Re: Roadmap for GNOME 2.0
- From: Havoc Pennington <hp redhat com>
- To: Federico Mena Quintero <federico redhat com>
- cc: gnome-hackers nuclecu unam mx, gnome-devel-list gnome org
- Subject: Re: Roadmap for GNOME 2.0
- Date: Thu, 30 Sep 1999 12:00:34 -0400 (EDT)
Hi,
- I like that Federico is going to be in charge of gnome-libs, everyone
please be nice when he does mean release-manager things. :-)
- it should be made clear that after we freeze, the release
happens "when it's stable," though a target date is nice
just so we know when we'd like to release and how
much time we are leaving for QA. That is the 1.5 months
for QA of gnome-libs is just an estimate.
- it makes sense to me to _release_ desktop environment and
developer toolkit at the same time; the reason is that
lots of desktop environment bugs will really be library
bugs or may require small API tweaks.
My proposal is that we feature freeze gnome-libs after 2-3 months,
and commit to "NO features, and source compatible unless we really,
really need to break it to fix a bug."
After gnome-libs is frozen we could release beta tarballs, but I think
making the actual release before finishing desktop environment will
just lead to gnome-libs creep. This time things should really, truly be
frozen after the release...
- We should have an initial (maybe 1 month) period of gnome-libs
development without porting any apps to the new gnome-libs.
This is because the new APIs will be in a state of flux,
and even the build setup (gnome-config) won't necessarily work,
so if we don't have this period then the whole CVS tree will
break.
2 weeks may be enough, hard to say.
- I think 5 months desktop environment devel and 3 months QA is more
conservative than 6 months devel and 2 months QA. I'd be in favor of
more conservative.
The release goals look very nice, I'm convinced that we can complete them
on time if we stick to them.
Can I make a final plea not to break things that work? If nothing else, we
are going to have to be running CVS panel and gnome-terminal and so on
over the next months, and if our desktop environment is broken we will not
get much work done.
I don't think Federico's current document has any major breakage planned,
so we just have to stick to it.
Havoc
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]