David Glasser wrote:
> On Jan 28, 2008 1:25 PM, David Glasser <glasser_at_davidglasser.net> wrote:
>> I think that given the amount of work that is needed, we should have a
>> relaxed approach to backporting until we're actually ready to roll a
>> release candidate (either not using the STATUS system at all, or
>> requiring just two votes (which can be on IRC), etc).
>
> Now that Mark has proposed a time for branching (which I am +1 to),
> I'd like to propose something more specific here. I think that for
> the short-term future, until we're closer to the point of an RC, that
> we treat the entire 1.5 branch with the policy we usually apply to
> non-core code. That is, we require one +1 from a full committer and
> one +0 from any committer to do a backport. (And this can happen over
> IRC or whatever and never even pass through the STATUS file, as long
> as the votes are recorded in the backporting log message.)
Sounds good, though I would recommend still having a STATUS file. It is
a convenient place to put nominations when nobody is around for an
immediate review. (Not that you were proposing to do away with STATUS
completely, I just wanted to make sure it was clear.)
-Hyrum
Received on 2008-01-29 04:36:37 CET