Home

Project outline

Final report

Press coverage

related research

people

 

Papers

How are Open Source Software Projects Organized and What Makes Them Successful?

Katherine Stewart, University of Maryland
kstewart@rhsmith.umd.edu

My research on open source software (OSS) has been concerned with trying to understand how OSS projects are organized and what factors facilitate (or inhibit) their success. I approach these questions from a social-psychological perspective.

There are many factors that might be considered in defining “success” in the context of OSS development. For example, attention has been focused on market penetration of the software, levels of participation of developers, and software quality (cf. Feller and Fitzgerald 2002; Halloran & Scherlis 2002; Weiss 2002). While some researchers have focused on technical and economic issues underlying the success of individual projects or the phenomenon as a whole, many have also recognized the importance of social coordination and control mechanisms in determining outcomes in OSS projects:

“[open source movements are] loosely coupled communities kept together by strong common values” (Ljungberg 2000: 208)

“[open source development is] a socio-technical process that entails the development of constructive social relationships, informally negotiated social agreements, and a commitment to participate through the sustained contribution of software discourse and shared representations.” (Scacchi, 2002: 37)

Focusing on the importance of social factors to OSS, there are two questions my colleagues and I have begun investigating. The first is concerned with modeling the organizational architecture of OSS projects and the second is concerned with empirically testing the effects of social coordination mechanisms on project outcomes. A brief description of each follows.

1. How should open source projects be viewed in light of the literature on organization theory? That is, do these groups represent a new organizational architecture and if they do, what are its distinguishing characteristics, and under what conditions does organizing work in this form lead to greater or lesser success? Our preliminary analysis (Seidel and Stewart, working paper) leads us to believe that important characteristics of OSS organization are (1) fluid, informal boundaries of membership, (2) dependence on volunteer labor, (3) a strong community culture (4) a focus on commercially viable information-based products (5) open sharing of organizational knowledge, and (6) inexpensive and efficient communication. Many of these characteristics are not unique to OSS development (e.g., see Von Hippel’s (2001) discussion of communities of sports enthusiasts). As communication and coordination costs have dropped with the diffusion of the Internet and other communication technologies, similar kinds of organizations have emerged in other domains – for example A Bell Tolls is a community of consumers producing long distance rate information for the domestic U.S. market (Steele 2001). We believe that investigating the organizational structure of OSS projects may lead to benefits not only in understanding the OSS phenomenon, but also in understanding when OSS-like organizational architectures might be fruitfully utilized in other domains.

2. While many have noted the importance of social coordination and control in OSS, a thorough understanding of what social mechanisms are important and how they influence success has yet to be developed. A second question we have begun investigating is: how do ideology, interpersonal trust, and different kinds of communication influence the effectiveness of OSS workgroups? Some preliminary data collection and analysis (Stewart & Gosain, working paper) lead us to believe that commitment to commonly recognized OSS ideals (e.g., as articulated in Raymond, 2000 and DiBona, Ockman, and Stone, 1999) and high levels of social as well as task communication among group members enhance affective and cognitive trust within OSS workgroups. Such trust is associated with increased group efficacy and higher levels of activity on the project. If these conclusions hold, they may be used to help foster more successful projects both within the OSS domain and possibly in other domains that can utilize similar organizational structures.

REFERENCES

DiBona, C., Ockman, S., and Stone, M. 1999. Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O'Reilly.

Halloran & Scherlis 2002. “High Quality and Open Source Software Practices,” <http://opensource.ucc.ie/icse2002/HalloranScherlis.pdf> Accessed 10/2/02.

Feller, J. and B. Fitzgerald. 2002. Understanding Open Source Software Development. London: Addison-Wesley.

Ljungberg, J. 2000. “Open source movements as a model for organizing,” European Journal of Information Systems (9), pp. 208-216.

Raymond, E. 2000 “Homesteading the Noosphere.” <http://www.tuxedo.org/~esr/writings/homesteading/homesteading.txt> Accessed 8/1/02.

Scacchi, W. 2002. “Understanding the requirements for developing open source software systems,” IEE Proceedings on Software (149), pp. 24-39.

Seidel, M.D. L. and Stewart, K. J. "An Initial Description of the C-Form Organization," working paper.

Steele, Jeffrey. 2001. “Are you well-connected?” Chicago Tribune, <http://www.chicagotribune.com/business/chi-0108210195aug21.story> Accessed 8/21/01.

Stewart, K. J. and Gosain, S. “Impacts of Social Coordination Mechanisms on Effectiveness in Open Source Development Groups” working paper.

Von Hippel, Eric. 2001. “Innovation by User Communities: Learning from Open Source Software,” Sloan Management Review, Summer, p. 82 – 86.

Weiss, G. 2002. “The state of enterprise Linux,” <http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2879104,00.html> Accessed 10/2/02.