Re: PING^2: resubmitted IRA improvement patches

[Vladimir Makarov <vmakarov at redhat dot com>]

 On 10/06/2010 10:08 AM, Vladimir Makarov wrote:
>  On 10/05/2010 10:57 PM, David Edelsohn wrote:
> > Many FOSS projects and companies operate very successfully and
> > efficiently on a "code review required" basis.  Nothing is perfect,
> > but I generally have heard good things from projects that implement
> > that model.  Having a second pair of eyes double-check a patch can
> > help a lot.
> Many FOSS projects are successful *with* "code reviewer required 
> basis" does not mean they are successful *because of* it.  I feel that 
> GCC is doing better than five years ago.  And IMHO that is mostly 
> because GCC got a critical mass of experienced professionally working 
> developers (although I guess improving reviewing process helped a bit 
> but IMO again not because of the policy but more because the 
> developers pay more attention to it).
> > Small patches by developers in their own area of expertise probably
> > will not gain a lot from a review, but I have witnessed many cases
> > when someone pointed out an API that the developer should have used or
> > a problematic corner case the developer missed.
> I don't think that only reviewer approach is right thing.  What if two 
> reviewers do not agree on a third party patch.  I saw recently a nasty 
> conflict on this base on GCC mailing list (I was myself in the same 
> situation recently in much less severe form).  What if a patch creator 
> (and also a reviewer in the patch code area too) does not agree with 
> his patch review and he believes he knows importance of the patch better.
>
> Instead of just a bazaar (although there are not negative sides of 
> such development) you might be creating a bazaar squared.

Sorry for the typo.  I meant "there are not *only* negative sides..."

> I think a natural way to resolve this issue is to have some soft 
> authority based in other words some credible person with status higher 
> than just reviewer.  You could call it a maintainer.