Just walking through reading it but here is a interesting study by the University of Victoria on code review processes in Open Source Projects.
http://opensource.mit.edu/papers/Rigby2006TR.pdf
Some of the research questions are:
- What types of review does a project use?
- Why are patches rejected?
- Who performs the review?
- When are reviews performed?
- How long do reviews take to perform?
Etc.
There is a ton of interesting data in the study from a breakdown of the Linux “Pyramid of Trust” method, two reviewer model that Mozilla uses to the voting model used by Apache.
Well worth the read.
