Monday, March 1, 2010

Predicting FOSS Fail

The Open Source Way is a great resource for developers who want to start their own open source project. It is a wiki of lessons learned and best practices gathered primarily from experiences working on the RedHat and Fedora projects. One page that I find particularly interesting is "How to tell if a FLOSS project is doomed to FAIL." The format assigns "fail points" for common (and uncommonly stupid) mistakes that project leads make. Making a few of these mistakes doesn't doom your project but making lots of them definitely stacks the odds against you.

While it was written from the perspective of a Linux library developer, the list also contains useful information for other types of open source projects. I was hoping to add some content around project leadership but I was discouraged to see that I was not able to create an account or contribute to the discussion page (FAIL!). Here is what I would have added:

  • There is no road map showing planned improvements to the software
  • The project leader(s) is(/are) not transparent about decisions [ +5 points of fail ]
  • There are no guidelines for contributing to the project [ +20 points of fail ]
  • The project does not have a way to attract and incorporate new contributors [ +10 points of fail ]
  • The project does not have a system for recognizing/rewarding contributions [ +5 points of fail ]