[Aptitude-devel] Removing mafm commit access

Thu Mar 6 16:58:51 UTC 2014

2014-03-06 00:01 Daniel Hartwig:
>On 5 March 2014 07:08, Axel Beckert <abe at debian.org> wrote:
>> Daniel,
>>
>> Daniel Hartwig wrote:
>>> I am not satisfied with the quality of commits Manuel is making.
>>
>> Hey, nobody's perfect!
>>
>> Did we remove your commit access just because we were not satisfied
>> with you being MIA for about 1.5 years? No, we didn't.
>>
>
>At what point did you determine I was MIA?  Did anyone follow the
>usual procedure and e.g. attempt to contact me directly without
>response?
>
>Even though there was no published activity, it is clear from my
>subsequent reports that indeed I have been active, and preparing the
>way to address some of the deep-rooted issues in time for Jessie.

It is clear that there's no Debian activity from you since mid 2013,
independently of what you might have done in private branches.  Which
BTW, either you even didn't fully publish yet, or if you did, it's a
bit less than impressive (most of it was already done in 2012 and was
released as "experimental").  And not integrated yet, after the couple
of weeks that you said that it would take (more than a month ago), so
it would be useful if you focused on working on this rather than
pissing off other people.

You had RC bugs for up to 6 months (#701243, #708812 and #710208) in
your package until other members of Debian were forced to spend time
to fix the only package that you maintan (which shouldn't happen under
normal circumstances, "In general it should be considered preferable
that maintainers take care of an issue themselves" [1]), and you
didn't incorporate the changes in more than 6 months after it was
NMUed [2], nor indeed published any development in all of those
months.

[1] http://www.debian.org/doc/manuals/developers-reference/pkgs.html#nmu-guidelines

[2] http://www.debian.org/doc/manuals/developers-reference/pkgs.html#nmu-maintainer

And in the case that you want to argue this... no, "having RCs fixed
in the VCS" (quite trivial fixes BTW) is not a valid excuse for having
your package FTBFS for months.  Having RC bugs for months in your
package is not nice, wastes many people's time and has a cost for
Debian as a whole.

That's the very definition of being MIA, independently of the
procedures to contact you (for that reason Daniel Burrows was not MIA
until 1 month ago, so neither you nor me should have taken over
aptitude).  So discussing that you are not MIA because you were not
contacted is not going to help you, and shows clearly how you don't
understand Debian ways.

The way in which you handled unstable/experimental releases and let
this work go to waste also shows your poor decisions as developer and
maintainer, BTW.

Anyway, the discussion is not about you being MIA or not, or being a
good or a bad developer or maintainer, it is of you decided who has
the right to collaborate in the project and in which ways, which is a
different matter altogether.

>> I expect you to apologize for your inappropriate behaviour and to
>> re-add Manuel as project member without him needing to do anything
>> himself like e.g. applying again.
>
>
>Just to be quite clear: I do not accept Manuel as a co-developer with
>access to the git repository at this time.  Such access will not be
>reinstated until I am generally satisfied that he and I can work
>together harmoniously.
>
>Manuel did formally remove himself from the position and duties of
>maintainer some time ago due to the same conflicts.

I already told you in previous messages that you are wrong about all
of this.  You keep thinking that you are entitled to be the gatekeeper
who decides who can collaborate and how but you are not, so it would
be less painful for everybody and more productive if you come to terms
with this sooner, rather than later -- or better, immediately; and did
what Axel suggested you to do.

In the same way that I decided to step aside in 2012 (when I thought
that you would take good care of the project), I now decided to come
back, because the project was poorly maintained again since late 2012
(or not maintained at all).

And now, as before, the decision is mine and not yours.  It's not for
you (nor for everybody else) to decide that I or any other contributor
should not be members the project or not have necessary permissions to
accomplish tasks (e.g. commit acces to VCS), and everybody is welcome
to participate as long as they don't cause intentional harm, respect
others, and do good work and sensible contributions.

BTW, I was the first to gain member and commit access to the project
[3], so I am more senior than you, if you want to play that game :-)

More importantly, you've never been a "long standing maintainer" (less
than 1 effective year of maintenance is not "long standing" by any
stretch) and you cannot even be properly called "maintainer" since
2012, so come down of your high horse please.

[3] http://lists.alioth.debian.org/pipermail/aptitude-devel/2012-January/001810.html

>Meanwhile I have done all the work _required_ for Wheezy, such as
>implementing multiarch, resolving all RC bugs, assisting transitions,
>etc.

Debian is not only Wheezy or stable releases, and RC bugs in unstable
are almost equally serious and need continuous attention.  For
example, people working on mipsel ports could not use aptitude at all.
So you cannot call you active maintainer if you don't pay attention to
these things.  You stopped being active mantainer long ago, by
definition.

Apart from that, it is not only you have not done all of the work
_required_ for wheezy, since other people participated in the
development in the months previous to the freeze (you have to count
the whole release cycle, not the freeze).

And even if you did, that doesn't mean that you have the right to veto
contributions from other people or remove project permissions, so it
would be helpful if you stopped discussing things not related to the
matter.

Cheers.
--
Manuel