1 .edu
Strategic Computing and Communications Technology, Fall 2005
Professors Hal Varian and Michael Franklin
Non-Software Open Source Opportunities
Group F:
Albin Louit
Chitra Madhwacharyula
Koji Murao
Mark Pearson
Mike Wooldridge
1. Introduction
Open source software has been one of the most unexpected technological successes of the past decade. The philosophy of open source runs counter to the traditional, common-sense notions of what is required to run a successful software enterprise—that you must charge money for your product, and that you must use those funds to support a centralized team of specialists, including developers who write the code, quality assurance teams that make sure it works as planned, and sales teams that distribute the compiled final product. In contrast, open source software projects give away the results of their work, make source code available for inspection and modification, and typically do most of their development through far-flung networks of part-time volunteers. As open source projects such as Linux and Apache have shown, the software that emerges can be on par with some of the most popular closed-source software products.
Now, many of the concepts and strategies underpinning open source are spreading outside the software realm to non-software-based content. Such content is being distributed for free with licenses that allow for the reuse and redistribution. In this paper, we analyze a few examples of open source content development and distribution, including the free online encyclopedia Wikipedia,[1] open access scholarly journals, and open source educational materials. We present a tabular comparison of the pricing and business strategies adopted by different open source content distributors[2]. We also suggest ways such projects can generate enough income to sustain themselves and even produce a profit for their investors.
2. Examples
2.1 Open Source Encyclopedias
2.1.1 Background
Wikipedia is a free, multilingual encyclopedia written collaboratively by contributors around the world. It has become the “common man’s encyclopedia” that provides free read/write access to its pages to anyone in the world. The Wikimedia Foundation Inc.[3] is the parent organization of Wikipedia. It is a non-profit organization whose main goal is to encourage the further growth and development of open-content, wiki-based projects and to provide the full contents of those projects to the public free of charge.
2.1.2 License
The GNU Free Documentation License[4] used by Wikipedia grants free access to the encyclopedia content in the same sense as GNU Public License software is licensed freely. This principle is known as copyleft. Wikipedia content can be copied, modified, and redistributed so long as the new version grants the same freedoms to others and acknowledges the authors of the Wikipedia article used. Wikipedia articles therefore will remain free forever and can be used by anybody subject to certain restrictions, most of which serve to ensure that freedom.
2.1.3 Expenditure
Server hardware, maintenance, and bandwidth make up the largest costs of running the Wikipedia. Although most of its servers and bandwidth are donated by corporations such as Yahoo! and Sun Microsystems, the site’s growing popularity has meant that the need for servers and bandwidth increases by the day. Domain registration costs also contribute to the expenses, as do salaries for staff and occasional travel costs.
2.1.4 Income
As of January 2005, Wikimedia is funded in most part through private donations and gifts of servers and hosting from corporations such as Yahoo! According to Wikipedia founder Jimmy Wales,[5] the Wikipedia community has been extremely generous on occasions when the Wikipedia asks for donations on its Web site. Wales says grants from large philanthropic organizations are less common, since Wikipedia’s unorthodox business structure has made it a hard sell (although this is changing as Wikipedia’s popularity increases). Wikipedia sites are free from advertisements and visitors are not charged to edit, read, or use the content.
However, as the popularity and content of Wikipedia grows, various other means of funding are under consideration. A few of the important ones are applying for grants, selling WikiReaders[6] (collections of related Wikipedia articles), selling a print version of the Wikipedia, implementing a paid membership system, and running advertisements. The Wikimedia Web site also sells clothes, accessories, and other merchandise with logos that are trademarked by the Wikimedia Foundation.
2.1.5 Industry Response
Wikipedia's two main competitors are Encyclopedia Britannica and Encarta. Microsoft earns revenue from Encarta by selling CDs and by selling the right to computer manufacturers to pre-install it on computers. It is also freely available online with accompanying advertisements. In response to the growth of Wikipedia, Microsoft has now made all Encarta articles editable, attempting to exploit the same content generating effect. Meanwhile, Britannica generates revenue mostly through charging libraries and schools fees for electronic access, with a little additional revenue coming from book sales. Britannica has reacted to Wikipedia's threat by emphasizing its accuracy and quality control procedures. The encyclopedia market was estimated to be worth 300 million in 2003 by a former chief executive of Encyclopedia Britannica[7].
