
40
Podracing usage by time of day, 1st week only
20 % 2:09:36
18 % 1:55:12
16 % 1:40:48
14 %
1:26:24
12 %
1:12:00 N_session
10 %
0:57:36 Time used
8%
0:43:12
6%
4% 0:28:48
2% 0:14:24
0% 0:00:00
00:00- 02:00- 04:00- 06:00- 08:00- 10:00- 12:00- 14:00- 16:00- 18:00- 20:00- 22:00-
02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00
Figure 21. Prime time of Podracing service during the first week.
Other analysis did not show substantial differences in the usage between the first week
and the rest of the trial period.
The idea of Podcasting was liked; it was perceived especially handy during holiday trips.
Podracing was independent from time and place. It allowed users to watch content in
privacy according to personal preferences. It also turned out to be convenient during
trips. The test users estimated that it increased TV watching in general, as expected.
“I liked the Podcasting concept very much. I used it mainly through 3G. But
it was very slow. It was very handy during summer holidays, especially
during our trips by car. But when I left outside Helsinki it was much slower
without 3G networks.” (Woman, 26 years)
Certain challenges were reported, concerning 3G networks, limited battery life and
lacks in usability.
Podracing prototype was used over both WLAN and 3G connections. The test users tried
to use WLAN in public places as much as possible, but often they had to find out for their
disappointment that the mobile phones did not always connect to the WLAN networks
when it was supposed to do so. 3G networks had lack of coverage even within cities.
41
“Only in Lasipalatsi I was able to connect to [3G] network.” (Woman,
51 years)
“In the bus 3G switches to 2G and then become the hiccups.” (Man, 27 years)
Battery life was another concern. Most of the test users were satisfied with mobile
phone battery life, if they loaded the battery once a day (the test users with N92 and
N93), and the use of Podracing and other media services was quite light during the day.
Yet, with increased media use, the battery life became obviously a problem (the test
users with N95).
“If I forgot to close some programs, I found very soon that the battery was
dead. Then if I did not use Podracing at all, I didn’t need to do anything for
it. Phone itself does not consume the battery too much, but all the other
media use discharges it quickly. Let’s say that if the battery was full in the
morning then, by the end of the afternoon, the battery ran out again. It was
enough only for just opening and closing Podracing.” (Man, 44 years)
Some users considered navigation cumbersome; it would have been easier to move
directly to right or left. There were also a number of popups and confirmation dialogs,
which were reported annoying. Some users were accessing downloaded content by
accessing them from mobile phone file explorers instead of Podracing service. In this
case the file names were not self-evident.
“You have to browse through one page before getting to the next. Of course
it was annoying when you have to go sort of round and round. Could not
just go to left and right.” (Man, 44 years)
“When upgrading to the latest version it began to nag if it can use the
information in the phone, it sometimes interrupted the download and began
the nag every minute. Had to press buttons all the time.” (Woman, 27 years)
“When I was on a train I had to guess, because I was not able to figure out
from the series of numbers, which episode was which.” (Woman, 35 years)
42
4.4 Use contexts and times
Elina Noppari
As stated in previous trials of the Podracing project, the usage of mobile services
usually occurs in three user areas: home, work and public. These were also the main use
environments during the third field trial. People were using the service in many different
locations from bathroom to bed. Mobile television was viewed when they were
commuting and travelling abroad, as well as in the bar or on the summer holiday. For
some users mobile television was clearly a private medium, while others sometimes
shared their viewing experiences and the usage had more social elements.
In this third, but also in the second field trial of Podracing project the users were asked
to take pictures, when they are using the service. Figure 22 shows a couple of them.
Figure 22. Pictures taken from where the service was used.
43
“I tested this all the way in Norway. During my one week holiday I watched
Elisa [YLE] news recordings. I did not mind one day delay. And then I was
sailing and travelling across Finland. The service did not work at the
summer cottage, which was annoying. I did like the idea though, that you
could like record with this. It is very good when you are travelling – you just
have to chase down WLAN spots.”
“You are able to watch it any time any place. It is good recreation during
long trips and nice playing, what series to watch this time. I usually don’t
watch much TV. It was possible to watch something at the end of the day at
work. It was not so much a social happening, but more like a private thing.
Although I also watched it with my friends mainly to show how it works.”
When asked more closely which content types they preferred in different use
environments and situations, test users said that the use of the service was more
versatile at home, than it was for example while commuting or travelling. Different
content types were tested more often at home than in public area.
To some degree this result may ascribe to 3G network problems and battery endurance.
The coverage of 3G was not flawless and interrupted viewing every now and then, also
in the city area. The users also had problems with WLAN: mobile phones did not
connect to WLAN in many places where the wireless network worked well with
portable computers.
“The only time I was able to get wireless connection was in Pärnu, on the
beach. Not even once I managed to do that in the city.”
Some interviewees stated that loading podcasts consumed the battery, so it was usually
done when the telephone was plugged. Usage problems probably channelled the use
conventions and it is hard to say how the testers would have used different content types
44
if the technical limitations had not been there. Some users criticized this emphasizing
that battery problems damaged the whole idea of mobile television.
“Couple of hours is the maximum. If you are transferring the data, battery
endurance is really weak. So in practise you have to keep the phone plugged
all the time. So is this television anywhere and anytime, as it is marketed?
You have to find the electric wall socket, if you really want to use this.”
However, most interviewees did not seem to mind very much that they had to plan their
use beforehand. Although typical mobile use is spontaneous, most users did not consider
it inconvenient to subscribe and download podcasts at home for later watching elsewhere.
