Carnegie Mellon University
Browse

Determinants of Open Source Software License Choice: A Social Influence Perspective

Download (239.33 kB)
journal contribution
posted on 2007-12-01, 00:00 authored by Param SinghParam Singh, Corey Phelps

This study presents a social influence model of open source license choice. For our theoretical foundation, we build on the heterogeneous diffusion model of social influence from the sociology literature. We characterize specific open source licenses as discrete practices that are at risk of being adopted by new open source projects. Specific hypotheses are proposed and tested on a sample of 5,307 open source projects hosted at SourceForge. Our findings suggest the most important factor that determines which license will be adopted by a new project is the type of license chosen by licensors more socially proximate in the social network of a new project’s licensor. Moreover, the likelihood that a new open source project adopts a particular license increases when additional similar open source projects have previously adopted such a license and when these projects are large and successful. We also find that project managers with longer tenure in open source environments or who have been members of successful open source projects are less susceptible to social influence. This suggests that project managers who are inexperienced with open source software development or who have contributed to unsuccessful open source projects are most influenced by their social context.

History

Date

2007-12-01