2.2 Open-Access Journals
2.2.1 Background
Open-access journals are open-source versions of scholarly research journals. Accessible via the Internet at no charge, they offer peer-reviewed articles from a variety of academic fields. One of the more popular open-access Web directories, the Directory of Open Access Journals,[8] lists more than 1,800 journals. Open-access journals are published by commercial (BioMedCentral and Internet Scientific Publications), non-profit as well as academic organizations.[9] Open-access journals compete with traditional scholarly journals that are available through subscription and are sold to large institutional libraries. The publishing market for research journals is worth $7 billion a year with the market leader, Reed Elsevier, taking in annual profits of $290 million.[10]
While their numbers are still small in comparison to subscription journals, two factors have helped open-access journals make inroads into the industry. The first is the increasing costs associated with the pay-based competition. A study conducted in 2001 showed that prices of for-profit journal content had tripled in the previous 15 years while prices for non-profit journal content had only risen by 50%.[11] Such increases have led to protests from library organizations and researchers who have argued traditional publishers are price gouging, and have been a source of positive PR for open-access journals.[12] Another factor is the apparent positive impact of free access on citation frequency. Recent studies have shown a positive correlation between the number of citations a paper receives and the probability that it is freely available online.[13] Given that citations are seen as a key measure of a paper’s merit, such correlations are making researchers think twice about publishing in journals that hide articles behind a subscription wall.
2.2.2 License
The definition of an open-access journal, from the Bethesda Statement on Open Access Publishing,[14] says that open-access authors must grant all users free access to journal content and allow for the redistribution and the creation of derivative works (subject to proper attribution). While journals differ in their licensing schemes, most of them follow this principle. Some open-access journals have different requirements for reuse depending on if the use is academic or commercial (i.e., some reuse may require written permission).
2.2.3 Expenditure
Generally, open-access journals operate in ways similar to those of traditional subscription journals, but on a more limited scale. According to a recent study of open-access journals, peer-review tends to be done more in-house instead of with outside reviewers, and copy-editing is performed less frequently. Even with their smaller staffs, most costs at open-access publishers are still related to people (editors and reviewers), and many rely on volunteers for some of their work. Other costs include computer hardware, software, and Internet service provider fees.
2.2.4 Income
To fund themselves, open-access journals rely on a mix of advertising, fees paid by authors (see section 3.2), and grants from industry and non-industry sources. In a 2005 study of open-access journals,[15] 35% expected to see a surplus over the latest fiscal year, 24% expected to break even, and 41% expected to realize a shortfall. (For subscription online journals, the study reported a better outlook: 74% expected to have a surplus, 3% expected to break even, and 22% expected a shortfall.) Interviews from the study suggest that open-access publishers are still experimenting with business models that will allow them to offer free content but still pay their bills.
2.2.5 Industry Response
A key effect that open-access journals have had in the industry is they’ve forced traditional publishers to experiment with new business models. One such model is to offer delayed open access, in which research articles are made freely available but only after a certain time. For example, Molecular Biology of the Cell allows free access to all their articles after a two-month delay.[16] Such a strategy is an example of versioning, i.e. marketing differently priced products to different audiences (similar to financial institutions charging extra money for real-time stock quotes). Traditional publishers appear to be seeking a middle ground where they are able to realize profits from content distribution but still offer the unrestricted access to information that has become expected in the Internet age.
2.3 Open Source Educational Materials
2.3.1 Background
In 2001, the Massachusetts Institute of Technology launched the OpenCourseWare initiative[17] and began placing digital copies of class lectures, handouts, and exams online, making them freely available to all. As of June 2005, the project features 1,100 lectures from more than 30 departments.
2.3.2 License
At M.I.T., these materials are available under a GPL-like license, allowing derivative non-commercial works provided the attribution and the license remains.
2.3.3 Expenditure
Since the course materials distributed on OpenCourseWare are already naturally generated by M.I.T. faculty (and hence require little additional time or effort on their part), most of the costs for the program stem from computers and hardware to distribute the content.
2.3.4 Income
M.I.T. accepts and encourages donations and recently received a million dollars from one M.I.T. alumnus. Most of the OpenCourseWare funding comes from M.I.T. itself; M.I.T. does not plan to generate revenue from this program. M.I.T. has strenuously proclaimed these materials will always remain free. It however remarks there are two items it is not providing: one, access to M.I.T.’s professors and two, a degree from M.I.T. In releasing these materials, it is implying that the value of the M.I.T. experience comes from contact with the people there and the degree/brand. In this way, M.I.T., while not making any money directly, is increasing the value of its other assets.
