Hi Andrey,
Am Fri, Jun 05, 2026 at 06:34:02PM +0500 schrieb Andrey Rakhmatullin:
> As a preface, I think the Cc list on these emails is horrible but I'm sure
> it's intentional so I'm not dropping anyone.
The CC list involves all people who posted to the two open bugs (+ MIA
team).
> I still think that formally orphaning unmaintained but not formally orphaned
> packages is the best thing (when they can't/shouldn't be RMed) because it's
> the only thing that explicitly marks a package as orphaned, i.e. that nobody
> cares about it. Shoving them into Debian Commons is not better than shoving
> them into random teams as was done before: it still hides their status, adds
> some busywork and prevents them from being formally orphaned for some more
> years.
I personally see a difference between the QA team and Debian Commons, but
even if we assume for the sake of discussion that moving the package to
the QA team would be preferable in this particular case, what process
would you suggest to formally orphan the package under the current
circumstances?
For clucene-core we have a maintainer who has not uploaded in nearly ten
years, contact attempts that went unanswered, repeated NMUs, and pending
fixes from contributors. Yet, as far as I understand, there is currently
no established procedure that would allow the package to become formally
orphaned without active involvement from the maintainer.
This is exactly the gap I am trying to discuss. If we believes that such
packages should become formally orphaned rather than move to Debian
Commons, what is the practical path to achieve that?
> Don't we have a mechanism where a specific really existing person can become
> a maintainer of an unmaintained but not formally orphaned package directly?
That would be ITS. Any volunteer(s)? I think most of the work needed for
the next upload is already done. However, even if somebody steps forward,
that would only solve the problem for this particular package.
BTW, we had a somewhat similar situation with libasyncns[1]. At the time,
the my mail received neither comments about how such situations should be
handled nor any volunteer willing to take over maintenance. I eventually
experimented with an "Intent to Orphan" process[2].
In hindsight, I agree that starting such a process without first seeking
consensus on the procedure was not a good idea. The criticism I received
on that point was fair.
However, that experience also reinforced my impression that we currently
lack a generally accepted process for dealing with packages that appear
to have no active maintenance but are not formally orphaned. In the case
of libasyncns, the package has since been adopted (#1123614), which is a
good outcome. My takeaway is not that the specific process I used should
be adopted, but that we would benefit from agreeing on some process for
handling such cases.
Without a shared procedure, we seem to be left with discussing individual
packages one by one, while contributors willing to help have no clear path
forward.
Kind regards
Andreas.
[1] https://lists.debian.org/debian-devel/2025/06/msg00131.html
[2] https://bugs.debian.org/1107475#10
--
https://fam-tille.de