Four users said that they loaded and viewed podcasts regularly in different locations.
Loading was usually done in the evening at home and viewing the content was
spontaneous. According to the testers, Nelonen podcasts were the content they used
most when they were away from home.
The number of the test users was limited, but it became quite clear, that each and every
one of them had their own daily routines and habits, which affected their use of mobile
television services. Some quite simple choices, like commuting with a private car or in
public transportation, effects the way mobile services are used. People who regularly
use public transportation have plenty of empty moments in their day to watch mobile
TV, while private car users view mobile television at home, sporadically at work, in
special occasions or do not find so much need for it. But interviews showed that mobile
television is not solely a public place medium, but is used in many different places in
the private area. It is presumable that the use of mobile television will become even
more versatile as people get used to the service and absorb it in their everyday routines.
Elina Noppari
As Figure 23 presents, test users found broadcasted news and podcasted entertainment
series most interesting. In the figure not only the content, but also the transfer mode and
channel brand is present and all these factors effect the interestedness of content to some
degree.
45
This result is similar to some earlier mobile television studies, saying that news is very
often the most popular mobile television content, and news viewers prefer to have live
feeds. And entertainment again might be popular podcasted or loaded content.
Elisa TV-channels
Theme channels
Nova
0 1 2 3 4 5 6
N
Figure 23. Content types that test users considered interesting or very interesting (N = 11).
As one notices, users did not like entertainment loops as much, saying that they want to
see the series from the beginning, even though mobile viewing is often short-term and
fragmentary. Some users stated that Finnish entertainment content is too rational for
loops: trash entertainment might work better; there is no need to see the whole show.
The irritation of not seeing the beginning of the show can be generalized to apply
also to broadcast entertainment in mobile television.
“Loops do not fit into a mobile phone. It is irritating to start viewing in the
middle of the program.”
Although the users preferred news loops, podcasted series of Nelonen came off well.
Only one user had tested podcasts before the trial. After the trial, all of them considered
the possibility of podcasting or subscribing programs as a major addition to a mobile
television. It makes the viewing even more flexible.
Most users did not view all the podcasts they have ordered during the field trial.
Anyhow, seven of them said that the podcast series of Nelonen were the content they
used most during the field test. Actual viewing of podcasts might have been even more
active if there weren’t problems with battery endurance and loading.
46
“It was really nice to watch the recordings. It was nice to watch them when
resting.”
Six users stated that also the channel brand effects on their viewing choices. Above all
this came up when talking about news and documentary content. Especially Finnish
Broadcasting Company YLE was considered a reliable news brand. Brands were not as
important when talking about entertainment. And naturally there were also those who
did not mind the brand as much at all. Some users simply chose the brand which was
first in the menu.
“I don’t care about who produces the news. The entertainment news of MTV
and Nelonen could have been in same heap.”
One has to remember, that also actual programs had a strong effect on users opinions.
Six of the test users did not find available content for their own preferences. Channel
Nelonen was evaluated more interesting as MTV, but this hasn’t so much to do with the
channel brand, but program supply. The pilot carried only two MTV entertainment
programs: a soap opera “Salatut elämät” and cooking program “Ruokala”. Nelonen with
its various entertainment programs had more to choose.
It is obvious that users do not view regularly a program they do not find interesting.
They might do it once or twice for testing purposes and leave it there. It is also
presumable that user preferences might have been different in a younger user group. In
that case, the entertainment content would probably have been more popular.
The length of the programs limited the use of entertainment content. Most test users
considered conventional TV series much too long for mobile viewing. According to
them, mobile television programs should be impulsive and short. The desirable length
could be 5 to 15 minutes. This result is also very similar to previous mobile television
studies saying that average mobile television use lasts less than 10 minutes. Use might
last a little longer when a special event is followed on mobile television. Mobile
television pilot done during the FIFA World cup showed that majority of testers viewed
mobile television 15 to 30 minutes, and there were those whose viewing lasted more
than an hour. But also in this trial use time reduced quickly after the event.1
1
http://www.sofi.unigoettingen.de/index.php?id=584&no_cache=1&tx_drblob_pi1%5BdownloadUid%5D=670.
47
Since the use situations are often very short, some users valued the text format as the
most suitable media type for mobile terminal. It is easy to flick through text headlines
and skip over uninteresting topics. Users emphasized that not all the news are
interesting and especially in mobile viewing situation it is frustrating to wait the
compelling topic.
“I was not inspired to view longer series. You suddenly have a tiny moment
for viewing, and viewing the series is always interrupted. – I watched short
humour pieces on Elisa-TV and news, and I may have put music on at the
same time. You have to follow longer series all the time to stay with them. If
there had been shorter podcast pieces, I probably would have subscribed
them.”
“I watched Channel Nelonen, because it was first on the menu. I did not
watch the series, because I do not normally watch them and they are too
long to be watched in mobile phone.”
Some users hoped for a mobile service, where text news would have been
complemented with short video clips, like it is done in several web pages. They also
considered news links, offering similar news topics as the original one, as useful.
Test users were also asked about the interactivity of mobile television: do they expect
mobile television to be more interactive than conventional television, and if so, what
interaction they consider as interesting. Three out of eleven users said they expect
mobile television to carry interactive elements. What those elements should be was even
more difficult question to answer.
None of testers were interested in chat, where users could discuss about the programs
with each others. Three of them stated that they might be interested in user-generated
content and even be willing to take part in content production (e.g. sending pictures or
videos to the program with mobile television). Four testers said that they would like to
give feedback, if only this possibility is implemented in a simple way.