2.3.5 Industry Response
Providing such free access disrupts the current market for higher education, especially in the areas of (for-fee) distance learning and higher-education materials. This is a substantial industry as the U.S. online higher education market alone is estimated at $6 billion for 2005 (Eduventures’ report “Online Distance Education Market Update”).[18] While it appears traditional online distance learning programs haven’t yet responded, others have taken the OpenCourseWare content and modified it. For instance, consortiums of universities in China and universities in Spanish- and Portuguese-speaking countries have translated much of M.I.T.’s content into their local languages. Also, other universities such as Ohio State and Tufts have followed M.I.T.’s lead and started opening up their course materials to everyone, [19] thereby echoing M.I.T.’s statement about where the value in higher-education resides. Like Wikipedia, one can easily imagine these bundles of knowledge growing and improving over time in the Metcalfian sense, as professors base classes off of these class resources, improve the presentation, add sections on knowledge not previously taught, and contribute the resulting lectures and handouts back into the pool.
2.4 Other Open Source Initiatives
As the open-source movement increases in popularity, the idea of freely sharing other types of digital content—such as audio, images, video, and text in the form of blogs—has been gaining popularity. Examples of other content-based, open-source initiatives can be found at Creative Commons.[20] Creative Commons is a non-profit organization that publishes open-source-friendly copyright licenses for authors and artists, and also serves as a clearinghouse for the sharing of free content. The Internet Archive[21] is another popular place for sharing all these types of free content. Also, Flickr,[22] a popular image-sharing site, allows searches by the type of license under which the image is shared – i.e., one could search for all images of cats that allow derivate works. Other open-source initiatives include environmental knowledge bases provided by the government.
3. Business Strategies Using Open Source Content
How does one make money to support the distribution of free content? This section looks at a variety of strategies for generating revenue from open-source content projects.
3.1 Repurposing Free Content
Republishing free content in another medium can be an effective way to generate revenue. Westlaw[23] (from Thomson West) and LexisNexis[24] (from Elsevier) are fee-based online legal research systems that take public legal documents (including state and federal statutes and caselaw), catalog them into databases, and resell the databases (or access to them) to the legal profession. Meanwhile, WikiBooks[25] is a Wikimedia project that produces textbooks and reference books. A possible business opportunity is to put some or all of the content from Wikibook projects on CD-ROM or in a printed book and sell it.[26] In fact, this strategy is being considered by the Wikimedia foundation as a way of raising funds for its projects. The main distinction between these two examples of repurposing is the first uses free content not yet accessible online while the second uses free content already accessible online.
3.2 The Author Pays
One way of financing the free distribution of content is by transferring costs to the author rather than the reader. This strategy is already used by some open-access scholarly journals, which require that authors pay a fee to cover the costs of publication. (Such fees are typically paid out of the author’s research grants, not out of the author’s own pocket.) The Public Library of Science (an open-access publisher) charges authors $1,500 per accepted article to cover the costs of peer review, technical editing, and electronic distribution.[27] BioMedCentral charges range between $585-$1680 depending on the journal.[28] Some subscription-based journals allow authors to pay fees to make their articles openly accessible online.[29]
3.3 Advertising
Advertising has been one of the most popular methods by which free content distribution and services on the Web are supported. News sites such as and search engines such as Google are two prominent examples. Some open-source content providers, including open-access journals, also use advertising as a primary means of funding. While it currently does not run ads on the Wikipedia site, the Wikimedia Foundation recently announced it will begin providing Wikipedia content to , a word definition Web site. will run adds with the Wikipedia content, and the two companies will split ad revenues[30]. Open-source content sites that focus on niche content—as all of the examples in Section 2 do—have the potential to offer targeted advertising, a possibility made easier by automatic advertisement brokers like Google’s AdSense.
3.4 Complementary Goods & Services
To bring in revenue, a firm could provide services that are complementary to its open source content. For example, after the U.S. government releases free census data, firms can take the data and add useful complementary data to it, or sell software that helps others analyze the census information. AnalyGIS’s[31] product Geonetx helps businesses select retail locations based on census data; its Terrapin product augments census data with earth sciences information to support environmental impact reports. The open-access publisher BioMedCentral[32] provides complementary services such as Adsumo (a career search service for researchers), Open Repository (a digital-document storage service for research institutions), and Current Controlled Trials (a suite of services related to clinical trials for medical research).
