Re: communication/information between forward-looking projects [was Re: Some info (Ref: GSOC 2008 advice)]

[Dave Neary <bolsh gnome org>]

Behdad Esfahbod wrote:
> On Mon, 2008-03-03 at 09:50 -0500, Luis Villa wrote:
>> Second, if there is any one thing which is agreed on, I think
>> it would be less 'no GNOME 3.0' and more 'no 3.0 for the sake of 3.0'.
>> In other words, if someone can come up with something that truly
>> improves the *user* experience (*not* the developer experience) then
>> that should at least be considered as the core of a new GNOME.
> 
> What most of the developers agree on is that we don't want yet another
> major version bump that means all apps using GNOME platform need to be
> practically rewritten, like the gtk 1.2 -> gtk2 transition.  You still
> see great apps using the old gtk eight years later.  That's what we know
> wouldn't work.

And this is a great synthesis of one part of the problem.

Some developers think "major version bump" is synonymous with "breaking
ABI and API stability". Many users think "major version bump" is
synonymous with "significant new features". These ideas are loosely
related, but not at all the same thing.

If you look at GNOME 2.0 and GNOME 2.22, they're nothing like one
another in terms of user experience. GNOME 2.22 is not just a polishing
of GNOME 2.0 - any commercial software company would have us on GNOME 4
or 5 at this stage (born out by the distros example of bumping major
version numbers in the same way Queen Elizabeth took baths - once a
year, whether they need to or not).

The first step in releasing GNOME 3.0 is to dissociate the in-built
assumption "will break everything" from the version jump.

Cheers,
Dave.

-- 
Dave Neary
GNOME Foundation member
bolsh gnome org