48
“I don’t usually participate anything using my mobile phone. Nor do I
participate on chats. I would probably take a picture if I was on the scene,
and I would send it to somewhere. The genre could be whatever from news
to entertainment. But I wouldn’t necessarily watch that kind of a program.”
Even though the test users were not so much into an interactive mobile television, their
interviews revealed that somehow user’s expectations of mobile television have
increased. Some users stated that Podracing pilot got positive attention in their circle of
friends and still has a certain status of novelty, but the others said that a mobile terminal
showing moving picture is no longer exceptional or interesting as such.
The users stated that a mobile television service should be able to give a feeling that a
user actually knows and gets more than others. Producing that kind of a real-time
service is challenging, especially in present media environment where everything is in
internet in the delay of ten minutes. If a person is leaving from home to work and opens
up his mobile television in the bus, he expects to find something extra that he wasn’t
able to get from the newspaper, morning TV show or internet at home.
“(There should be) some new stuff that I could show to my colleagues, hey
look what is going on.”
Users hoped that a mobile service would match up to their real needs and use situations.
Whilst the use of mobile television is usually very spontaneous and low commitment, it
seems that some people expect to have niche content even for those short moments of
viewing. Just surfing around and viewing meaningless program flow is not necessarily
enough. Even one of key motives of mobile television use is killing time people may
want to kill boredom with special content, which is just suitable for that moment.
“You have a minute to use the mobile phone. The service must have an
answer for that moment.”
As people’s daily routines and needs are very different, there is a demand for feature
rich mobile television with customized elements. Some users stated that they would like
to have tailored services and the idea of theme channel got positive attention. In this
pilot the theme channel key words were written in advance and they were not
adjustable. This naturally cut down the actual use of the channels. However, the concept
of theme channel got understood and seven out of eleven users liked the idea.
49
“Opinions and preferences change, so you have to be able to change the
content, too.”
To summarize this up it seems that news are still the most interesting content in mobile
television, even though people also appreciate the possibility of subscribing the
podcasts. However, conventional entertainment TV shows are considered too long for
mobile television. Many people prefer short and even niche content for their brief
moments of mobile viewing. Their expectations have increased: many of them hope for
customized services, which could give them a feeling that they actually get more than
others, thanks to a mobile television.
Elina Noppari
To demonstrate the variety of needs mobile television users may have, we created four
user profiles based on the test user interviews. To some degree all the users can be
described as early adopters, because they voluntarily took part in this kind of a study
and were interested in new technology. However, their motives and preferences were
different, as were their every day routines and habits. These profiles are more like ideal
types: they are based on real characters, but generalised and blended. These fictional
user ideal types authenticate that people have many different expectations for mobile
television and all these needs should be taken into account in concept designing.
An independent content seeker is a twenty something male, who is commuting with his
private car, is socially active, interested in new technology and is a very experienced
media consumer. Independent content seeker makes the most of peer to peer networks,
his internet use is versatile and he is neither dependent on television program schedules
nor the recording devices. He is an individual user, who finds the most content he is
interested in, in the internet.
“I don’t watch so much television, mostly DVDs and loaded internet series.
Actually I watch only Formulas as live. I prefer to watch series at my own
pace, not when the channel says so. The programs I’m interested in, I find in
the internet. I’ve been using the net so long, I know what you can find there.
MTV3 is charging one Euro of Salatut elämät previews, but you can easily
reach those episodes in peer to peer networks for free.”
50
The independent content seeker does not use public transportation, and this affects his
mobile television use. Mobile services are often used for killing time when waiting, but
this kind of a user does not have so many idle moments in his day. When he is at home
or at work, the internet is available, and he does not feel he needs mobile terminal so
much for up-dating or leisure. He is a selective user, who does not listen to the radio in
his car, but his favourite CDs. His mobile television use does not take place in public
sphere, but sporadically at work and at home.
As he is used to get quickly what he wants, his expectations of mobile television are
relatively high. He wants it to run smoothly and he is not willing to wait. An
independent user might choose the text as a mobile media format, because uninteresting
topics are easily skipped over. Because he is a selective television user and arranges
special movie nights, he finds regular programs, let alone the films, much too long for
mobile television. Formula race kind of an event might be an exception, if there
wouldn’t be conventional television at hand.
Content seeker is an experienced media consumer and does not mind so much about the
brands. He trusts that he is able to interpret and evaluate current content, no matter who
has produced it.
Socially oriented user is a thirty something female, who often shares her mobile
television use experience with other people. Even though mobile terminals are usually
considered as private devices, this user likes to show clips of her favourite content to her
friends. When she goes visiting her grandmother to the hospital she wants to cheer her
up with mobile television.
51
“My friends are still interested in mobile television. It is fun to watch this
with friends. I showed this to my 80 year old grandmother, who has senile
dementia and who seldom responses to any impulses. She said “Television!”.
It was a touching moment.”
Occasionally she uses mobile television at home and wants to share this experience with
other family members. For this user, mobile television has same kind of characteristics
as conventional television. But unlike the normal television, she is able to take the
mobile device with her. Viewing situations may occur when she is lying with her
husband in bed, side by side.
“I was watching this with my significant other. We were in bed and it was
nice.”
As mobile television still has some novelty value for her, she often asks her son for tips
and counselling. Not only the technical advice, but also hints about interesting programs
and content. Mobile television works as a good conversation topic with a teenager. Her
son has introduced her to some alternative content, like user generated programs. She
even finds it possible to participate in content production – at least if she would get
some compensation out of it.
