Re: [Backports-queue] Processing of qemu-kvm_0.11.1+dfsg-1~bpo50+1_amd64.changes
On Thu, 18 Mar 2010 19:52:39 +0100
Gerfried Fuchs <email@example.com> wrote:
> * Andres Salomon <firstname.lastname@example.org> [2010-03-18 18:52:17 CET]:
> > > > Contacting the former uploader is on bpo's best practices but
> > > > it's on a strict rule of bpo, is it? Not that it's a bad idea,
> > > > it's just not always very practical.
> > I'd love to see this formalized as a strict rule. Say, a formal
> > requirement would be to email the uploader (using the email address
> > listed in the upload), and if there's no response within 30 days,
> > permission is implicitly granted to maintain the backport.
> I still wonder why this would be needed. Do we have it as strict rule
> for regular Debian uploads too? Actually I fail to see the big
> difference in this respect and why people should see it differently.
> If it is actually needed to get it strictly written out it should be
> in the developer's reference - and the backports documentation might
> refer to that.
I think this is actually a failing within Debian itself, the fact that
it's so hard to take over neglected packages (and that it it involves
delays of undetermined amounts of time, especially if the maintainer
initially replies w/ "oh, I'm busy, but I'll get to it next week" and
then never does). There's no reason why bpo should suffer the same fate,
w/ packages stuck in limbo. Bpo might actually pave the way here for
future developer's reference changes.
I threw out the 30 days thing as an example period of time; it's short
enough that one probably won't get discouraged/forget about the package
(it can be hosted in a private repo in the meantime), while long enough
to account for busy people or accidents (having recently been
hit by a car, I can attest to sometimes requiring more than just a week
or two to respond to Debian-related stuff).
> Having said that:
> ,------------------> quoting developer's reference <------------------
> | It is not OK to simply take over a package that you feel is
> neglected — | that would be package hijacking. You can, of course,
> contact the current | maintainer and ask them if you may take over
> the package. If you have | reason to believe a maintainer has gone
> AWOL (absent without leave), see | Section 7.4, “Dealing with
> inactive and/or unreachable maintainers”. `------------------>
> quoting developer's reference <------------------
> That paragraph is quite exactly the formalized strict rule you would
> have hoped for, and it is the one that should be taken into account
> for backports, too.
Not really. Wording like this isn't particularly strict:
"and wait a
reasonable time for a response. It is quite hard to define reasonable
time, but it is important to take into account that real life is
sometimes very hectic"
"If the maintainer doesn't reply within four weeks [...] you should
investigate further, and try to gather as much useful information about
the maintainer in question as possible."
(Well, no. If the maintainer doesn't respond to both a private and
public ping within a month, the onus should be on the maintainer to
prove that they still care about the package. Imho.)
How long should one wait if the maintainer claims they're actively
working on their packages, but sees no progress month after month?
There are plenty of examples of this sort of thing happening within
Debian. Note that I'm referring specifically to maintainer comments of
"I'm working on this", rather than "No, that should never happen" or
other kinds of technical disagreements between users and maintainers
(those would be outside the scope of this discussion).