Re: Goal for GNOME Mobile?

[Paul Cooper <pgc linux intel com>]

On Mon, 2008-11-10 at 14:08 +0200, Quim Gil wrote:
> 
> ext Paul Cooper wrote:
> >> "This is not GNOME", many said in the past and nowadays.
> > 
> > But for GStreamer and Telepathy we identify them as part of the GM
> > platform, how are we saying they're not part of GNOME?
> 
> Mmm you are right that the diagram says so. The "Technologies in
> Incubation" include more freedesktop.org projects, also true.
> 
> However, if we go and ask to the GNOME community (including many
> maintainers of those projects), what answers we would get? GStreamer,
> Telepathy and such are out of GNOME for the average GNOME contributor.

OK, without something more specific I can't see this. As you mentioned
if you look at the schedule and attendees of GUADEC, if you read p.g.o
and look at what's in the platform (both desktop and mobile) and I can't
think how the 'average GNOME contributor' would think many of these
things are not part of GNOME?

> Now, perhaps the
> http://library.gnome.org/misc/release-notes/2.24/#rnmobile released some
> weeks ago indeed changes the situation and my worries are already part
> of the past.
> 
> 
> > BTW Clutter isn't part of freedesktop.org.
> 
> Yes, that was unaccurate from my side. Apologies. The example of Clutter
> is still valid, though: a technology born in the GNOME context that
> might or might not play together with GTK+.

Huh? Clutter has the same dependencies as GTK+ (glib, pango, etc) and
the play perfectly nicely together (even more so once GTK gets offscreen
rendering)

> >> GTK+ is essential to the GNOME project but, is it the cornerstone to
> >> sustain the whole building? In the desktop years it was, and this is why
> >> "This is not GNOME" was a clear assumption: you can run GStreamer,
> >> Telepathy and etc with other GUI toolkits on top, they are not dependent
> >> on GTK+, therefore they are not GNOME.
> > 
> > Are you suggesting we should somehow make gstreamer, telepathy, etc
> > dependent on GTK+? I can't see how that would be good for either.
> 
> See my following paragraph:
> 
> >> Is this the best strategy to fulfill the GNOME vision in the mobile
> >> context and the forthcoming years, though? But what is the GNOME vision,
> >> again? You see the trap.
> 
> The trap of keeping the GNOME identity tied to a specific technology
> (GTK+) as a best replacement for a lack of identity and vision (*). This
> approach worked during the first decade of GNOME, I just wonder whether
> it can be kept for the mobile decade.

That we're lacking identity and vision is a reasonable argument and
subject for another (wider) thread. Somehow mixing GTK+ as the problem
doesn't make sense to me.

> (*) Benchmarks to test: can you explain GNOME without getting into the
> details of its building blocks? Can you think of a GNOME based product
> skipping GTK+?

Of course you can explain GNOME at a high level without discussing the
building blocks - but why wouldn't you want to talk about them if more
detail is needed / wanted.

No current GNOME based product skips GTK+ because there hasn't really be
any other option. Clutter maybe an option for some devices in the
future, but I really don't get the point - what about thinking of a
device that skip Glib, or Pango, or DBus?

> > What does 'over-covered by corporate backing' mean?
> 
> An impressive list of corporate sponsors supporting a reasonably
> structured and resourced GNOME Foundation, compared to the ad-hoc
> reality of freedesktop.org and some of its critical projects. Maybe I'm
> again wrong and freedesktop.org doesn't need additional help or
> attention, but if so the GNOME Foundation would be an obvious channel to
> support them.

Ah, so what you're saying is that corporate interest has been focused at
the higher most visible end of GNOME, to the detriment of the lower
level freedesktop.org plumbing?

> Again comparing to the Desktop decade, on that time it was important to
> extend the GNOME concept to the application layer making sure the main
> features were covered: browser, email, etc. However, in the Mobile
> decade applications become less of a problem since many parties seem
> interested in providing them. The attention needs to turn more on the
> inner layers, since the user experience and the success in fron of
> androids and etc depends a lot more on solid, well maintained and
> innovative building blocks.

No arguement here - I agree that we need to provide a solid, well
maintained and innovative set of building blocks. With the first release
of the GNOME Mobile stack I think we're delivering on solid and well
maintained. Now we have that release cycle under our belts we need to
start including some of the incubator stuff which should be adding the
innovation (although I shudder every time I type that word, it's so
overused). I would note, however, that Android does not necessarily sit
on solid, well maintained*, or innovative building blocks (just ask
Lefty, if you have a while to spare ;-) and the jury is still out as to
whether it's a success, either as an open-source project, or a product.

Paul

* well maintained, in the sense we would say an open source project is
well maintained. It's only been released a few days, let see how they do
over time
 

-- 
Intel Open Source Technology Centre
http://oss.intel.com/