“If I were on the scene, I could take a picture and even send it. Especially, if
I get some benefit out of it – price reduction for example.”
Sometimes socially active user views mobile television in public places. She does not
want to miss a common viewing experience of big, once in the life time events.
However, when watching Eurovision song contest in a bar, she was a bit embarrassed.
Even though she wants to share her mobile television experiences, she does not want to
attract too much attention.
“I noticed that when I tried this during the Eurovision song contest, I got
some... people had this attitude; is she bragging or what. An adult playing
with her telephone...”
A rational time killer is a forty something male, who uses mobile television quite
traditionally when he is commuting or is travelling for business purposes. Mobile
phones are often considered as devices used primarily in public area.
52
Every morning rational time killer takes a bus to work and during his 25 minutes trip to
town, he wants to reach fresh and new content. He is also pretty selective. Even though
the mobile television is used for killing time, he is not interested in trivial content. He
wants a mobile television to give him a feeling that he actually knows more than others.
This “knowing more” has a social aspect: he likes his role as a source of information
and wants to present new conversation topics to his colleagues when he gets to work.
“I’m interested in current topics. This service should give me a feeling that
I know more than others. If I get same information from other media, why
I should follow this? If I have a mobile terminal and I’m on the move, I’m
interested in topics that are relevant in that mobile situation.”
As he expects mobile television to give him something extra he did not get from the
morning newspaper or morning TV show, he is interested in customized and personified
content. For that reason he values the theme channels in Podracing.
“This (mobile phone) is my private thing. Today you expect to have more
tailored features. Theme channels with updated keywords would probably
have been the service I used most. It was customized. If I have 50 news
headlines and I have to hunt for interesting topics. If someone is classifying
those for me, I read them. And I’m ready to pay for them.”
Rational time killer does not use his mobile television in private area, and he does not
even understand how the device could be used at home. He is not interested in mobile
entertainment and uses his mobile television in short sequences.
“No one would watch programs on mobile device at home! And if I were
hooked on some TV-series, I would record them, instead of watching them
on small screen outside of home. I’m too old; I’m not attracted to watch 40
minutes long series on this. This is far too small and the programs are too
long.”
Channel brands have some relevance for this user. Even though he seeks also for
alternative sources, he thinks that for example Finnish Broadcasting Company YLE is a
very reliable news brand. Because he is first and foremost interested in fresh
perspectives, he hopes for mobile services where short video clips from different
sources are put together with text news. Even though he is a bit traditionalist, he is
willing to view and even to produce user-generated content – if only he would witness
some newsworthy situations.
53
Rational time killer is also a bit of a techno freak. He expects the service to make good
use of all technical features.
“One generation has passed, but the content has not developed as well as
the device. The clarity was not as good as the terminal would have enabled.
If you think about mobile device with multi media features, you expect the
service to be consonant with it.”
D. Entertainment enthusiast
Entertainment enthusiast is a thirty something female, who finds the entertainment most
interesting mobile television content. She is hooked on many TV series, and does not
want to miss an episode while she does not have a chance to watch conventional
television. Even she has recording equipment she finds mobile television more
convenient way to stay updated in courses of her favourite shows.
Especially she likes the possibility of podcasting. Subscribing podcasts makes mobile
television even more flexible. She values the TV anytime and anywhere experience.
“It was possible to watch whenever you wanted to watch something. You
had that chance. And it was possible to watch programs you liked. The idea
was great and amplified my television viewing.”
As entertainment enthusiast regularly follows several television series, her life is pretty
much in routines. Her week scheme is strongly affected by media, and following it
makes her feel secure and gives her the feeling of life control. Because of that she even
wants to follow domestic programs when she is abroad. Even though she is not
normally into mobile news, any domestic program will do when she is travelling.
“I was using this in Norway. It was really great to watch recorded YLE
news in Norway! I also tested this in Tallinn. I liked the concept. It was
great when you had a summer holiday.”
These four ideal types were based on eleven test user interviews. Because of small number
of users, no wide generalizations can be drawn, but even this amount of users embodied
several use routines and needs. If there were more test users from different age groups and
social classes, the spectrum of user profiles would have been even more diverse.
54
5. Mobile TV pricing and business models
Jorma Riihikoski
Mobile broadcasting can generate revenues from three sources, namely from
• customers
• advertising
• broadcast network access fees.
Customer revenues are based on the additional value of mobile TV services. There are
several charging mechanisms to be used: A monthly subscription fee is considered the
most acceptable according to nearly all pilot market research, but pay-per-view or time-
based charging are other options. Mobile broadcasting provides a new distribution
channel for TV content, including advertising. Thus, selling air time to advertisers can
be considered as an additional income source for broadcasters. [Sattler 2006.]
At this moment in Finland mobile TV audience and markets are so small that the
advertising revenue streams don’t play any significant role. There are also different
ways for revenue:
a) One time revenue which is created when device is sold or bundled to the
operator’s service.
b) Monthly access fee which is paid with telephone bill.
c) Pay-per-view or time-based charging for other extra options and services which
user selects personally. All these revenues will be shared to some extent between
all players in mobile TV value chain.
Companies get profit already from some mobile services like mobile votes, ring tones
and interactive TV, but the mobile TV’s value capture models and roadmaps are still
under the construction. One question is if the value is captured from user’s payments or
from advertisement incomes. One possibility is that it could come from pay-TV package
which’s one attribute is to have mobile TV in same contract. The investments are going
to be at low level because of a) waiting for more clear business model, b) higher
equipment penetration c) and better knowledge about what the users really want to
watch.
