Code review required for commits - formal Team vote.

Jeremy Allison jra at
Sun Oct 14 21:55:47 MDT 2012

On Sat, Oct 13, 2012 at 05:48:23PM +0200, Michael Adam wrote:
> Interestingly, Jeremy who has proposed this move at this time
> is by himself someone who has not been into the branch-and-rebase
> business to an exaggerated level, and rumour is that he has also
> pushed patches prematurely in the past (as most of us have).
> So guess what his commitment to doing mandatory code reviews
> and non-author push also means: that he is not only trying to
> discipline others, but also himself... :-)

Guilty as charged :-). The problem with the current system
of reviewer is that they have no "skin in the game". It's
easy to look over work, think "yeah this looks kind of ok"
and say it's ok to merge without any more real thought.

I am guilty of the above as well :-(.

If you actually have to push that code yourself it is
psychologically now *YOUR* code, and brings a much
deeper level of commitment than merely looking over
code someone else will push.


More information about the samba-technical mailing list