Free Software: Uses of Free Software and Its Implications in The Software Industry
Free Software: Uses of Free Software and Its Implications in The Software Industry
Free Software: Uses of Free Software and Its Implications in The Software Industry
Peer Reviewed
Vivek Shah is a Professor of Quantitative Methods and James Keefe jk07@txstate.edu is a Senior
Lecturer in the Department of Computer Information Systems at Texas State University at San
Marcos.
1
Abstract
Throughout the world, millions of ordinary consumers as well as businesses now
use “freeware” - open source software applications that are available either for free or
at a small price. This paper presents pros and cons of using free software either for
personal or business use. The economic impact of free software on proprietary
software developers is explored, as well as the business case for developing or
servicing freeware for public use.
Introduction
“Freeware” are open source software (OSS) applications that are either available
for free or a comparatively low price. Freeware is different from free software in that the
source code that is usually proprietary or not publicly available. Freeware is also
distributed for free (or at a very low price) and is usually a light version of commercial
software. Many “Free” software applications, on the other hand, have open source
codes; as a result, anyone can make modifications (including updates) to the code.
“Free,” in this sense, is not only about price but also about the freedom to access and
modify the code as well as the freedom to distribute the software.
To address the ambiguity between “free as in free speech” versus “free as in free
beer,” a number of alternative terms have been suggested. The expression “free and
open source software” (FOSS) includes both “free” aspects of free software. Software
that is distributed at no cost is usually referred to as “gratis,” whereas OSS is usually
referred to as “libre”, which means that anyone has the freedom to read, modify, and
redistribute the source code (http:// GNU.org/philosophy/free-sw.html).
2
This paper will provide an overview of free software as well as a discussion of the
pros and cons of using free software for either consumer or business use. Factors
considered include software performance, security, reliability as well as associated
costs. Further, the impact of free software on proprietary software developers as well
as the business case for developing or servicing FOSS for public use is discussed.
There has been a recent surge of interest in “open source” software development,
which helps developers at various different locations and organizations share code to
develop and refine programs.
In 1980, the US patent office included software as patentable art. Prior to 1980,
software development was a collaborative effort between colleagues in the computer
industry. With the new patent law in place, companies began developing proprietary
software that they could license to users for a fee. This proprietary software gave
companies a competitive advantage, but did not allow for public access to the source
code, thus eliminating the possibility for collaboration. In 1983, Richard Stallman, a
programmer at MIT, was not happy with the way software development was headed.
By 1984, Stallman began a mission to ensure that FOSS is available to anyone who
desired it.
“GNU” (a recursive algorithm for “GNU is not UNIX”) is a free and open source
operating system, first developed by Stallman in 1985. Since UNIX was a popular
proprietary operating system, the idea was to create a UNIX-compatible operating
system that would be made available to the public.
Stallman also founded the nonprofit Free Software Foundation in 1985, which
exists to promote the development of the GNU project and to protect the legal status of
free software (www.fsf.org). Although the foundation, in principle, is opposed to
software patents and copyrights, the foundation holds the copyrights for the GNU
operating system in an attempt to prevent it from becoming proprietary.
3
UNIX-based but also Microsoft-based systems such as the Firefox web browser
and openoffice.org, which competes with the Microsoft Office package.
A number of open source products, such as Linux and the Apache web server,
dominate their product categories. Linux running on corporate servers is one of the
most common uses of open-source software among businesses. According to a
September 2008 Gartner survey of 274 companies worldwide that either currently use
open source or are planning to do so in the next 12 months. About 52% of the
companies surveyed are already using open-source server software and another 23%
plan to use it within the next 12 months (King, 2008).
Over the last decade, numerous major corporations, including Hewlett Packard,
IBM, and Sun, have launched projects to develop and use OSS. Meanwhile, a number
of companies specializing in commercializing Linux, such as Red Hat and VA Linux,
have completed initial public offerings, and other open source-based companies such
as Cobalt Networks and Sendmail have received venture capital financing. Also,
commercial software developer like Oracle‟s bid acquire open-source database maker
mySQL was a clear sign of the profound changes commercial software giants are willing
to make as they adapt to the increasingly significant collaborative programming
philosophy (Shankland, 2006).
Free Software applications are first, second, and third-running products in terms of
market share in several markets, including web servers, server and desktop operating
systems, web browsers, e-mail, and other infrastructure applications.