Also, while consultancy is a common source of revenue for open-source software companies, we did not encounter any non-software open-source businesses taking advantage of this.
3.5 Merchandising
A firm can merchandise the open-source goods. Wikipedia is currently selling T-shirts, mugs, mouse pads, journals, bags, and other items with its logo printed on them through CafePress.[33] The Wikimedia Foundation, Wikipedia’s parent organization, earns 20% of the list price.
3.6 Grants/Government Financing
Organizations working for the public interest can receive grants from benefactors or funding from the government to provide their open-source content. Wikimedia received $94,648 from private and corporate benefactors during its fund raising session of the first quarter 2005.[34] Governments also provide funding for open-source projects. An example is the Global LCI Directory,[35] an environmental knowledge base available on the Environmental Protection Agency’s website. This knowledge base provides links to articles about Life Cycle Assessment (LCA), a specific recursive method to assess the environmental impact of various human activities. In LCA, scientists use previous LCA assessment to make new ones. Therefore, as more laboratories share their data on LCA, this method becomes more accurate.
3.7 What if you aren’t the original provider or distributor of the content?
Even if one is not the original provider of the content, most of the above business strategies still work (although most are most effective if they can be branded and marketed by the original distributor). Some, like the repurposing the content, depend on the free content being available under a license that allows commercial reuse. Others, like selling complementary products and services, can be used regardless of the license under which the original content is distributed. Finally, companies can incorporate the open source content into their own products to improve their product quality. For example, Yahoo! and Google embed Wikipedia excerpts directly on a search result page for certain queries; users no longer have to click on a link to the find the answer to the query.
4. Future Trends
The Open Source movement shows all signs of being the next biggest trend after the World Wide Web. With the increasing popularity of open-source content, organizations are left with no choice but to recognize its potential and try to find ways to use it to further their businesses. A few emerging trends that will potentially gain prominence in the future open source world are as follows:
4.1 Advertising
It would be interesting to see how long it will be before open source content sites like Wikipedia succumb to the lure of advertising. Advertising brokers like Google’s AdSense make it increasingly easy to automatically pair relevant advertisements with content. With the growth of popularity of open source freeware, these content sites will become prime targets for advertisers who would like to leverage the popularity of these sites to promote their products. And it may eventually become a symbiotic relationship by providing the open source organizations a steady stream of income.
Open source publishers who want to include advertising might even be able to use some of the key concepts of open source to their advantage. For instance, publishers could include viral advertising clauses in their licenses that state that ads on the original content must be included on repurposed content (similar to how GPL licenses require repurpose versions of free software to also be free). This could allow open-source projects to charge advertisers more based on the additional exposure that would be realized through future redistribution (although predicting the extent of redistribution might be a challenge).
While “nagware”[36]—i.e., interstitial advertisements that customers can avoid by paying a fee—is a strategy that some Web sites and software manufacturers use to make money, such a strategy runs counter to Wikipedia’s current practices and could threaten the organization’s positive brand image. Instituting nagware could also make it vulnerable to competitors who could simply repurpose the Wikipedia content without the nagware (or with less annoying advertisements), since the content is open source. Switching costs for users would be minimal.
4.2 Collective Intelligence Systems
Time magazine[37] in its October 2004 issue published a discussion on identifying the trends that are most likely to affect the future. According to the discussion participants, collective intelligence is the buzzword of the future. Considering how open source media like Wikipedia function, how [38] leverages the user annotation on its site and how photo-sharing sites like Flickr[39] are becoming a big threat to other similar applications, it is evident that an era of global knowledge sharing has begun. Organizations gain knowledge from their customers and leverage this knowledge to improve the quality of their products and services in turn to provide that knowledge to other customers, thus in effect creating a gigantic knowledge management system. This effect–driven by Metcalfe’s law–will help open content networks to provide more utility to the user than closed ones and many closed networks (such as Westlaw or the legal side of LexisNexis) will need to become more open or risk obsolescence.
5. Appendix
Comparison of Non-Software Open Source Projects
[pic]
Comparison of Non-Software Open Source Projects (cont’d)
[pic]
-----------------------
[1]
[2] See Appendix, pp. 9-10
[3]
[4]
[5] In answer to a question about fundraising from Group F’s Koji Murao, at a lecture at UC Berkeley SIMS on Oct. 3, 2005.
