Re: Developer Status
Bas Wijnen wrote:
> On Thu, Oct 23, 2008 at 05:40:32PM -0300, Felipe Sateler wrote:
>> >> Basically, they need to pass the ID check, agree to the Social
>> >> Contract/DFSG and have successfully answered a set of questions
>> >> similar to the ones used in the current first P&P step, to keep doing
>> >> the same thing they have been doing all this time.
>> >
>> > No. Current Debian Maintainers also need an ID check, agree to
>> > SC/DFSG/DMUP and be advocated. The only thing that is added (and that
>> > was made clear by Joerg), is that they need to answer a very limited set
>> > of questions.
>>
>> I am talking about the DNDCs here. DNDCs have no priviledge whatsoever
>> besides getting included in a list.
>
> Yes, and possibly getting a @contributor.debian.org e-mail, appearantly.
> So what does Debian want to do? We want to show those people we
> appreciate their work, and we want them to be able to tell others that
> they do work for Debian. We also want this to be worth something, so we
> shouldn't add just anyone to the list, but only people who agree with
> our philosophy.
>
> In order to be able to say anything, we need e-mail most of the time,
> and in order to identify someone as "the person I'm talking about", we
> need a signed key. So the ID check is going to stay. For the
> philosophy thing, we need agreement with our foundational documents
> (which isn't any problem, of course). The need to be advocated seems
> reasonable to me as well, to maintain the status of that list. So
> there's only answering the questions, and I think that was only the case
> for DDC, not DNDC.
My name is on the Maintainer field of 2 packages in main, and I think we can
consider the Maintainer fields as "a list of contributors" (evidently, not all
of them). I haven't (formally) agreed to any document, my key is not signed by
anyone in the project, I haven't been advocated by anyone and certainly have
not answered any set of predefined questions. Why should the bar for
non-developing contributors should be different than mine (ie, they have to do
more than just the work they are contributing)?
>> > Becoming a Debian Maintainer is supposed to be a light-weight version of
>> > the New Maintainer process. It's not a "I'll skip the New Maintainer
>> > process entirely"-version.
>>
>> If the current Debian Maintainer process is failing for some reason, please
>> elaborate. If it's not, then I don't see why adding more checks is useful.
>
> I don't think it's failing, but I also don't see where the "more checks"
> would be. You're talking about the very limited T&S questions? Jörg
> made it clear that this wouldn't be much trouble, and that people should
> be able to finish the checks in a very short time. You may be right
> that there's no reason for them, and in that case it would be better to
> remove them. But it's also not a big issue IMO.
Maybe. But then, big issues are IME created usually by small incremental issues.
I think the "burden of proof" is on the proposer's side, not the other way
around.
>
>> But I think that for general upload rights the bar is way lower. As I
>> said in another message, 1 year is enough to do a lot of work, but
>> spending half of that year waiting is not useful, I think.
>
> If a person needs to learn about Debian packaging at the start of the
> year, then I don't think it's reasonable to expect much work on more
> than a few packages, at least in the first 6 months. And for a few
> packages, there's no need to get full upload rights. Just becoming a DC
> is enough for that, and that needs no waiting.
Only if this person applies for DC status at the start of said year, which may
or may not be what people are actually doing: my first package was uploaded in
2006 and I still haven't applied for DD or DM. Others may do it differently.
>
>> > You seem to want to rush total outsiders into the most priviledged
>> > positions of the project. Why would that be a good thing? What is the
>> > problem of letting people work 6 months with slightly fewer rights?
>>
>> I don't want to rush people into privileged positions. I object arbitrary
>> limits, specially when I think the limit will miss many important cases.
>
> I don't see the many cases you are talking about. One effect of this
> proposal is that people should apply for DC when they are getting
> started. If people don't do that, but instead are active but not in any
> keyring, then 6 months is a long time to wait before being able to apply
> for DM.
Apparently I was thinking mostly in the latter case than the former. I consider
it nonsense to apply for official status while doing my first contributions to
a project. I don't usually go to a project and say "hey, I want to contribute,
please give me (restricted) access to your VCS". I think it would also cause
more work for the people administering the Dx queues, which I hear are already
overworked.
> It could be good to allow skipping of the delay for one month
> per advocate, which means you need to get seven advocations (one to
> start, plus one for each month) to start immediately with a DM
> application instead of DC. If people are really active, getting seven
> advocations shouldn't be too hard. If they aren't, then the waiting
> isn't a bad thing. That sounds like a good idea, actually. Jörg, what
> do you think about this?
In your scenario, how is one supposed to get an advocate before doing any
contributions? Also, given that the time limit is arbitrary (in the sense that
it's there in hopes of other things to be done, but there is no real check that
these things are being done), I could argue that if the contributor can be
tracked within the project for X time, then the waiting time should be 6
months - X.
>
>> > Of course there technically is a full and almost full rights membership.
>> > What I think he means to say, is that DNDMs should not be looked down
>> > upon, and that they do get everything they need from the project.
>>
>> That's why I said "you might not intend that". If they are effectively
>> almost-DDM, there is a large room for looking down.
>
> Which is also why I prefer my naming scheme. Almost all the time,
> people will be talking about DMs then, so there's no reason to look
> down; we won't even remember who is DDM and who is DNDM.
I would find a negative form more downgrading. Ie, the Non-Developing Member is
clearly inferior to the Developing Member.
--
Felipe Sateler
Reply to: