Introduction: new community guy, and some discussion

[Adam Williamson]

On Thu, 2009-02-05 at 21:06 +0000, Anne Wilson wrote:

> There are no winners, Rahul.  Personally I've been using both Fedora and 
> Mandriva, and I can see advantages in both approaches.
> 
> I'm not asking for anything to change, just some guidance that make sure that 
> we understand what is and what is not a Fedora issue.

There's two sides to this, theoretical and practical. The theoretical is
simple but worth writing down for clarity: if it's in code from upstream
it's an upstream bug, if it's in code from a Fedora patch or Fedora
supplementary source or it's a bug caused by Fedora's packaging in some
way, it's a Fedora bug.

The practical side is actually determining the above. That can be really
tough, I'm not sure if it's practical for reporters to always be able to
do this - to me it's one of the primary responsibilities of Triage,
however that's defined.

There are also wrinkles. Mainly, there are several cases where upstream
bugs have to be addressed in Fedora. I'd say the two most important
cases are:

1: really critical bugs. Even if these are being fixed upstream with the
intent to use the updated release to fix the problem in Fedora or to
backport the fix, it is good to have a bug in the Fedora system to
monitor this. Say a bug that's release critical for F11 is in upstream
code - we should still have a bug in Fedora Bugzilla that's marked
release_critical in order to track the progress on the bug being fixed
upstream and the fix getting into Fedora.

2: bugs that will not be addressed upstream. The most obvious case is
the one where upstream's dead, where even 'upstream' bugs will need to
be addressed in Fedora. But there's also the case where, for instance,
it's something upstream won't change but it makes sense to change it for
Fedora (though we try to minimize that), or the case where it's an
important issue that upstream will not be able to fix before it's too
late for some Fedora release.

As I see it there's two areas for action here: deciding whose
responsibility it is to determine what bugs should be in Fedora BZ and
what bugs should be in upstream (and codifying that somewhere in the
Wiki), and deciding if it's a good idea to have some kind of policy for
'exceptions' where an upstream bug is tracked in Fedora BZ or if it
should just be handled on a case-by-case basis.

What do you guys think?
-- 
adamw