Some of these market developments are easy to account for. For example,
software vendors support open source platforms because failure to do so would
significantly limit their market size. Another factor driving change in the software
industry‟s business model is that software applications fit into specific niches within
larger systems and cannot be readily substituted for one another. Software is different
from a sack of concrete or a reel of wire cable. Some vendors behave as if software
prices are set by the market, but you cannot buy a quantity of software from one vendor
and an equal amount from another vendor and expect them to be interchangeable—
something you could expect to do with cable or concrete.
4
Free Software in the Business Market
Lower cost of ownership: The lower cost ownership is one of the main reasons
for enterprises to adopt Free Software. The companies spend heavily on commercial
software products for licenses. OSS offers low license and maintenance costs, which
helps to reduce the technology cost for a company. Forrest Research found that 87
percent of their respondents have gained expected cost savings from using OSS
(Forrester Consulting, 2007). E*Trade is one of the companies that have used open
source technology to reduce their costs and they were able to save $13 million a year
through the use of open source applications (King, 2008).
5
started to move into new markets without proprietary rivals. For instance, a company
called Cloudera distributes a version of Hadoop, a program that helps firms process and
analyze the unprecedented volumes of data generated by large websites. The small
businesses are using Asterisk, which is an open source telephony engine and platform,
to reduce their communication cost (Rupley, 2009).
When companies are operating under a tight budget, more companies are looking
for open source technologies. Usually, open source is typically a tenth of the cost of
traditional enterprise software (Howells, 2009).
Better quality: More people are involved in reviewing OSS source code than
proprietary software. Issues such as security breaches are fixed and communicated
relatively fast to the user community. Therefore, OSS has better quality than
commercial software. Forrest Research showed that 92 percent of respondents found
the quality of OSS to exceed their quality expectation levels (Forrester Consulting,
2007). E*Trade‟s Thompson has found that their systems are more reliable under OSS.
On January 22, 2009, when the interest rate was changed by the government, 55,000
customers logged into their website at once. Not only did the site perform well under the
heavy load, it performed better than its competitors (Ebert, 2008).
6
reduces work time and provides predictably reliable results. E*Trade‟s Thompson found
that their engineers spend less time on contract negotiation and more time on the
technology when they use OSS.
Unlimited upgrades: Commercial software sometimes does not share its code,
and therefore, upgrades are not possible. Some commercial software companies might
provide limited upgrades for a specific time. On the other hand, upgrades can be
extended indefinitely.
Poor Code: From the business perspective, OSS may involve issues and risks
for Program Managers. Considering that the size of the OSS community is not
sufficiently large, the people who are involved in it may be unreliable. Thus, this
situation often leads to poor code development and fails to attract skilled workers to
work with the systems.
7
Complex legal restrictions: According to Koohgoli, the CEO of Protecode, the
issue is not about the use of open source, but with the unmanaged copyright and
licensing issue (Pearlman, 2009). Open source has complex legal restrictions that can
create copyright and patent compliance issues and corporate transaction challenges for
companies that rely heavily on customized software or that distribute software to
partners or customers. There are two categories of licenses. The first is an attribution-
type license. This acknowledges that the authorship of the software is included in some
manner, such as source code comments and help files. The second is the reciprocal-
type license, known as “copyleft”, which requires users who make modifications or
extractions of copyleft software keep the derivative products free as well (Free Software
Foundation). The idea is to ensure compliance with the applicable licensing
requirements.
Difficult initial adoption: Commercial software might be costly, but the systems
are easier to adopt. Meanwhile, OSS is associated with an indirect cost. This means
more salary and other labor costs could be wasted due to uncommon knowledge
required to use OSS. Organizations that do not have experience with OSS projects will
find it easier to get commercial software running. Moreover, if the previous project or
assignment was created using commercial software, moving to OSS could cause
inefficiencies within the whole business process (Rangaswami, 2008).
Business Implications
8
market segment, there are significant financial and strategic impacts on the entire
industry.
There are some who think that free software is easier to hack into because
everyone has access to the code. Likewise, the notion has been advanced that since
the source code is hidden, proprietary software is therefore inherently more secure. This
has proven false, since programmers can hack into proprietary software just as easily
as OSS. If hackers really want to get access to proprietary source code, there are also
software packages that can decode the source code from the binary executable files.
