argues that code inspections are as effective in finding bugs as testing - but they tend to find different bugs - so testing and code inspections are complementing each other. And also it requires less effort to fix the bugs found by code review than those found by testing. One would expect that the code in Open Source projects is constantly inspected by the users of it - but in fact I suspect that people only look under the hood when they encounter a problem (at least that is what I was doing). And also I would say that the ego play in OS projects is somehow a bit stronger than in proprietary projects - in fact that should not be a surprise once you realize that pride of their code is one of the driving forces for OS programmers.
So here comes my question - has anyone tried to do a systematic code review on an OS project? Did it work? Do you think it can work?