On Fri, 2009-04-24 at 20:38 +0100, Peter Robinson wrote:
>> > I hereby proclaim my candidacy to become
provenpackager.
>>
>> It doesn't help you to close all these tickets, because you cannot
>> simultaneously be both reviewer of package and package owner.
>
> The provenpackager would not be the owner, but just the one who applies
> fixes for issues that are found (or would be found) during the merge-review.
But the issue with that is that even though the provenpackager
wouldn't own the package they are making the change to the package
that they're reviewing and the problem with that is that there is no
review of the review to confirm that they are correct. The whole point
of the review/merge requests is that more than one person is reviewing
the package. In the case where the reviewer applies the fixes there is
no second person.
One one hand you are correct, but on the other hand
http://fedoraproject.org/wiki/Who_is_allowed_to_modify_which_packages
clearly states:
"If ... the changes are quite minor or considered as a general cleanup
to a lot of packages then provenpackagers are allowed to fix stuff in
other peoples packages."
From this I'd say that provenpackagers can make trivial package
review
cleanups to make the package pass the review, if the maintainer doesn't
react in some time.
Making corrections to the spec file doesn't mean building the new spec
file, not to mention pushing the new build as an update. That should be
left to the maintainer, if at all possible.
We need to shorten the review queue. Any measures that help doing so
without compromising the quality of the reviews should IMHO be done.
--
Jussi Lehtola
Fedora Project Contributor
jussilehtola(a)fedoraproject.org