Re: 2.4 timetable

Tomas Frydrych wrote:
>
>
> I would like us to release 2.5 week on Fri, July 15 -- I think the HEAD
> is as stable as the current stable release, and we have made a good
> inrows into the bug count. If you feel there are reasons not to release
> next Friday, then best if you indicate so in the bugzilla.
>

There are at least 15 open regressions in Bugzilla that were introduced
during 2.3.x alone; there are 81 known bugs that cause AbiWord to hang,
crash, or lose data; there are 897 known bugs in total, although that
number may be a bit high due to every bug being marked as new since the
recent Bugzilla upgrade.

CVS-HEAD is not stable enough to soon become 2.4.0 or even replace
2.2.8, and that fact is made worse by having had only two 2.3.x
releases. This feature freeze hasn't seen enough bug fixing to say it's
made good inroads on the bug count, especially considering the fact that
a lot of fixes have caused regressions or introduced new bugs.

I think it's quite clear that I'm against having 2.4 so early and it's
not solely because of the most recent regressions; it's because AbiWord
simply gets buggier with every release. It may not crash or corrupt your
data as easily as it used to, but AbiWord still lives up to the buggy
reputation it's earned. Furthermore, feature freezes don't work as well
as they should because they're either ignored or cut too short.

It was agreed that some serious bugs that blocked 2.2.0 would be fixed
during the early parts of 2.2.x and block the branching of 2.3.x. Those
bugs are still valid and it appears that a quick 2.5.x branching is
planned. I'm quite aware that 2.3.x was branched early due to lack of
coder morale, but at some point the ever growing bug count needs to be
addressed if AbiWord wants to seriously be considered a stable and
polished, production-ready word processor. If it's not done during this
cycle, it needs to be done soon because features new to 2.0-2.4 are less
than useful due to the numerous bugs with each and some haven't even
been tested.

Now, I realize a few of the new features are in the form of plug-ins,
but if you're going to tout them as highlights of 2.4, you really ought
to make sure they've at least been eyeballed. And with only two limited
releases, I can guarantee this hasn't happened - 2.3.x really needs more
testing and bug fixing before 2.4 can be released. Sorry for the length
of this email, but these things definitely need to be considered before
a release occurs.
Received on Tue Jul 5 02:39:02 2005