Re: Developer Status
Joerg Jaspert wrote:
> On 11547 March 1977, Felipe Sateler wrote:
>>> While, strictly speaking, this increases the barrier to get DM compared
>>> to the current implementation of DM, we do not think it is an
>>> unreasonable or too high level. Anyone who is able to get a package put
>>> together in a lintian clean way will be able to get DM without much
>>> effort or time used.
>> So this basically requires DMs to do the (somewhat reduced) P&P and T&S
>> questions, and I don't see the real reason for this. The idea behind DMs is
>> to maintain a package one knows how to maintain. The only reason I can see
>> here is that DDs are not being trusted in their advocations, which is a far
>> worse problem that won't get solved by this.
>
> As written above, the actual amount of checks DM have to do is left to
> the NM Committee and can be adjusted as needed.
> This does IMO not contradict DM, but it enhances the value of it in
> the long term. Right now you can be DM, but have to pass the full NM
> before you can get DD. In future you get DM (and yes, you have a tiny
> amount of more work to do for this), be it for a few months and then you
> can get DD relatively easy by just doing the last few steps
> needed. Ie. once you are a DM you are already about half through NM.
What you are doing is moving the work from one side to the other (pre-DD to
pre-DM). Given that the idea of DMs was to reduce the barrier of entry, it kind
of defeats the purpose.
>
>>> Those two "classes" are the initial set in which every NM will end
>>> up. After six months as DC or DM one might chose to become a
>>> Debian Member or Debian Developer. This
>>> - ensures that the interest in Debian isn't short-term.
>> Why do people keep thinking this is a good thing?
>
> Because short term involvement usually creates more work than it helps, IME?
I would like some numbers to back this claim. Also, what is short term? I would
find one year to be sufficient to do a great amount of good work, but then a >
6-month process is too much bureaucracy.
>
>>> - enables them to learn more about the workings in Debian and generally
>>> helps them for the next step.
>> They should be doing this on their own, and not force an arbitrary limit on
>> them. What if they did this before applying for DD/DME/DM/DC status?
>
> Then it will be a no-brainer to pass the rest in this
> process.
That is not the point. Why do I have to wait 6 months to "learn more" when I
already did that?
> Unfortunately a large number of people trying to join Debian
> did not do that before applying, as past has shown us.
I really don't see why NM should be a mentoring process. Instead of asking NMs
to do stuff during the process, I think it would be much better to ask them to
show what they have already done. This way you get less work for the
AM/whatever and "force" NMs to do their homework before applying and not while.
>
>> This all smells like a whole lot of bureocracy for no gain to me.
>
> I do see the gain for everyone that wants to join and also for those
> non-technical people that do help to make Debian better. Not everything
> in Debian is about packaging alone, even if some people (not attacking
> you, thats a general some) love to think there is nothing besides
> packages. We do want translators, we do want documentation writers, they
> all help to make Debian even more the Universal OS by opening it up to
> those not speaking english. Or those that aren't the most clueful ones
> in technical things and just want to use a great OS to get their daily
> work done.
> Yes, the most important part in Debian is about packaging, without this
> there won't be a place for translators/documentation writers, but that
> doesn't mean we should ignore or deny their existance.
I do want translators, documentation writers, etc to be recognized for their
work. I also want them to have the tools to do their work. The DMe thing would
only accomplish adding your name to a list, since it gives you no
tool/privilege to do your work better/easier. I want them to do, eg, i18n
campaigns like Christian Perrier does. For that they require technical skills,
that is true. But then every contributor to any software project has to learn a
few things before they can be effective (eg, learn about patching or the VCS in
use).
--
Felipe Sateler
Reply to: