DVCS comparison for our organisation: decided

[Nicholas Allen]

> 
> Except you then have a different race condition. Which is that 2 people go to
> commit at the same time, and the pre-commit *passes* on both sides, but only 1
> of them can win. Otherwise the second commit is not based on the first commit
> to succeed.

Yes exactly. You put it better than I ever could have ;-) The PQM not 
only ensures that each commit passes the test suite but that it passes 
even after applying the commits that were accepted before it. This is 
something that a pre commit hook that Ben would like would not guarantee 
(unless it blocked as I proposed). Personally I would rather the PQM 
approach even for smallish projects but if the pre commit hook is 
adequate and the team small enough then it is slightly simpler to setup 
using the hooks I guess.

Nick