Weber, S. (2004).The Success of Open Source. Harvard University Press. ISBN: 0674012925 (COinS)

By Eric Lease Morgan

Introduction

The Success of Open Source by Steven Weber details the history, process, motivations, and possible long-term effects of open source software (OSS). This scholarly yet easy-to-read, well-written, and provocative book is worth the time of anyone who wants to understand how open source software is affecting information technology. Using Linux as its primary example, the book describes how the process of open source software may affect business & economics, methods of governance, and concepts of intellectual property. It is also a great read for those of us librarians who desire to play a role in the building of “next generation” library catalogs and other library-related information systems.My acquisition of the book was rather embarrassing, and at the same time typical. As the leader of an open source software project called MyLibrary, I asked some of my fellow hackers and open source software aficionados for advice on how to promote MyLibrary and build a larger community around it. One of the suggestions was to read Weber’s book. Like most people, I searched Google for the title, and Google returned it as the first hit. I was able to read the entire text online, if I desired. I didn’t. The librarian in me then went to WorldCat where I learned the book was located down the street in another academic library. I could borrow the book for free. All I had to do was visit the library and check it out. I didn’t. Instead, I looked the book up on Amazon.com and found a “new” copy from an Amazon.com Associate. Twelve dollars and four days later my book arrived. Easy. Convenient. Cheap. Food for thought.The book can be divided into four overarching topics:

1. the history of open source,
2. the process of open source software development
3. business models and open source software’s relationship to the idea of a “commons”, and finally,
4. a summary as well as a look to the future describing how the process of open source software might affect other human endeavors.

History of OSS

Weber traces the history of open source software from its roots in AT&T UNIX and the Berkeley Software Distribution (BSD) to the present day Linux operating systems. Weaved throughout is the development of networking technologies, specifically the Internet. The history brings to light two very influential computing philosophies: the “Unix Way” and software as something to be bought and sold.The first computing philosophy is the “Unix Way”, an outline of three engineering principles for making good software:

1. write programs that do one thing and do it well,
2. write programs that work well together, and
3. write programs that handle text streams because that is a universal interface.

Software that adheres to these principles is typically considered more useful than software trying to be all things to all people. Such software is modular, portable, easy to create and maintain, and can be applied in any number of settings.The second philosophy revolves around ideas of intellectual property. Suppose I spend time and creative energy writing a piece of software. Through this expenditure I have the right to sell the software to other people in order to gain compensation for my efforts. Software, like other goods, can be exchanged for other things of value, namely money. Moreover, if people copy my software and give it other others, then such a process is just like stealing from me since I am not being compensated for my efforts. This was the attitude of Bill Gates as he stated it in his “open letter to hobbyists” as they distributed his implementation of the BASIC programming language in order to run programs they had written against it. This perspective regarding software as something to be bought and sold ultimately led to the creation of Microsoft. At the time of Gates’ writing of the 1976 “letter” computers were bought and sold. Software sort of just came along for the ride.While it is important to note that I, the author of this review, am certainly an advocate for open source software, the book takes no sides one way or another. Instead, throughout the book, Steven Weber tows a middle ground by simply asking questions and then does his best to answer them as objectively as possible. The book neither advocates nor condemns open source software. It simply observes the environment and makes generalizations accordingly.

OSS as a process not a thing

The chapters I found most interesting described the open source software process, how it works, and the motivations of its participants. I learned from the book that open source software is more about a particular process and less about a thing or product. Consider two statements a priori:

1. software can be copied an infinite number of times and not denigrate the original version, and
2. globally networked computers allow the tiniest numbers of like-minded individuals to find each other easily.

Given such an environment the open source software process flourishes, and that process is outlined here:

1. Someone has a computing problem they want to solve — an itch to scratch.
2. The person builds on the good work of others and writes a computer program.
3. The person “freely” shares their software with others under some sort of license agreement.
4. A community forms along with norms of behavior and guidelines (governance) for contributing back to the solution.
5. The software grows and matures, hopefully
6. Go to Step #1 until the software is “done” or until someone else wants to take on the leadership role.

The resulting open source software application, the book points out, is not necessarily better (or worse) than “closed source” software. Instead, it is simply different. It is more easily modified. It is vetted through the eyes of end-users — a set of self-described pragmatists wanting to scratch their own itch. Furthermore, since the process easily accommodates the philosophy of letting a thousand flowers bloom, the resulting software is not necessarily designed for the mass market: The software is not aimed at a lowest common denominator.What motivates open source software participants? Weber shies away from the altruistic motives espoused in Eric Raymond’s The Cathedral and the Bazaar. Instead, Weber imagines that the motivations stem from a desire for artistry & craftsmanship, the desire to create better software, and the desire for recognition from peers (“ego-boosting”). Many of the people who participate in open source software seem to enjoy puzzle solving. Some of them like reading and writing beautiful code — software as poetry. They are engineers looking to build better solutions to known problems, to “build a better mouse trap”. Like scholars and researchers in academia, peer-review is an important aspect of the work, and if you write something that is used (cited) by many people, then in the eyes of others your value is increased.

