[Aptitude-devel] Stepping down as maintainer/developer

[Manuel A. Fernandez Montecelo]

Hi,

2012/3/2 Daniel Hartwig <mandyke at gmail.com>:
> On 2 March 2012 21:29, Stefano Zacchiroli <zack at debian.org> wrote:
>> [bit re-added]
>> Hi folks,
>>  I happen to be on the aptitude-devel list and also being an aptitude
>> user. I've been lurking activity here in the past months, the amazing
>> evolution of the code base and of package bug count, and this discussion
>> in the past few days.

A not very important observation (for the discussion at hand):

In my opinion the codebase didn't change that much so far, in the
sense that, even if some of them are important changes, I would not
say that it has suffered an "amazing evolution", not so far anyway.


>> I'm sad to see you two seem not to get along. I'm even more sad that, as
>> a result, the state of aptitude code base is not progressing as swiftly
>> and as quickly as it could.

(Maybe another unimportant observation, feel free to skip in your replies).

Even if I agree with the general idea/intention of your email, I'd say
that it's not fair to say whether the progress since "the conflict"
has been slowed down or not.  For two reasons:

1) There has been no time enough to measure if the progress is slowed
down, this "outburst" happened only last weekend.  It was not as if
previous disagreements hindered the development of aptitude for
weeks/months -- or not that I am aware of, anyway.

2) I was not joking or ironising when saying that I'm sure that Daniel
(and eventually others) would take good care of it, because Daniel
Hartwig has been looking and fixing problems for longer time, goes
deeper addressing more important bugs, and knows what to do better
than me in general.

So yes, it's conceivable that less manpower will slow down the
development, but in general I don't think that it can be assessed in
one way or another at this point.  As I said before maybe this is not
important, but I wanted to clear that up because it surprised me that
you arrived to such conclusion.


>> None of us has magic wands that can make people who don't get along
>> suddenly get along. But let me state the obvious here: moving away from
>> contributing code is a net loss for everybody. If you two --- and
>> everybody else who is willing to contribute code --- do not manage to
>> get along and step back from contributing, you'll make aptitude suffer.
>> And that is bad for Debian users, no matter who is right or who is
>> wrong. Please think about what is at stake.
>>
>> People might have technical or procedural divergences that block
>> contributions, but we need to find ways to fix them. If you've already
>> tried discussing and that didn't work, we have other ways. For instance
>> we can have someone mediating among the existing positions. I'll be
>> happy to mediate on this, if you want me to.
>
> [...]
> Thank you for your offer to assist in the matter.  I'm happy to
> work with yourself or anyone else as a mediator if this will
> dissolve the tension.
>
> Manuel, what do you think?

Well, I think that I overreacted after seeing several emails
critisising my work (sometimes in too much detail, I think) when
arriving monday morning, after a weekend of sleep (and street)
deprivation working in aptitude; so I felt quite bad and frustrated.
I will try to get over it and act less passionately, but I will need a
bit of time for that.

Also, I will consider Daniel Hartwig the primary developer for the
time being, that will probably help me to get more
emotionally-detached from the project and avoid rants/etc in the
future.  Other than that, I will try to pay more attention to keep the
same coding style etc.

Despite that, I still firmly believe that my claims with respect to
what's understood with "increasing quality", how to not delay
uploading the packages because of certain class of lintian warnings
and so on are still valid and justified, and that refactoring as one
goes is good thing to do (but I will not pursue it further).

I also think that more aggressive/intrusive changes are needed to
address many of the current and upcoming problems, and that the new
people shouldn't be too constrained by previous decisions.  Maybe the
overhead of carrying such legacy is partly the reason why previous
folk (esp. Daniel Burrows, of course) didn't continue with the project
now.  I'm talking about things like disabling aptitude-gtk altogether
until/unless somebody takes responsibility for it; or functionality
like a separate resolver maybe is unneeded and that aptitude should be
more integrated with apt (although I haven't looked up to the
implementation and don't know its current status or advantages over
apt, the complexity that it adds, etc).

I think that it was appropriate to mention these things concerning me,
so you'll know where I'm coming in my future opinions.  Having said
this, I think that I'll refrain from comments like this (possibly
highly controversial) or to get too involved in the future, just in
case.

Cheers.