Felix Stalder on Thu, 14 Aug 2003 18:52:27 +0200 (CEST) |
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
<nettime> Six Limitations to the Current Open Source Development Methodology |
Six Limitations to the Current Open Source Development Methodology The "Open Source Approach" to develop informational goods has been spectacularly successful, particularly in the area for which it was developed, software. Also beyond software, there are important, successfull Open Source projects such as the free Encyclopedia, Wikipedia; collaborative sites writing/publishing projects such as koro5hin.org; and the Distributed Proofreading Project, attached to the Gutenberg Project. However, particularly outside the software domain, the Open Source projects remain relatively marginal. Why? Some of it can be explained by the relative newness of the approach. It takes time for new ideas to take hold and to be transferred successfully from one context to another. But this is only part of the story. The other part is that the current development model is based on a number of specific, yet unacknowledged conditions that limit its applicability to more diverse contexts, say the music distribution or drug research. The boundaries to the open production model as it has been established in the last decade are set by six conditions characterizing virtually all of the success stories of what Benkler called "commons-based peer production." The following list is a conceptual abstraction, a kind of ideal-type. The actual configuration and relative importance of each condition varies from project to project, but taken together they indicate the boundaries of the current model. In this elaboration, I draw from examples of free and open source software, but it would be simple to illustrate these limitation based on open content projects. 1) Producers are not sellers The majority professional, i.e. highly-skilled, programmers do not draw their economic livelihood from directly selling the code they write. Many work for organizations that use software but do not sell it, for example as system administrators. For them the efficient solution of particular problems is of interest, and if that solution can be found and maintained by collaborating with others, the sharing of code is not an issue. For others employed in private sector companies, for example at IBM, the development of free software is the basis for selling services based on that code. The fact that some people can use that code without purchasing the services is more than off-set by being able to base the service on the collective creativity of the developer community at large. From IBM's point of view, the costs of participating in open software development can be regarded as 'capital investment' necessary for the selling of the resulting product: services. For members of academia (faculty and students) writing code, but not selling (often explicitly prohibited), contributes to their professional goals, be it as part of their education, be it as part of their professional reputation-building. For them, sharing of code is not only part of their professional advancement, but an integral part of the professional culture that sustains them also economically,. in form of salaries for the faculty and stipends for the (graduate) students. Last but not least are all those who use their professional skills outside the professional setting, for example at home on evenings and weekends. Having already secured their financial stability, they can now pursue other interests using the same skill set. 2) Limited capital investment Particularly the last, and very important group of people, whose who work outside the institutional framework on projects based on their own idiosyncratic interests, can only exist due to the fact that the means of production are extraordinarily inexpensive and accessible. Materially, all that is needed is a standard computer (often even a substandard one would already suffice) and a fast, reliable connection to the communication forums of the community. Of course, the computer and the network rely on a level of infrastructure that cannot be taken for granted in large parts of the world, but for most people in the centers of development, they are within relatively easy reach. Once this access to be means of communication is secured, the skills necessary to participate in the development of code can also be acquired collaboratively, free of charge. The number of self-taught programmers is significant. Since no expensive diplomas are necessary to become active, the financial hurdle is, indeed, extraordinarily low. 3) High number of potential contributors Programming knowledge is becoming relatively common knowledge, no longer restricted to an engineering elite, but widely distributed throughout society. Of course, truly great programmers are rare as truly great artists are, but average professional knowledge is widely available. This has a quantitative and a qualitative dimensions. Quantitatively, the number of able programmers is in the millions, and rising. Qualitatively, the range of people capable programmers is also unusually wide, not the least because the material hurdles are so low and the learning can take place outside of institutions with entry exams and tuition fees. This large and diversified pool of talents makes it possible to create the critical mass of contributors out of only a fraction of population. 4) Modularized Production A large software program consists of many smaller code segments (libraries, plug-ins etc.)This makes it possible to break down the production process into many small steps which can be carried out by distributed contributors. If the act of integration is relatively straight forward, it allows to keep the amount of work that each has to contribute highly flexible and also make use of smaller contributions (bug reports, patches). Furthermore, the modularity of the production process allows a high number of people to work in parallel without creating significant interferences. 5) Producers Are Users According to Eric S. Raymond, a good open source projects starts with a programmer scratching his own itch and finding out in the process that there are many others with the same problem. Wanting to use a program is a great motivation of contributing to developing it. Often, it's much more efficient that waiting, hoping that someone will write and sell a program that will address one's particular need. 6) No Liability Last, but not least, software has no product liability. Paragraph 11 of the GPL states, similar to most other licenses, that "the copyright holders and/or other parties provide the program 'as is' without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose" (GPL, v2). The absence of liability makes it possible to produce a program without having to assign clear ownership, or other markers allowing to determine liability, to it. The space delimited by these condition is large and still not fully explored. We can expect that the current open production model will find additional niches in which it can thrive. Few could have predicted the success of Wikipedia only three years ago, even though Open Source Software had already been very successful at the time. However, it is also clear that many information goods fall outside of this space. Not always are the means of production inexpensive and readily available or the production process modular. Sometimes, the number of potential producers is small, more often than not are the producers not the users of their own products, and, in many cases, product liability is desirable. This does not mean that the "open source model" cannot apply to, say, the production of literary works, music, or medical drugs. What it means, however, is that to make it viable, another round of social innovation is required. This is slowly happening. The growth of "Open Access Journals" or discussions around "compulsory licensing" are good, though very early examples. ----+-------+---------+--- http://felix.openflows.org ------------------------------------------------------- -- ----+-------+---------+--- http://felix.openflows.org # distributed via <nettime>: no commercial use without permission # <nettime> is a moderated mailing list for net criticism, # collaborative text filtering and cultural politics of the nets # more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body # archive: http://www.nettime.org contact: nettime@bbs.thing.net