OSS and business

The sections and chapters regarding open source software and business are probably the most significant aspects of the book. They cover issues that are the least understood by the wider community including the definition of “free”, the concept of property and the “commons”, the differences between the BSD and GPL licenses, how these differences effect business opportunities, and business models in an environment where the primary thing to be exchanged for value is bound by rights of distribution as opposed to exclusion.In our increasingly commercial society, the concept of “free” software is something most people find confusing. Weber compares and contrasts “free” from the point of view of the Free Software Foundation (led by Richard Stallman) and the point of view of dot-com companies such as Red Hat (a commercial distributor of open source software). In both cases the concept of “free” should be equated with the word “liberty” as opposed to “gratis”. However, the FSF approach has a moral slant while the Red Hat approach emphasizes practicality and the ability to easily improve software solutions.The concept of property plays a big role in business models. How can you sell something that is “free”? How can you earn money on a thing that is a part of the community commons? Who owns this intellectual property? These questions can be answered, according to Weber, by interpreting the BSD and GNU (or General) Public License. In both cased the software is give away gratis. The differences lie in redistribution. Under the GPL license, new software based on GPL-licensed software must be re-distributed under the GPL or a less restrictive license. The BSD license does not have this stipulation; new software based on BSD licensed software does not have to be redistributed for “free”. Mac OSX is an example of BSD-based software which has been commercialized: Apple , Inc. started with BSD Unix, enhanced it, then redistributed their modified version for a fee. Weber outlines a number of possible business models for open source software:

1. support sellers,
2. loss leaders,
3. “sell it, free it”,
4. accessorizing,
5. service enablers, and
6. branding.

For each of these models Weber points to a number of examples.

The book’s summary

The last two pages of the book summarize much of Weber’s observations, and listing them here feels a bit like spoiling the ending of a mystery novel. The effective open source software process/project needs to take into account and support:

• disaggregated contributions,
• the need for a critical mass of users,
• peer review,
• the positive effects of the Internet,
• belief that a small group can generate something truly useful, and
• a voluntary community.

Weber also outlines the characteristics of effective agents (open source software participants). They:

• can judge the viability of the product,
• can make an informed bet their contributions will be used,
• are driven by things beyond simple economic gain,
• gain personal knowledge through the process, and
• hold a positive ethical valence towards the process.

OSS and libraries

Naturally, I read the book through my rose-colored glasses of librarianship. After reading it and combining it with more recent personal experiences, I am now less of a “believer” in open source software. I take away a more realistic perspective on the definition of open source, its process, and what motivates its participants. This does not in any way diminish my belief that the open source software process can benefit the library community and therefore library users.Throughout the book I kept comparing the kernel of the Linux operating system to the integrated library system (ILS) of libraries. Both provide fundamental interface functions between two entities. In the case of an operating system kernel, the interface is between hardware and people. In the case of an ILS the interface is between a library and patrons.Many library open source projects already exist. The extent they meet with continued success and wider adoption, I predict, will be measured by the extent that they can accomplish the following things. First, an easy-to-understand vision statement needs to be outlined by one or more people who possess leadership qualities. Second, those leaders need to amass the resources required to make their vision a reality. Third, they need to put their vision into practice allowing as many people to participate as possible. Fourth, start small and work up. Encourage the building and re-use of existing core applications, such as databases, indexers, editors, and server platforms. Make sure the applications are modular and standards-compliant. Practice the Unix Way. Make it work first, then improve things. Don’t even attempt to create the perfect system the first time. The process won’t be quick. The process won’t be easy. The process won’t be “free”. On the other hand, the process will empower and enable the profession. It will give it increased choice and opportunity. Weber’s book, the Success of Open Source, can be used as a set of guidelines – a description of a framework – for building software solutions for the computing problems facing libraries.

About the Author

Eric Lease Morgan <emorgan@nd.edu> is the Head of the Digital Access and Information Architecture Department at the University Libraries of Notre Dame. He considers himself to be a librarian first and a computer user second. He and his fellow teammates have primary responsibilities for the University library’s website, the campus-wide search engine, and a number of digital library projects. He is also on the editorial committee of the Code4Lib Journal. In his spare time he can be seen folding defective floppy disks into intricate origami flora and fauna.