actually, I'm serious: forbid bugzilla filing against FC4?

[Matej Cepl]

Matthew Miller <mattdm at mattdm.org> writes:
> Things which stay in needinfo too long should get closed as
> INSUFFICIENT_DATA. When a release is obsoleted, all bugs in the devel
> version which were closed in the timeframe of that release could be moved to
> the Fedora Pasture as well.

Being a bugmaster for the desktop team, let me chime here as well
-- I would actually love if the automatic closing of NEEDINFO
bugs was re-established. The period after which it is closed can
be pretty long (couple of months even), there may be some
preliminary warning to owners of the components and the target of
NEEDINFO (e.g., email saying "These bugs on the components you
own will be closed in two weeks for INSUFFICIENT_DATA"), but
something should be done.

We have now hundreds of bugs which are NEEDINFO because somebody
was trying to hide that he is too lazy to think about the
particular bug. Or there are many NEEDINFOs which are actually
meant as CLOSE, but they are not marked as such. Or there are
NEEDINFOs which probably should be translated as NEEDINFO(God)
"Give me some clue what to do with this." I have nothing against
religion, but I don't think bugzilla is the right place to
excercise it ;-).

It should be IMNSHO re-established that NEEDINFO means, that
somebody should provide information. I.e., the last comment in
the bug should specifically point to somebody what specific
piece of information is needed and who is to provide it. When that
information is provided, NEEDINFO should be removed and the bug
should be (ASAP, which depends) ASSIGNed to somebody who will
take care of it. And of course, when this piece of information is
not provided in reasonable time, even after some warning is
given, and especially when it is NEEDINFO(reporter), it means
that somebody is not willing to cooperate and it doesn't make
sense to keep the bug alive.

Yes, our bugzilla is IMHO in much better shape than some other
ones (cough, cough, bugzilla.mozilla.org), but still I am afraid
that it is not as helpful tool for developers as it could
be. TODO list with hundreds of items is just useless.

And last but not least, thank you Matthew for the last couple of
actions you made on the cleaning up the -test* mess. It is very
helpful.

Best,

Matej Cepl