Article Preview
TopIntroduction
Over the last 10 years, open source software development has increasingly attracted the attention of scholars in the fields of economics, management, and social sciences in general (for sociological contributions, see Himanen, Torvalds, & Castells, 2001; Weber, 2004; see Maurer & Scotchmer, 2006, for an account of the phenomenon from the economist’s perspective). Although the significance of the software industry in modern economic systems can partially explain the increasing number of research contributions in this area, it is clear that the chief reason behind this growing interest is the fact that open source software development seems to represent a form of innovation process that challenges many facets of the current conventional wisdom concerning the generation of innovations in market economies (Lerner & Tirole, 2001).
Traditionally, economists have considered technological knowledge as a public good, that is, a good endowed with two fundamental features: (a) nonrivalry and (b) nonexcludability. Nonrivalry states that when one actor consumes or uses the good, this does not prevent other actors from consuming or using it. Obviously, this does not hold for standard economic goods: If Paul eats the apple, it is clear that Nathan cannot eat the same apple. On the other hand, both Paul and Nathan can breathe the fresh air of the park. Nonexcludability refers to the fact that when technological knowledge is in the public domain, it is no longer possible to prevent other actors from using it. Again, while Paul may force Nathan to pay for the apple, he cannot (legally) prevent Nathan from breathing the fresh air of the park. The traditional economist’s viewpoint contends that market economies are characterized by a systematic underprovision of public goods as their production is, due to the two properties described above, not profitable for private firms. In these circumstances, the standard prescription is that governments should intervene, using tax revenues to supply directly the appropriate quantity of public goods. This reasoning is at the heart of the argument that is commonly used in making the case for the public support of scientific research (Nelson, 1959). It is worth noting that, historically, the allocation of public resources for the production of scientific knowledge has been organized around a rather particular institutional arrangement (“open science”) capable of producing both incentives to create new knowledge and the public disclosure of scientific finding (Dasgupta & David, 1994).
Public funding, however, is not the only answer. Another solution put forward by the literature is based on the idea of inducing private firms to invest in the production of technological knowledge by means of an artificial system of property rights (Arrow, 1962). The most common example, in this respect, is the patent system. A patent assigns temporarily to its inventor the complete control of the new technological knowledge discovered. The rationale for this institutional device is straightforward: The prospect of the commercial exploitation of this temporary monopoly right will induce private firms to invest resources in inventive activities, that is, in the production of new technological knowledge.
In this context, open source software represents a case of the production of new technological knowledge (high-quality computer programs) carried out by individuals without any direct attempt of “appropriating” the related economic returns. Clearly, all this is at odds with the conventional wisdom summarized above.