[Kde-scm-interest] atomicity, again

[Ian Monroe]

On Wed, Jun 17, 2009 at 11:45 AM, Ingmar Vanhassel <ingmar at exherbo.org> wrote:
>
> Excerpts from Boyd Stephen Smith Jr.'s message of Wed Jun 17 18:35:05 +0200 2009:
> > So, I'm on your side of the fight J6t, but I also believe it to be
> > unwinnable.  I also believe it's not *really* on-topic for the list; how
> > many people have a commit-bit is completely orthogonal to what version
> > control software is in use.
>
> A DVCS makse _not_ having push access to official repositories far
> nicer though. Qt's git repository has become a real nice example of
> this, with developers pulling in patches from thirdparty contributors in
> a much shorter timeframe than before. At least, that's my impression.
>
> While I disagree that we need to have this commit-bit discussion now, we
> shouldn't be blindly reject other workflows without carefully evaluating
> their merit. I do think that what Johannes suggests has merits, in terms
> of improving quality.

I could actually see individual projects picking up the "have the
maintainer pull from branches" method. But yea, we should support the
current development method, projects can switch to the pull method
when and if they want to. It doesn't even need to be enforced by
technical constraints: people already routinely submit patches to
projects they have svn commit access to.

Ian