A Multi-agents Model of the Open Source Development Process

By:
Dr. Lik Mui,
Dr. Mojdeh Mohtashemi,
Dr. Sameer Verma
To add a paper, Login.

The open source development efforts of the past decade have brought much benefit for the users of such free software as Apache, GNOME, GNU/Linux, JBoss, among others. Many questions have been raised about the motivation behind contributions of developers around the world to create useful software and then freely give it away Traditional rational agency theory predicts that rational, self-interested individuals are not likely to volunteer contributing to the common good (Olson, 1965; Hardin, 1982). In contrast, the voluntary actions of the open source developers seem altruistic and defy our intuitive notion of how a market-driven, capitalistic society works. Von Hippel and von Krogh (2003), Lerner and Tirole (2000) among others have documented field observations on possible motives for why top developers around the world would volunteer to participate in open source projects. In this paper, we develop a multi-agents model of open source development with a focus on how reputation and social groups influence the emergence of volunteerism for open source developers. We analytically derive conditions for how voluntary actions among the developers would emerge to dominate over overt self-interested concerns. We also report simulation results which correlate well with field observations about voluntary actions in the open source world. For the management community, results of this paper illustrate how such implicit and indirect incentives as reputation gain and group identification can have powerful influence on motivating voluntary actions in software development.


Keywords: Multi-agents System, Open Source Development, Reputation, Social Groups, Volunteerism, Software Development
Stream: Technology in Community
Presentation Type: Paper Presentation in English
Paper: Multi-agents Model of the Open Source Development Process, A


Dr. Lik Mui

Senior Science Developer, Stanford Functional Genomics Facility, Stanford University
USA


Dr. Mojdeh Mohtashemi

Mitre Corporation


Dr. Sameer Verma

California State University, San Francisco


Ref: T05P0226