55
5.2 Mobile TV cost streams
• infrastructure
• content
• device subsidies (possibly), but also
• marketing and sales expenditures.
The largest part of infrastructure costs attains to broadcast network rollout and
operation. Mobile broadcast networks are expected to have good indoor coverage,
because pilots have shown a high usage even indoors. So coverage and quality of
reception are heavy cost drivers. Infrastructure costs also depend on the planed
coverage.
The main content costs relate to content creation and provision, the adaptation of
content formats to the requirements of mobile reception, but also the acquisition of
content rights and licenses. [Sattler 2006.]
In most of the studies on mobile broadcast business models, the following main roles of
the value chain have been identified [Sattler 2006]:
In the most cases the revenue is split between evenly between content provider, mobile
operator and broadcast service provider.
56
Financial design criteria in this context relate to the costs of the network build-out, the
revenue sharing agreements, and the business-to-consumer billing formulas.
Mobile network operator led model (Figure 24) seems to be quite near the situation in
Finnish mobile TV trials.
In the mobile network operator-led model, the mobile network operator handles the role
of mobile broadcast service provider, manages the end-relationship with customers and
is responsible for service provision, marketing and customer care. For the service, the
mobile network operator will need to purchase content from broadcasters and other
content providers.
There are also several decisions to be made by operators, broadcasters and consumers.
The following criteria were used to describe the financial design decisions [Braet &
Ballon, 2008]:
(a) Cost sharing agreements. This first financial criterion describes how different
actors carry the costs of the service rollout. Three cost categories are taken into
account. First, the device cost refers to the primary purchase cost of the handsets
and to what degree the consumer has to pay the entire cost of the handset, or
whether device subsidies are allowed. Second, the network infrastructure costs
refer to the cost of building the transmission infrastructure. Third, the content
57
and application costs refer to which partner carries what part of the content
and/or application development cost.
(b) End-user billing. This criterion describes the ways in which the user pays for the
services provided. The billing formula will depend on the kinds of product
bundles offered, but does not follow directly from that criterion.
(c) Revenue sharing agreements. The last criterion describes the ways in which the
service supplier(s) agree on how the revenues generated through end-user billing
are distributed throughout the value network, including the broadcasters, other
content providers, and the mobile network operators.
Braet and Ballon [2008] also found out that no clear business case can be made before
one knows in what ways the collected revenues will be redistributed from the ‘customer
owner’ to the broadcast network operator and the content aggregators. From the little
information they obtained during interviews with the executives, it appeared that the
broadcast network owner and the customer owner will divide the lion’s share of
revenues. Content aggregators that do not own spectrum do not build and operate the
network or do not have a direct customer relationship, have expressed fears that revenue
share agreements may turn out to be suboptimal for them.
This section is based on interviewing mobile TV content and service providers: Teemu
Lehtonen / MTV Oy, Marcus Wiklund / SWelcom Oy, Timo Argillander / Digital Media Oy,
Jonas Kronlund / Elisa Corporation and Ari Pöyhtäri / Sofia Digital Oy.
There will be all kinds of possible contents mixed up in Podracing service and it doesn’t
matter who owns them. That was the original plan when the project started. Aim was
also create best possible and useful platform and interface for users (customers).
Content should be available divided to different media formats and brands such as
MTV3, Radio Nova or Nelonen. Fundamental aspect is that everybody should be in
control for their own media and be able to sell advertisement for their own brand.
Example of that is Google news which aggregates services from different providers,
which is good for them but is not thought very satisfied model by the service producers.
While it may be beneficial from customers, the model for value capturing is prone to
become complicated. There should always be an aggregating host who is taking care of
58
all services and business activities. In this respect the model presented in Podracing,
where content is accessed through media brands, could be a successful compromise.
The content providers are key players when we are thinking about different sectors of
users and their models for watching the mobile TV. The service could be also produced
only by one media company in co-operation with the operator. Relevant question is that
if Podracing is very wide national service it could take long time to get it together and
meanwhile there could be other technologies like IPTV where people could customize
their own media- and web-services by RSS-feeds. Getting together services and
contents from different content providers and selling them to right user segments may
turn out to be a very hard task for company which is acting as an aggregator. So it is
easier to come to the conclusion that services are in future produced by companies or
their brands and not collected as wider spectrum service. It is possible that users in the
future can filter their contents through web’s search pages.
Media companies’ portfolios are today already in internet. Their usability and
commission gives the roadmap for future development. There are also services in IPTV
which are very near mobile TV business and benefiting of mobile connections. For
example UK satellite TV company BSkyB is taking a clear benefit of broadband by
complementing the satellite service with IP based TV transmissions:
Still several major questions remain unsolved: Are the operator’s content portals the
places where customers start to seek the service contents they prefer or are they willing
to go straight to the origins of the contents which means in this case the content
providers and broadcasters portfolios or portals? Which is the menu that customers use
to order the services for their mobile phones? It is also often the question of promoting,
marketing and training the customers to use the portal and services. Customers want to
59
choose the services from the menu, but there always have to be someone who is able to
create this supply. In many cases supplier is content provider or broadcaster. It could
also be the operators. But it raises the question why operators, because the broadcasters
could have direct relationship with customers. On the other hand, the operator is billing
the mobile users and they have direct customer relationship with the operator.
