Code review required for commits - Discuss.

[Jelmer Vernooij]

On Thu, 2012-10-11 at 19:15 -0700, Jeremy Allison wrote:
> On Thu, Oct 11, 2012 at 09:58:12PM -0400, simo wrote:
> > On Thu, 2012-10-11 at 18:23 -0400, Scott Lovenberg wrote:
> > > 
> > > Ira, we discussed this a bit in the IRC channel today, but I'm
> > > curious.  How does this work for a non-member that has code in Samba?
> > > Can I maintain the parts of code that I've written without two team
> > > members signing off?  It's really not a big deal, but we might as well
> > > get these things ironed out while there is discussion going on.
> > 
> > As far as I am concerned 1 signoff (author) and 1 ack should be
> > sufficient, independently of who is the author.
> > However it would be really desirable that who acks is the maintainer.
> 
> I'm flexible on this. I only suggested 2 Team members as
> that's what we currently do for bug reports.
> 
> The main thing is to move to reviewer-push IMHO.
What's the advantage of reviewer-push? I would prefer to limit the
amount of work the reviewer has to do, so that people are more inclined
to do reviews and require less time to do reviews.

Cheers,

Jelmer