GNU bug report logs - #61894
[PATCH RFC] Team approval for patches

Previous Next

Full log

Message #17 received at 61894 <at> debbugs.gnu.org (full text, mbox):

From: Ludovic Courtès <ludo <at> gnu.org>
To: Christopher Baines <mail <at> cbaines.net>
Cc: guix-devel <at> gnu.org, 61894 <at> debbugs.gnu.org, guix-maintainers <at> gnu.org
Subject: Re: bug#61894: [PATCH RFC] Team approval for patches
Date: Wed, 01 Mar 2023 23:45:27 +0100

Hi Chris,

Christopher Baines <mail <at> cbaines.net> skribis:

> I guess I'm still a team sceptic, I think the idea is interesting and I
> have added myself as a member of some teams. But the main impact on me
> so far is that I've just been getting some unwanted personal email,
> messages that previously wouldn't have landed in my inbox have been
> doing so.

Same for me (took me a while to understand why I was suddenly Cc’d on
some many patches.  :-))  I’m not sure how to improve on that.

> Regarding this change specifically though, I'm unclear how it would
> impact the things I push for others. I pushed some patches today, would
> this mean that I'd have to look at what team/teams are involved
> (according to /etc/teams.scm.in) for each commit/series, and then either
> continue if I'm a member of that team, or skip it if I'm not?
>
> If I'm going to not be pushing stuff I would have previously pushed
> because I'm not in the relevant teams, maybe I should just add myself to
> every team? I guess this is not a serious question, but I'm more making
> the point that if teams become a formal part of patch review, then some
> formalities over membership of a team is probably a prerequsite.
>
> As a point of clarification, if a patch or series touches files that
> fall within the scope of several teams, am I correct in saying that this
> change would require approval from all teams?

Good questions.

For teams like ‘core’ or ‘home’, there should be no overlap, so it’s
quite easy to see who’s in charge.

Teams related to packages are more likely to overlap, and it’s also an
area where we generally want more flexibility.  The example you
give—pushing patches even though you’re not on the corresponding
team(s)—is something we’d still want to allow most of the time.

There seems to be different requirements depending on teams.  I’d like
more coordination and clearer responsibilities for subsystems (guix/*,
gnu/{services,system,build}/*, etc.) and key packages/tools (Python,
ocaml-build-system, etc.).  For “random packages”, I’m fine with the
status quo.

WDYT?

Thanks,
Ludo’.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.