5.5.2 Business models and value added services created for users
The most popular services at Mobile TV at that moment are text news, videos and
simulated broadcasting. These services must offer so much added value that the price
level to pay is lowered and at the same time one must keep the promise given to
customers. Consumers are eager to pay just for the exact services that they feel will suit
them. The level to pay for whole portal is high, because the average watching time is
short. One solution might be a package agreement with the customer so that he gets
more services and that the ones he is interested in are included in the package.
One business model is so called 360 degree model which includes broadcast, radio, net
TV, mobile broadcast as a cross media. The mobile services are not spearhead services
when launching the business. In this model all players work in co-operation. Mobile
services are an extra business which gives added value to the users while they are
related to the mainstream services already available. The key idea is to provide the same
services to the different media so that they support each other. Individual mobile content
creates added value to the other existing media services. That way the product is more
planned and linked to the services in other medias. The model is built around the
traditional linear TV. Most popular shows will be popular also in mobile TV and web-tv.
The main channels should be included to attract the majority of potential users, but there
is also long tail thinking that opens mobile TV markets for different subcultures. What
is available in the web is sooner or later usable also in mobile phones. Interactivities are
going to be personified services and multimedia experience with radio channel (+audio
information about music) with simulcasted broadcast.
Another model is to keep the brands independent and choose suitable contents for the
right audience and sell ads to enhance revenue. Part of these contents are free and other
part are payable. That is model for straight business wise thinking. Operator have no
role in this model all the income is coming in from advertisement and pay-TV from
niche services which give exact added value to user. Though this model requires
limitless data transfer agreement with operator.
60
pay-TV channels. It seems that the end users of the services have to pay in future
otherwise the schema doesn’t work. Capturing the value through advertisement income
seems not to be enough to make Mobile TV activities in Finland profitable.
At this moment the role of Mobile TV is very small in business. The turnover is a very
small proportion of the revenue if compared to the one from web services. However
development in this sector is encouraging. Mobile internet is coming on strong in future.
What will be the new medias for using mobile web is to be seen. Short video material is
going to increase. Content for mobiles is multimedia: news, messages search services
and moving pics (short cuts). Customers definitely are attracted to live happenings as
Formula racing and other sport events. Longer movie cuts which are watched from
phone’s HD in special occasions as during travel will be common. Even now people use
PSP consoles in planes and trains to play and watch movies.
Social media is already in web and in that way those contents like FaceBook will be
available in mobile phones. They are also free to use. Mobile versions for these
applications are under construction. Mobile is no distant island, but is part of the
systems that are used to share services in different media, such as computers, game
consoles, laptops and TV sets. The real challenge for media companies is to create
compact mobile TV service aggregate for users. Media companies might try to create
new shorter mixes from their TV and internet services for mobile phones, which will
suit better mobile TV and its user’s needs and habits. Fitting example of this is Formula
One highlight news mixes specially made for mobile TV and music videos. The step to
make content which is only aimed for mobile TV use is very high. Media companies
already make Web-only Content and it seems logical that this development will reach
the mobile TV in future. The border between Mobile TV and Internet is slowly
disappearing which makes only mobile TV aimed content production problematic.
Mobisodes [Short cuts] and highlights from news and different series are probably the
most wanted content in mobile TV. Channels collect and recommend their TV-
favourites this way to the user. Chatting, playing and voting are part of interactive TV.
Mobile phone is the best device for that, because it is more personified than TV. Social
media in mobile TV as a format might be the other favourite. Market is too insignificant
in Finland for large scale mobile TV business and it is not going be real profitable
business in few years time. But it is definitely going to follow the stronger and faster
development happening in larger international markets.
61
5.6 Results of field test: Mobile TV and value added services
The test users were asked about their willingness to buy different kind of mobile TV
services and products. The field test users were most eager to watch live happenings
like concerts as Figure 25 presents.
6
5
4
3 users
2
1
0
1 2 3 4
1 fully agree - 4 fully disagree
Nine users out on ten agreed on that issue. Test users were also willing to buy episodes
of old favourite series or collections of them from some kind retro-TV shop as Figure
26 shows.
7
6
5
4
users
3
2
1
0
1 2 3 4
1 fully agree - 4 fully disagree
62
Buying and watching movies were not so popular, as Figure 27 demonstrates: only three
users out of ten showed interest in it. It seemed that size of the mobile screen and the
longer length of the movies prevented the watching.
Buying movies
6
5
4
3 users
2
1
0
1 2 3 4
1 fully agree - fully disagree
Test users were not so interested buying TV series supplementary products like DVD-
boxes with mobile devices, but they were more interested in buying supplementary
services like ringtones etc. which have connections to TV series. Half of the test users
were ready to subscribe or podcast the unseen episodes of TV series beforehand and
paying extra for that.
Test users preferred paying for mobile TV with the operator’s bill. It was also asked if
users would like the pay with different bill, but only two test users agreed on that issue
and eight wanted to pay for the mobile TV with operator’s mobile phone bill.
Most users thought that best principle for pricing is to have fixed package price for one
month (Figure 28).
63
Pricing principle
6
5
4
3 users
2
1
0
ay
th
k
m
ee
on
/d
ra
/w
ice
/m
og
ice
ice
pr
pr
pr
li d
pr
of
l id
so
nt
li d
so
ou
so
am
Figure 28. Fixed price per month was the most acceptable pricing model.
Table 7 present test users answers to the questions how much they were willing to pay
for mobile TV content.
