Re: Idea for maintaining packages up for adoption
Russ Allbery <email@example.com> writes:
> Steve Langasek <firstname.lastname@example.org> writes:
>> Packages need to have maintainers -- meaning, someone needs to take
>> responsibility for the package. Orphaned packages *routinely* slip into
>> stable releases with release critical bugs that have been in the package
>> for a year or more, sometimes even introduced by a QA upload. We don't
>> know if these packages have users, but we *do* know there's no one in a
>> position of responsibility over the packages who's using them and is
>> fixing bugs that appear during use!
> After having spent a couple of hours looking at the bug database for
> orphaned packages yesterday, my feeling is that what orphaned packages
> really need are users who will actually use reportbug.
I think you are right about this.
Of course, there are a couple of problems with the way reportbug and the
BTS work that I think impact on this.
One is that if you have a bug you don't know you need to use reportbug!
Sounds silly, but for users coming from the 'doze background, persuading
them to use reportbug to file a report is very hard.
Second is that you get very little feedback from reportbug. Yes, you get
emails telling you that the bug is filed, and yes you often get a really
fast response from an active maintainer, but in the worst case you file
a BTS report and never hear anything again - and that is *not*
A mail a month saying "your bug XXXXXXX is still open. N comments have
been submitted. In the last month M bugs were fixed on this package. If
you don't want any more of these mails, reply to this one" would be
positive in this respect.
Third: reportbug should start up with a commented example of a good bug
report, to give people something to aim for. This could be different
each time, chosen from a pool of reports developers had found to be good.
And should the submitter of a bug be auto-subscribed to the bug? I
think at present it is possible to carry on a long debate on the BTS
without the original submitter knowing - which seems to miss the point?
rich walker | Shadow Robot Company | email@example.com
technical director 251 Liverpool Road |
need a Hand? London N1 1LX | +UK 20 7700 2487