Though not 100 percent accurate, programmers can get a good idea as to what the
source code looks like. So, just having access to the source code does not make it
possible to hack into systems. Source code availability and security are independent.
On the other hand, FOSS developers claim that since a larger community has
access to the source code, there are more people available to look for vulnerabilities
than there are hackers who seek to exploit software vulnerabilities. In the case of
proprietary software, the community of hackers is likely larger than the potentially small
community of developers who have access to the source code. Although the absolute
number of developers versus the number of hackers is not a perfect metric for
measuring software security, it is nonetheless a factor to consider when evaluating the
probability of security breaches.
Some may think that proprietary software is more reliable since there are paid
professionals who are updating the code. This theory too is not entirely true since FOSS
is updated regularly by professionals. Just because FOSS programmers are not paid by
large corporations does not mean that the quality is any less. In addition, some FOSS
developers have a business model that includes updating and maintaining business
FOSS for a fee. Although the software is free and open source, a company can still
create a business model by maintaining the software for others, such as Red Hat and
the Free Software Foundation. One test by Bloor Research in 1999 (Boulanger
2005121) compared the reliability of a Windows NT server and a GNU/Linux server.
The test showed that Windows system had an uptime of 99.26%, while the FOSS
system had and uptime of 99.95%. This may not seem like too much difference, but
9
over the one-year test, the Windows system had 60 more hours of downtime. This test
shows that a FOSS system can be at least competitive with a proprietary system.
Another important factor that affects both security and reliability is the speed of
software updates and fixes. All software code will be written with errors, usually
measured in the number of errors per number of lines of code (usually defects per 100
lines of code. For example a common defect rate is about 1 percent. So, based on the
total number of lines of code, “Windows XP and Red Hat Linux would be estimated to
have approximately 40,000 and 30,000 undiscovered defects, respectively” (Boulanger
2008). When an FOSS release is made, all users serve as a test-bed to verify the
software. In contrast, a proprietary software release has a much smaller test capability,
since fewer people have access to the code, and there are limited test resources within
the firm. Since the FOSS community is much larger than proprietary software
developers, FOSS system flaws are typically patched much faster than proprietary
software vendors.
Another argument against proprietary software is that since the code is hidden, the
application may have additional functions that the user is not aware of. For example,
seemingly benign software could be designed to gather data from the user. The data
gathered may be personal or simply non-personal trend data. Regardless of the type of
data gathered, however, it should not be done without the user‟s knowledge. With OSS,
it is much easier for users to know the type of information that the software is gathering.
For individual and small business users, software cost of purchase is not the only
cost. Just as relevant is immediate usefulness of the software, the ability to produce
documents that can be read by a wide variety of recipients, and to do this without
incurring additional training costs. For instance, school systems from high school
through college teach Microsoft Office, and hiring people who are ready to be
productive using these applications is much easier than finding people who are
comfortable with OpenOffice, Lotus Symphony, or other free offerings. Training can be
10
arranged, but this involves extra effort and time. This is as true for individuals (who may
not have to be trained on MS-Office) as it is for businesses. Conversion tools for
converting OpenOffice documents and Office are readily available (some built into
OpenOffice, others feely available online), but users may prefer the convenience of
software that does not require any additional downloads, plug-ins, conversion tools, or
the extra learning time (however small) involved in mastering them.
The idea that the “free” software reduces software cost is due to a conflation of the
term “cost” with that of “price.” Just as consumers at the convenience store are acting
rationally when paying higher prices because they factor time and effort into their overall
cost calculation, software purchases must factor in convenience and time (including
training and search costs), into their cost model.
However, cost is not the only reason for the growing popularity of open source.
Open source software offers more flexibility than proprietary programs, the licenses for
which often include restrictions on how they can be used, and companies no longer
perceive free software as riskier. Getting sued for running programs that inadvertently
violate somebody else‟s intellectual property, for instance, has proven not to be as big
an issue as once feared (Anonymous, 2009).
Software adopters who do not want to invest resources of time and personnel into
a FOSS project may find that freedom (from initial software purchase outlays) is
outweighed by the convenience of using products in which training has already been
invested.
Economic Logic
11
blade sales stream, where frequent upgrades and software maintenance contracts are
the razor blades.