Some of the users compared the price of the concert and movie to the price of a ticket.
Result was that highest price they were willing to pay was about 10–20% of the ticket
price. Some of the users were not willing to pay anything for the mobile TV products or
services.
Most of the test users would like to have mobile package which includes mobile TV
services for certain amount of money as Figure 29 shows.
64
Package with mobile TV services
4,5
4
3,5
3
2,5
users
2
1,5
1
0,5
0
1 2 3 4
1 fully agree - 4 fully disagree
Test users were also asked the open question what would be suitable contents to be
tailored for the mobile TV. Presented as an open question, answering turned out to be
difficult, but following contents and services were mentioned:
65
6. Conclusions
Ville Ollikainen, Elina Noppari, Esa Reunanen, Jorma Riihikoski
The aim of Podracing project was to compare three different media formats (text, audio
and video), and three delivery methods (broadcast, unicast and pre-download) from end
user’s perspective.
In field trials there were no winners and no losers: Although the number of participants
in the trials was limited, mobile terminals were used in such a variety of situations that
no single media format and no delivery method was able to fulfil all needs. When
interviewed, the users presented a wish that a mobile service, such as mobile television,
should match up to their real needs and use situations. Whilst the use of mobile
television is usually spontaneous, it seems that some people expect to have niche
content even for those short moments of viewing.
Podracing had a unified user interface for all content and all delivery methods. The
users appreciated having similar logic in user interface. Furthermore, it was presented
that right content should be found within 30 seconds from launching the service.
Keeping in mind that a media brand is important, the model presented in Podracing
could be a successful compromise: the content was accessed through selecting media
brand first.
Previous studies suggested that mobile television involves low commitment. This was
contrary to the findings of Podracing: The users did not usually do anything else while
watching. Furthermore, just surfing around or viewing an arbitrary program flow seems
not be enough. Even though one of key motives of mobile television use is to kill time,
people may want to kill boredom with special content, which is just suitable for that
moment.
Just as mobile television, the Podracing service was not solely a public place medium. It
was used in many different places in the private area. It is presumable that the use of
mobile television will become even more versatile as people get used to the service and
absorb it into their everyday routines, thus increasing the complexity required.
Recent mobile television development has been focusing on broadcast. Mobile user
does not choose time to watch television. In Podracing, broadcast was emulated using
programme loops. Users did not like entertainment loops, saying that they want to see
the series from the beginning, even though mobile viewing is often short-term and
fragmentary. Some users also stated that Finnish entertainment content is too rational
for loops: trash entertainment might work better, as there is no need to see the whole
66
show. The irritation of not seeing the beginning of the show can be expected to apply
also to broadcast entertainment in mobile television in general.
Another trend has been to include video playback capabilities to MP3 devices, mobile
phones and even GPS navigators. Although typical mobile use is spontaneous, most
users did not consider it inconvenient to subscribe and download podcasts at home for
later watching. This lacks real time effect, which is not as important for entertainment.
Text and streaming video got almost equal amount of usage. Some users hoped for a
mobile service, where text news would have been complemented with short video clips,
like it is done in several web pages. In news services they also appreciated recommendation
links to similar news articles.
From business point of view, one prominent business model is so called 360 degree
model which includes broadcast, radio, net TV, mobile broadcast as a cross media.
Test users preferred paying for mobile TV with the operator’s bill. It was also asked if
users would like the pay with different bill, but only two test users agreed on that issue
and eight wanted to pay for the mobile TV with operator’s mobile phone bill. Most
users thought that best principle for pricing is to have fixed package price for one
month.
Even though the testers were not so much into an interactive mobile television, their
interviews revealed that somehow user’s expectations of mobile television have
increased. Some users stated that Podracing pilot got positive attention in their circle of
friends and still has a certain status of novelty, but the others said that a mobile terminal
showing moving picture is no longer exceptional or interesting as such.
The preferences and motives of different users were different, as were their every day
routines and habits. It was possible to create four user ideal types: independent content
seeker, socially oriented user, rational time killer and entertainment enthusiast. Because
of small number of test users, no wide generalizations can be drawn, but even this
amount of users embodied several use routines and needs. If there were more test users
from different age groups and social classes, the spectrum of user profiles would have
been even more diverse.
67
Acknowledgements
The authors wish to thank the partners of the Podracing project: Digita Oy, Elisa Oyj,
Lingsoft Oy, MTV Oy, Sofia Digital Oy, SWelcom Oy, the Finnish Broadcasting
Company (YLE), and Tekes – the Finnish Funding Agency for Technology and
Innovation.
68
References
Braet, O. & Ballon, P. 2008. Cooperation models for mobile television in Europe.
Telematics and Informatics, Vol. 25, Iss. 3, pp. 216–236.
doi:10.1016/j.tele.2007.03.003.
Burk, M. & McQueen, D. 2006. Mobile TV: Broadcast and Multimedia. 2nd edition.
http://www.informa.com.au/marlin/30000001001/MARKT_EFFORT/marketingid/2000
1235585/marlinsource/PDCinformaTMprodcatalog.
Hartung, F., Horn, U., Huschke, J., Kampmann, M., Lohmar, T. & Lundevall, M. 2007.
Delivery of broadcast services in 3G networks. IEEE Transactions on Broadcasting,
Vol. 53, No. 1, pp. 188–189.
Kim, Keun do. 2003. DMB in South Korea. 9th Meeting of the World DAB Forum
General Assembly. Available at http://www.worlddab.org/images/WorldDAB-398.pdf.