[6]
[7]
[8]
[9] (p.6)
[10]
[11]
[12]
[13]
[14]
[15]
[16] (p. 3)
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
-----------------------
Opportunity
GIS
Content Distributed under Creative
Commons Licenses
Legal search services
Description
A geographic information system (GIS): a
system for creating and managing spatial
data and associated attributes.
Digital contents license initiative.
Provides 11 different sets of licenses
based on 4 key concepts. This licensing
schema makes it easier to share digital
contents, described in the words “some
rights reserved”.
Cornell’s Legal Information
Institute provides access to
some legal rulings and filings,
but no comprehensive open
service exists yet.
Competitors
Traditional paper-base maps
Any business living on copyrights:
Movie studios, music publishers etc
Westlaw, LexisNexis
Fundings
Sale of complementary goods
Donations, Merchandize
University-funded
# of users
Depends on application
Initiated in 37 countries
Not available
# of data available
Depends on application
50 million objects linked to Creative
Commons
Not available
Author/Provider fee
None
None
None
Reader/consumer fee
Bundled data and software analysis tools
requires payment.
None (restricted to the type of license
on the object)
Free
Lock-ins
Yes: Currently no substitute
Yes: Users of contents will enjoy
freedom they could not have under “All
rights reserved” license schema.
Not prominent
Network effects
Some. There are various applications.
Very important. The more users and
providers there are, the more useful it
will become.
Some, when Creative
Commons licenses allow use of
the content in creating
derivative works.
Pricing strategies
Depends on application
None
Almost always none, but
depends on license.
Standards
Standardized data format
Varies for each object provided
Varies for type of content.
Opportunity
Open source encyclopedias
Open Access journals
Educational materials
Environmental databases
Description
Wikipedia: Free multilingual, Web-based,
free-content encyclopedia that anyone can
edit and view. It is written collaboratively
by volunteers with wiki software. The
project began on January 15, 2001 and is
now operated by the non-profit
Wikimedia Foundation
The Directory of Open Access Journal:
free, full text, quality controlled
scientific and scholarly journals
MIT OpenCourseWare: an
initiative of the MIT to put all
of the educational materials
from MIT's undergraduate-
and graduate-level courses
online, free and openly
available to anyone, anywhere,
by the year 2007.
LCI directory: a database on
EPA’s website that provides
links to existing articles on a
specific method (LCA)
according to specific research
criteria
Competitors
Britannica, Encarta etc
Traditional scholarly journals that are
available through subscription
Higher education institutions,
distance learning, higher
education materials
Lab-owned smaller
environmental databases
Fundings
Donations, Merchandize
A mix of advertising, fees paid by the
authors, and grants
University-funded, donations
Government’s for the website,
Grant/fees for the data
providers
# of users
915,000 visits per day. Operate under
189 languages worldwide. 535,378
registered users for the English page.
Spin-off projects in China,
Japan, Vietnam, Spain,
Portugal, France, 5 other
colleges and universities in
North America
Around 50 labs worldwide,
not growing much anymore
# of data available
787,000+ articles for English, 2,044,000
articles under 10 major languages
1,848 journals. 79,827 articles
In September 2004, 900 MIT
courses were available online.
Materials of all 1,800 courses
will be available by 2007.
Around 3,000 articles,
Growing steadily
Author/Provider fee
None
Sometimes
None
None
Reader/consumer fee
None
None
None
Around 25% of the data
require payments
Lock-ins
Search costs of users for as good
alternatives as Wikipedia are large. A
traditional encyclopedia is more
expensive than “free” encyclopedia.
The more journals are added, the more
the switching cost (search cost) will
Increase.
The M.I.T brand
Yes: when you start using the
methodology that this database
refers to (LCA), it is difficult to
switch to another method
Network effects
Very prominent
Very important: the more journals are
added, the more useful it will become.
Very important: the more
educational institutes base their
classes on the materials here,
and contribute back to the
pool, the more useful the site
will be.
Very important: As more
users uses LCA, the more
effective this method
becomes.
Pricing strategies
None for the data. Wikipedia is currently
selling merchandized goods with its logo.
Users cannot use the logo outside for
business use.
It depends on each journal.
None
As more articles become free,
the LCA method becomes
more efficient, which benefits
all laboratories using it.
Standards
Articles are posted in the standard web
environment. Users do not have to learn
any language.
All the journals must be available on
the web.
Course materials are provided
in the PDF format
Varies for each article
provided
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.