Strawman: eliminating debdiffs

[James Westby]

On Wed, 2008-10-01 at 12:08 +0200, Luca Falavigna wrote:
> James Westby ha scritto:
> >   * The contributor finds a fix to a problem, and forwards the
> >     patch upstream. They follow the progress of the bug and
> >     work with upstream to get it committed.
> 
> This should be the procedure, reporting our changes upstream (Debian,
> upstream bugtracker, whatever) as long as it's applicable there ...

Yes, it's my belief that our current procedures emphasise this
as a secondary concern, I would like to change the procedures to
make it a primary concern, along with making Ubuntu a great 
distribution.

> >   * Once the fix is committed, or some time has passed with no 
> >     comment from upstream, if the contributor deems the fix
> >     important enough to warrant an upload before the new upstream
> >     is packaged they seek a sponsor.
> 
> ... but upstream projects and Debian have different timings than ours,
> we would want to push a fix to solve someone's problem ASAP, waiting for
> someone to release the fix for us could lead to longer delays. Ok, we
> can upload the fix directly, but we must discuss if a fix is important
> enough to have it in the archives immediately or not and check
> constantly if upstream committed our changes to finally close our bugs.
> I'm not sure this is the right approach because it would require much
> time and efforts to have it in place, what about ...

Currently for a lot of packages the only fixes we apply are those for
which someone in Ubuntu development requested sponsorship. That 
generally means that a bug was filed by a user, someone wrote or
found the patch, and then did the work to get it sponsored. Is that
the best criteria for which patches we pull in?

It relies on people "running in to" important bugs. We probably
do ok from this, and without more people watching upstreams we
can't spot important fixes originating from elsewhere anyway.

However, it means we have a lot of patches for quite un-important
problems. If they are not forwarded upstream then we are just
making more work for ourselves. If they are forwarded upstream
then my only concern would be where the effort is being expended.
For the Intrepid release now I'd rather fix important bugs than
more typos in package descriptions.

> >   * The sponsor grabs the patch and reviews it, with more scrutiny
> >     if there has been no comment upstream. They drop it in the
> >     package and add a changelog entry, which will be easy because
> >     contributors will be encouraged to provide a lot of information
> >     about the fix.
> 
> ... including upstream reference into the changelog entry? A link to
> upstream bugtracker, VCS commit or whatever should be enough to track
> status of our submitted patches, maybe enforcing [1] as our policy.
> 
> [1] https://lists.ubuntu.com/archives/ubuntu-devel/2008-June/025551.html

I'm aware of the discussion, and in support of that policy, and in fact
that is how I write my changelog entries. However, I haven't seen a big
change since the discussion. It's clear that not everyone thinks it
is worthwhile, or sponsors aren't willing to require contributors to do
this, and aren't doing it themselves when they upload their own fixes.

Also, the thing that this proposal will often miss, that mine tries to
incorporate, is feedback from upstream on the patches, which I find very
valuable.

Thanks,

James