[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: what to do is maintainer is lacking? (was: wine-unstable in Debian)

On Wed, 2012-04-18 at 10:16 +0100, Jon Dowland wrote:
> We have hijacks, we have the tech committee: weapons of last resort.

I was thinking about the same issue before this thread.  My feeling is
that over the same time period in which we've moved towards team
maintenance, and moved to a lower threshold for NMUs, people have become
much less ready to hijack packages.  

Do others feel the same (or indeed does someone have data on the
subject), and, if so, do people think that is a good thing or not?

My impression is that the existence of the MIA procedures has scared
people off hijacking packages where the maintainer is not MIA but
clearly does not have time for that particular package.

Here's another example of where hijacking (or a greater possibility of
hijacking if the maintainer doesn't agree to hand on the package) might
be useful, giving real data but anonymised to avoid making the
discussion personal.  My intention is not to criticise maintainers;
indeed, I would agree with the suggestion of previous posts in thinking
that the problem arises where maintainers feel responsible for the state
of a package and want to sort it out before passing it on, but don't
realistically have time for this.

This is the package that came to my attention:

package1: last maintainer-created upload 2009-07-26,
    then four by non-maintainer but authorised, 
    then NMUs 2011-06-30, 2011-08-06
  package at 1.6.5, upstream at 1.8.3 (major improvements upstream)
  83 open bugs (few replies from maintainer), 3 lintian warnings

Maintainer's other packages:

package2: last upload 2008-01-28
  12 open bugs (replied to one in 2008), 4 lintian warnings
package3: last upload 2009-11-11,
  then NMUs 2010-10-15, 2011-03-18, 2011-07-12
  1 lintian error, 3 lintian warnings
package4: last upload 2011-04-26 (released 2011-03-25)
  1 lintian warning
package5: last upload 2009-11-20
  package at 0.90.0, upstream 0.90.2 in 2010, now dead?
  1 (unanswered) open bug, 1 lintian error, 4 lintian warnings
package6: last upload 2010-10-11, NMU 2011-08-23
  package at 1.5.3, upstream at 1.8.91
  53 open bugs (replied to few), 4 lintian warnings
package7: last upload 2012-02-12 (released 2012-01-12)
  11 (unanswered) open bugs

Since package7 has been updated recently, fairly soon after the upstream
release, the maintainer is clearly not MIA.  But I also don't believe
that it's a good idea for the maintainer to retain tight control over
all these packages in the apparently unrealistic hope of finding time
for them.


Reply to: