Darwin at play on Sourceforge
Robert Galoppini and colleagues set out to get a good estimate of how many active, stable open-source projects there are. While Sourceforge shows well over 100,000 projects, and other source-code repositories add tens of thousands more, the number of projects that are actively being developed globally is actually a relatively small number:
18,000.
Should we be surprised that there are so few? No. Really we should be surprised that there are so many.
While Goloppini and colleages found that only 6.8% of Sourceforge projects have seen a reasonable level of activity in the past six months, the real question is those 6.8% of total projects. The 80/20 rule applies to open source as much as anything else: just as a few developers do most of the core development work on any given project, it makes sense that a relatively small percentage (~10%) of any given open-source projects will succeed in acquiring a population of active developers.
No matter how big the open-source development community grows, it will always be somewhat limited, just as the proprietary-software world is. How many proprietary projects succeed in attracting development and financial resources? 10%, if that. The proprietary world is subject to capitalistic darwinism, just as open source is.
The big difference, of course, is that with open source far less money is wasted on the path to discovery of whether you're hot...or not. This is true if you're a customer looking to buy software/services, for whom open-source software dramatically lowers the cost of failure. It's equally true if you're building software and want to determine your market opportunity.
If you can't attract downloads and developers, you're better off contributing to an existing project. In fact, I would bet that many of the moribund projects on Sourceforge were originally staffed by developers who eventually joined other projects.
Open source separates the wheat from the chaff. What is most shocking is that there are 18,000 "wheat" projects out there for developers to adopt and adapt. That software is worth billions, but costs $0.00.
Matt Asay brings a decade of in-the-trenches open-source business and legal experience to The Open Road, with an emphasis on emerging open-source business strategies and opportunities. Matt is vice president of business development at Alfresco, a company that develops open-source software for content management. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can follow Matt on Twitter @mjasay. 
There is one more interesting paper with results from empirical research on contribution to open source software. "An Empirical Investigation of Code Contribution, Communication Participation, and Release Strategy in Open Source Software Development: A Conditional Hazard Model Approach" By Param Vir Singh, Ming Fan, Yong Tan. You can download it here:
http://opensource.mit.edu/papers/singh_fan_tan.pdf
Some interesting facts from it:
- averagely 81% of code is committed by the top 20% of the most active developers
- Almost the same group of the developers is responsible for 80% of communication
Mark
http://www.sourcekibitzer.org/Bio.ext?sp=l8
thank you very much for the citation! I must tell you that
despite I recently started a joint research activity with the FLOSSMETRICS project, the post you mentioned was written solely by Carlo Daffara, that I am honored to guest time by time on my personal blog. Stay tuned, lots more results to come!
FLOSSMETRICS website: http://flossmetrics.org