Re: ignoring bugs with no maintainer (Re: Removal of emacs23 from unstable/testing)

[Ian Jackson <ijackson@chiark.greenend.org.uk>]

Andrei POPESCU writes ("Re: ignoring bugs with no maintainer (Re: Removal of emacs23 from unstable/testing)"):
> Last time someone (Bcc'd) tried to tackle these (admittedly without
> contacting the maintainer in advance) the contributor was prevented
> from doing so and was requested to either check them against emacs24
> or leave them alone. It's not hard to imagine what happened...

One of my bugs was involved in this situation and I was one of the
people (the person?) who objected.  I contacted the maintainer who
agreed with me that the bugs should not have been closed.  After
discussion the contributor's actions were reverted.

I don't think it is appropriate to mass-close bug reports without
approval from the maintainer.  Failing to get the maintainer's opinion
was not a minor communication error - it was a critical mistake,
because it led to harmful (if well-meaning) action by the contributor.

It's true that some packages are so overwhelmed with bugs, and have so
much churn in the code, that old bug reports are nearly worthless.  In
such a case I can see that the maintainer might have a policy of
closing some categories of old bug reports.  This is obviously not a
desirable situation, but it might be the least bad approach to a
difficult situation.

BUT: Firstly, this should not be the usual approach in Debian.  And
secondly, whether to adopt such a policy is for the maintainer to
decide.

I think that the Developers's Reference could do with a section, near
the beginning, on the role of the maintainer in Debian.  That would
make all of this much clearer.  I will see if I can come up with some
wording.

Ian.