When Red Hat, MySQL AB, Pentaho, or any other open source-based vendor
gives away some or entire software portion of their product, they do so in the
expectation that potential customers attracted by the free software will then purchase
support contracts, consulting and training services, and other value-added services.
Making the software open tends to attract members of a community that were never
likely to purchase the software or related services, such as independent developers and
companies that are too small, for now, to afford enterprise-level solutions.
So, there are good reasons for much of the ongoing embrace of open source
solutions by enterprise software vendors, but we have yet to grapple with the problem of
why programmers would develop OSS in the first place. Clearly, it is time for a new
economic model. Programmers acting as individuals in a labor market could expect
some reasonable payment for the time and effort they expend on developing OSS. They
decline that remuneration, and that seems to indicate that they do not operate under the
free market model.
12
There are many reasons besides charity or the pure joy of creativity that prompt
people to write OSS. There‟s also a substantial and growing developer contingent
working on free or open source software that serves as the basis for a commercial
software product. OpenOffice development is sponsored by Sun Microsystems.
OpenOffice is free, but Sun rolls OpenOffice improvements into its commercial
StarOffice suite. MySQL is available either free or in a commercial version with added
configuration tools and other proprietary bells and whistles, and at least half a dozen
popular web content management and ecommerce packages also fall into the dual-
licensed, dual-branded category.
Trojan Horses?
Michael Tiemann (2008) observes with dismay that Microsoft‟s embrace of the
Open Source concept is actually a corruption of the concept, since its license to open
software for noncommercial use is antithetical to the generally agreed-upon definition of
Open Source, which clearly rules out restrictions against commercial use.
13
Closing Observations
So far, we have considered adoption of FOSS in the context of businesses that are
large and sophisticated enough to become active participants in the FOSS community,
but future research would do well to consider individual and small businesses software
users of software (such as database or desktop applications as well as larger
businesses that do not wish to participate in the ongoing development of software
platforms or sophisticated applications. Cost, for these users, may entail more than the
initial software price, and restrictive license agreements that prohibit source code
modification may have no relevance to their adoption decisions. These users have no
desire to modify the software they purchase, but instead may value other factors such
as ease of use and the ability to share documents with other users over the Internet.
Moreover, while lack of familiarity with low-cost software alternatives might cause
this class of smaller-scale users to spend money on brand-name software that has been
heavily promoted by expensive marketing campaigns, it is not at all obvious that FOSS
presents a more attractive or even cost-effective solution when their priorities are
considered. Indeed, familiarity with the use of FOSS as strategic weapons by
companies such as SUN could be additional reasons for them to doubt whether the aim
of corporate FOSS sponsors is to produce high quality software or whether the software
merely represents a short-term ruse to sabotage marketplace competitors.
14
References
Anonymous (2009), “Business Born Free; Open Source Software in the Recession,”
The Economist, 391:8633, May 30, pp. 69.
Asay, Matt (2008), “The Dying Embers of Microsoft‟s IP Claims against Open Source”
October 20, http://news.cnet.com/8301-13505_3-10070008-16.html.
Ebert, Christof (2008). “How Open Source Tools can Benefit Industry,” IEEE Computer
Society, 26:2, March/April, pp. 50-51.
Howells, Ian (2009). “The Top 5 FUD Myths Used against Open Source,”
ComputerWorldUK,
http://www.computerworlduk.com/community/blogs/index.cfm?blogId=16&entryId=2019,
March 23.
15
Mobity, Tony (2007). “So, why, why do People and Companies Develop Free
Software?” Free Software Magazine, Issue 20, November 7,
http://www.freesoftwaremagazine.com/articles/editorial_20.
Shankland, Stephen (2006). “Oracle Tried to Buy Open-source MySQL,” February 15,
http://www.builderau.com.au/news/soa/Oracle-tried-to-buy-open-source-
MySQL/0,339028227,339234504,00.htm,
Stallman, Richard (2007), “Why Open Source Misses the Point of Free Software,”
http://www.gnu.org/philosophy/open-source-misses-the-point.html.
Wichmann, Thorsten, and Glott, Ruediger (2002). “Free/Libre and Open Source
Software: Survey and Study,” FLOSS FINAL REPORT, Berlecon Research GmbH
Berlin, http://www.flossproject.org/report/FLOSS_Final4.pdf.
16