Re: next CommitFest

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: jd(at)commandprompt(dot)com
Cc: Chris Browne <cbbrowne(at)acm(dot)org>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: next CommitFest
Date: 2009-11-16 18:23:53
Message-ID: 603c8f070911161023t70a36756n3ca83a955983b56f@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, Nov 16, 2009 at 1:08 PM, Joshua D. Drake <jd(at)commandprompt(dot)com> wrote:
>> True.  But "not enough reviewers to review all the patches we get" is
>> also a barrier to contribution.
>
> No. It is a barrier of contribution not to contribution.

I am not sure exactly what that means, but I agree that it isn't quite
the same. Backing up a minute, AIUI, the CommitFest process was
created to solve the problem that patches weren't getting reviewed in
a timely fashion. To address that problem, dedicated times were
created in each release cycle for people to stop working on their own
patches and review patches from other contributors. I haven't been
around long enough to be able to compare from personal experience, but
I think generally what I've heard is that the new process is a big
improvement. But, there are some problems, and speaking from
experience, one of those problems is that reviewing patches and
running CommitFests is long, hard, and difficult when not enough
people volunteer to review, or not enough committers volunteer to
commit.

I guess I agree with your statement that the structures that are being
proposed are punitive, although perhaps I might choose the word
coercive instead. Clearly, the preferable solution is for people to
volunteer. But if they don't, we haven't got a lot of options.
Perhaps by encouraging them to volunteer and recognizing their
contributions when they do volunteer, we can get the number of
volunteers back up to an adequate level. If after doing those things
we still don't have enough volunteers, we're not going to be able to
review all the patches. Should that occur, we'll have to decide which
ones to review and which ones to skip. Maybe we'll just let people
volunteer and any patches for which nobody volunteers will fall on the
floor or be forever postponed to the next CommitFest. Maybe we'll try
to assign reviewers preferentially to first-time contributors and
those who are themselves reviewing, as I'm suggesting. Or maybe we'll
handle it some other way. I don't know. It seems we don't have to
decide yet.

...Robert

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message SebiF 2009-11-16 18:31:05 Update on Insert
Previous Message Markus Wanner 2009-11-16 18:19:58 Re: write ahead logging in standby (streaming replication)