Knoche, H. & McCarthy, J.D. 2005. Good News for Mobile TV. Proceedings of
WWRF14. 7–8 July, San Diego, CA, USA.
Lee, J. & Jayant, N. 2006. Mixed-initiative multimedia for mobile devices: a voting-
based user interface for news videos. Proceedings of the 14th ACM International
conference on Multimedia. Pp. 611–614.
Repo, P., Hyvönen, K., Pantzar, M. & Timonen, P. 2003. Mobile Video. Research
Report 2003:5. Helsinki: National Consumer Research Centre.
http://www.kuluttajatutkimuskeskus.fi/files/4719/mobilevideo_2003_5.pdf.
Sattler, C. 2006. Mobile Broadcast Business Models. A State of the Art Study. Berlin:
bmcoforum. Link: http://www.bmcoforum.org/fileadmin/user_upload/Downloads/
Business_Models_Study.pdf.
69
Further reading
Cui, Y., Chipchase, J. & Jung, Y. 2007. Personal TV: A Qualitative study of mobile TV
Users. In: Caesar, P., Chronianopoulos, K. & Jensen, J. F. (Eds.) Interactive TV: A
shared experience. 5th European Conference, EuroITV Proceedings, Amsterdam, the
Netherlands. Pp. 195–204.
ISO. 1998. Ergonomic requirements for office work with visual display terminals
(VDTs) – Part 11: Guidance on usability (ISO9241-11:1998(E)). Geneva, Switzerland.
Murdoch, G. 2000. Television Trends: Organization and representation. In: Wieten, J.,
Murdoch, G. & Dahgren, P. (Eds.) Television across Europe. A comparative introduction.
London: Sage.
Perrot, P. 2001. DVB-HTML – an optional declarative language within MHP 1.1. EBU
Technical Review.
Perry, M., O’Hara, K., Sellen, A., Broen, B. & Harper, R. 2001. Dealing with Mobility:
undestanding access anytime, anywhere. ACM Transactions on Computer-Human
Interaction, Vol. 8, No. 4, pp. 323–347.
Picard, R.G. 2005. Mobile telephony and broadcasting: are they compatible for
consumers. International Journal of Mobile Communications, Vol. 3, No. 1, pp. 19–28.
Sauro, J. & Kindlund, E. 2005. A method to standardize usability metrics into a single
score. Conference on Human Factors in Computing Systems. Proceedings of the
SIGCHI Conference on Human Factors in Computing Systems. Portland, Oregon, USA.
70
Södergård, C. (Ed.) 2003. Mobile television – technology and user experiences. Report
on the Mobile-TV project. Espoo: VTT.
Wieten, J. Murdoch, G. & Dahgren, P. 2000. Preface. In: Wieten, J., Murdoch, G. &
Dahgren, P. (Eds.) Television across Europe. A comparative introduction. London: Sage.
71
Appendix A: End user questionnaire
A1
4. Missä paikoissa ja tilanteissa useimmin käytit Podracing-palvelun eri osioita?
Kerro myös vuorokauden aika, sisältötyyppi ja käytön kesto.
Muistele käyttöaikoja ja tilanteita.
Liikennevälineissä
1. Miksi valitsit ko. tilanteessa
Harrastuksissa
kyseisen
Ympyröi numero, brändin/palvelumuodon?
Matkalla
Mökillä
joka vastaa
Kotona
Töissä
lähinnä sopivaa Muualla, missä? 2. Mitä sisältöjä silloin
vaihtoehtoa: Logi: ____________ kuuntelit/luit/katsoit?
Nelonen entertainment loop 1 2 3 4 5 6 7 ______________________
Nova ondemand
1 2 3 4 5 6 7 ______________________
puheohjelmat
Teemakanavat 1 2 3 4 5 6 7 ______________________
A2
5. Mobiili-tv, sosiaalisuus ja interaktiivisuus
En osaa sanoa
Ympyröi numero,
joka vastaa
lähinnä sopivaa
vaihtoehtoa: Perustele miksi?
Mobiili-tv:n katsomista julkisella paikalla
1 2 3 4 5 _________________________________
pidetään brassailemisena.
A3
Appendix B: Business studies questionnary
Mikä olisi mielestäsi paras laskutusmuoto:
1. Puhelinlaskun yhteydessä
2. Erillinen laskutus
______________________________________________________________________
B1
Series title, number and
report code of publication
ISBN
978-951-38-7217-5 (soft back ed.)
978-951-38-7218-2 (URL: http://www.vtt.fi/publications/index.jsp)
Series title and ISSN Project number
VTT Tiedotteita – Research Notes 1376
1235-0605 (soft back ed.)
1455-0865 (URL: http://www.vtt.fi/publications/index.jsp)
Date Language Pages
June 2008 English, Finnish abstr. 71 p. + app. 4 p.
Name of project Commissioned by
Podracing Tekes, Elisa, Lingsoft, MTV3, Radio Nova,
Sofia Digital, SWelcom, YLE
Keywords Publisher
mobile TV, podcasting, DVB-H, media usability, VTT Technical Research Centre of Finland
business models P.O. Box 1000, FI-02044 VTT, Finland
Phone internat. +358 20 722 4520
Fax +358 20 722 4374
ESPOO 2008 VTT RESEARCH NOTES 2439
Mobile TV should be more than a television. The final report of Podracing project
The primary study question was formulated as following: If a user had
the possibility to watch the latest television news from the mobile phone,
or listen to the news on the radio, or read text news with the mobile,
what would he or she choose?