Re: On maintainers not responding to bugs
* Sune Vuorela <nospam@vuorela.dk> [070226 14:28]:
> > Hey, I was speaking about not releasing such packages at all (perhaps
> > except libc and the kernel (at least until hurd is ready ;->)). So
>
> Eh? you say that they should not migrate to testing ?
To cite myself:
| Sorry to say it that way. But after I read this, not letting such
| packages in a stable release before they get enough manpower to be
| handled, sound a better idea then before.
There can be said many things why stopping testing propagation in
unanswered bug reports is a bad idea: It may be circumvented by
sending empty mails or something clearly automated. It needs some
additional tracking to make sure everything is answered, even
things added by the maintainer itself. And there may be situations
where a big packages have to prioritize on other things first so
it is totaly is unpractical, unless it at least ignores recent
bug reports (for long enough values of recent).
But Pierre argued that handling bugs is "tedious" work and there
is no time for all of them and still packaging new upstreams.
Given that older bugs are usually harder to deal with than new
ones, means that there will be a permanent ever increasing backlog
of bug reports, i.e. no handling of non RC-bugs at all. Citing such
a situation is no reason against anything but against not letting
that package in as there are not enough volunteers to support it.
> I think 'forcing' is bad, but it could be a starting-platform for those
> on hold in am-wait.
People doing the package maintainer route already have to show some
sponsored package and when AMs then ask the sponsor if they showed
they can handle a package and with bug reports, there is little to
say as little packages often do not get many bug reports. Making
people deal with real bug reports might make it easier to assess
their qualities (and thus helping them faster in) and help our users.
(I wished my day had 72 hours more, AM work would really tempt me
when I had much more free time)
> And you are still welcome to help.
And you are welcome to invent a time machine. I always try to look
into some other bugs of a package when I send in a bugreport or
a patch. But especially if no bugs has any hints from the maintainer
and there a many old bugs where I would need old packages to find
out if I just fail to reproduce or i they are fixed a long time,
I often give up and invest more time to being sucked into some upstream
work...
Hochachtungsvoll,
Bernhard R. Link
Reply to: