Re: Is 2.7 being tested?

[Elijah P Newren <newren math utah edu>]

On Thu, 2004-08-12 at 09:27 +0200, Murray Cumming wrote:
> A question for people who are keeping an eye on bugzilla:
> 
> Are we getting the usual number of bugs for GNOME 2.7? I fear that far
> fewer people are testing 2.7 than tested previous versions, because
> garnome is not really being supported/suggested this time, and we don't
> have a full jhbuild-from-tarballs solution yet (Nobody's fault. I meant to
> figure that out myself, but never found the time). Also, we've had less
> releases, at less-predictable times.

I'm having a really hard time trying to answer that.  I tend to triage
in spurts (especially since I've split what time I do have into
assisting more projects this release cycle than in the past).  Plus we
had several months with bug-buddy submissions being queued instead of
added to bugzilla--and instead of working more evenly on new submissions
with occasional work on older bugs, I've spent more time on clearing out
these bugs that appeared all at once.

However...  I believe you are right about there being fewer testers this
release than previous ones.  Luis said that there has actually been a
trend since 2.0 that each release has fewer testers than the previous.

Although my guess is that we have fewer people testing, I think it might
not be far fewer.  In particular, the fact that gnome-love had a large
effort a few months ago to assist people in learning to compile from CVS
using jhbuild has probably helped.

> And what are you general impressions about the quality/stability of 2.7 so
> far?

We haven't had bug days to try to find bugs that should be marked with
TARGET2.8.0, er, I mean, Gnome Milestone=2.8 (the TARGET keywords are
gone now).  So this is also hard to answer.  However, I think we don't
have any big worries here, because I don't think any major destabilizing
changes have occurred--without such changes, of course, code tends to
become more stable over time.  So I think 2.7 should be more stable than
2.5 on that general premise.  It would be nice to have more info to back
that up, but all I can really say is that for the subset of stuff that I
use (which unfortunately is probably a smaller and different subset than
the normal user), that seems to hold true.

Sorry this isn't any more helpful,
Elijah