Re: very rough pre-gep tentative new modules list

[Michael Meeks <michael ximian com>]

On Wed, 2002-09-18 at 10:52, John Sheehan wrote:
> >Unless I'm misunderstanding something, this really isn't an issue at
> >all. Companies can just as easily pop stuff in and out as they please -
> >I know Sun, for one, has done this - and I'm pretty sure RedHat do this
> >too.
>
> Yes, companies like Sun can choose to omit applications from their release.
> However, once they include an app in their supported core, they cannot
> easily pop it out again or replace it with a different current flavor app.
> Having a minimal minimum Gnome Desktop that changes as a result of clear
> and long-term planning is a critical issue.

	Precisely - indeed, one role of the Gnome foundation is to define 'What
Gnome is' - clearly that has to be something that is sensibly defined,
and not changing on a 6month basis - more like a 6 year roadmap basis
would be good ;-)

	Clearly that's difficult to do; currently the best guess we have is the
core development platform will continue to provide a certain set of
APIs, and we're shrinking and cleaning those up.

	Also, things like Nautilus, Panel etc. seem to be with us on a much
longer scale than things like 'totem[1]'; so - all that concerns me is
that we have some (sensible) division between things that are the long
term future that everyone has to work to maintain, and things that are
convenience throwaways so we can say "Gnome has XYZ feature"  ;-)

	So - I'm just concerned that in an attempt to 'encourage' projects, we
destroy a very useful delineation, and demolish a potentially useful
definition of "What Gnome is".

	Anyhow; incidentally - one advantage of the GEP - "Red tape etc." is
that the discussion period has an end - at which point we all compromise
and move on; so in fact, the lack of a GEP leads to protracted, vague
arguments, with no conclusions, clear flow of thought, nor decision at
the end. This has _far_ worse long term consequences than having a
lively, discussion with an end-point. Really !

	Regards,

		Michael.

[1] - Whether Totem is the right way to go long term, is of course
another (interesting) question, it's just an example.

-- 
 mmeeks gnu org  <><, Pseudo Engineer, itinerant idiot