Monetary donations to an open source software platform
Citation: Sandeep Krishnamurthy, Arvind K. Tripathi (2009) Monetary donations to an open source software platform. Research Policy (RSS)
DOI (original publisher): 10.1016/j.respol.2008.11.004
Semantic Scholar (metadata): 10.1016/j.respol.2008.11.004
Sci-Hub (fulltext): 10.1016/j.respol.2008.11.004
Internet Archive Scholar (search for fulltext): Monetary donations to an open source software platform
Tagged: Economics
(RSS) open source (RSS), donations (RSS), Open Source Software Community (RSS), Economics (RSS)
Summary
Krishnamurthy and Tripathi's article is an analysis of donations to the company which runs SourceForge (SF) as support for the platform. It is the donations to the platform itself, and not to the individual projects, that the authors seems to have data for (although donations to individual projects within their dataset happen too). The authors suggest that the question of how these projects can be sustained is of importance.
The paper is broken into two studies. In the first study, the authors look at what predicts donations to SF. The authors build a dataset of the top 15% of SF projects by activity. They then build a dataset of people listed as members or contributed to those projects. They use this sample of users in a logistic regression that models the likelihood of a user donating to support SF itself based on a number of factors and testing four hypotheses. These hypotheses include:
- H1: Subscribers are more likely to donate. Supported
- H2: Individuals who have a longer tenure will be more likely to donate. Not supported
- H3: Individuals who donate to projects are more likely to donate to the platform. Supported
- H4: Individuals who accept donations will be more likely to donate. Supported
The authors do note seem to fully define what a subscriber is and argue that tenure is an operationalization of identity because users with longer tenure will have a stronger identity as a SF member and that accepting donations is a measure of reciprocation.
The paper also includes short qualitative study that analyzes comments left by donors on the SF website which is presented in an appendix.
In the second study, the authors reduce their sample to include only the users who have donated to the platform and instead try to ask how much users have donated. Because this information is not perfectly visible, the authors are able to extract estimates from badges displayed on the site. They offer an additional four hypotheses:
- H5: Subscribers are will donate more. Supported
- H6: Individuals who have a longer tenure will donate more. Supported
- H7: Individuals who are associated with active projects will donate more to the platform. Not supported
- H8: Individuals who accept donations will donate more. Not supported
Theoretical and Practical Relevance
Although the question of financial support FLOSS projects is of wide interest, the question of supporting the platform seems of more questionable and less broad relevance. Not all FLOSS projects use these platforms and there only a very small number of these. Additionally, these platforms have additional sources of income and may not be dependent on these donations to any degree.
An interesting framing explored in the paper is around the idea that SF is a for-profit venture. A short survey the authors present suggests that most users who responded (only about 1/7 of those surveyed) knew that SF was a for-profit and those that didn't would probably have donated anyway. However, this articles does not follow this fact up in any depth.
The fact that the donations were, in some form, public, seems to complicate questions about the mechanisms.
