Re: GNOME 2.0 Schedule

[Havoc Pennington <hp redhat com>]

Iain <iain ximian com> writes: 
> And if we're releasing gnome 2.0, but it's not going to be any better
> (user wise) than gnome 1.4 (and lets be honest, it's not), and then
> releasing gnome 2.0.1 (or even 2.2) a few months later, then why not
> delay the gnome 2.0 release by a month?

Don't worry, I have little doubt it'll be delayed a month with no
extra effort expended on our part. ;-) j/k, sortof.

My attitude is: we decided ages ago on the approximate release
date. To get big user features by this date we had to do less goofing
with libraries. We didn't want to do that. So now we made our bed, and
we have to lie in it. The libraries are the feature we decided to
implement. Let's not be ashamed of that; it's what this release is
about. The libraries are good. They will help us make beautiful user
features faster.

(Nonetheless, I do think replacing the control center, decent keynav,
lack of flicker, AA fonts, accessibility, and a few random things that
will inevitably happen, will be a noticeable user improvement. Not
huge, no, but enough. Plus hopefully we can have lots of stability and
a reasonable release date, by aggressively punting any attempts to add
features beyond these. Releasing the software is a valuable
feature. As is freezing and stabilizing. Users appreciate both of
those things quite a bit. I think we can do quite a nice GNOME 2 given
the currently-proposed schedule, it will just take a good effort all
around. Honestly most user complaints I see are about platform
features like AA fonts, and about bugs. So GNOME 2 can address a lot
of that.)

Of course it makes total sense to start working on user features today
even if they won't necessarily make 2.0, life isn't all about porting
and bugfixing. Everyone should have a fun project that is utterly
pointless/experimental and/or absurdly long-term.

> Or another option is: We release the gnome 2.0 platform when we are
> happy with it and then when we've got what would be this mystical
> wonder release you suggest for 2.0.1 (or 2.2 (heck, talk about
> version inflation)) release it as the gnome 2.0 desktop.

I think this is needless version number metaphysics. The "2.0" means
"platform revision," and "3.0" will come at the next platform
revision. Simple, straightforward, easy to explain. "When we are happy
with it" has never happened, ever, for any software project I've ever
been involved in or heard about...

Personally, I start my projects at version 2.3, just to keep the
versioning angst under control. Very therapeutic. ;-)


Don't be afraid of releases; they're good for you. All experience
suggests that frequent releases lead to more and better software than
constant release-stalling. The more often a project releases the
better off it is, no question. And there is really no reason we can't
make the currently proposed date with something respectable.
Let's just do it.

MHO,
Havoc