Re: Goal for GNOME Mobile?

[Dave Neary <dneary gnome org>]

Hi Ben,

Ben Leslie wrote:
> 1/ GNOME mobile is an umbrella banner for libraries and infrastructure
> that is related to GNOME and useful on mobile/embedded platforms.
> 
> 2/ GNOME mobile should be 'something more' with some vague ideas about
> what 'something more' means. The closest to an articulation of 
> 'something more' appears to be an 'application framework' or 'mobile phone
> stack'.
> 
> The problem with #1 is that it does not seem to be inspiring enough, or
> visionary enough, for some people's taste.
> 
> The problem with #2 is that there are already a bunch of other different
> projects or efforts for these things, and it is not clear what the world
> gains by GNOME mobile being one of those things as well. (I'm being 
> intentionally vague about what 'these things' are, but think product/
> application stack/software development kit/standard/etc).

Completely agree with your analysis.

> Anyway, if I had to come up with the 'GNOME Mobile vision' it would
> be something along the lines of:
> 
> 'GNOME Mobile is a set of GNOME and related open-source technologies
> that have been optimised for mobile and embedded devices'

Disagree with this. I would definitely not say that GNOME Mobile is
specifically optimised for mobile & embedded. The whole point is to take
*standard desktop components* and use them in mobile.

We are, as you say, optimising those components for mobile (or we should
be), but we're not making "GTK+ mobile/embedded edition", and I don't
think it would be desirable to do that.

> I would follow up with something a FAQ along the lines of:

<snip>

I like the FAQ idea, and mostly agree with what's in it. One thing which
is still missing from this all is the call to action. The stuff that
needs to get done for mobile through GNOME Mobile.

I'm going to kill the analogy completely, but bear with me...

GNOME Mobile is the builder's raw materials. We're the lumber, tools,
cement, bricks, paint, plaster, etc. Poky Linux builds houses with us,
LiMi builds houses with us, Maemo builds houses with us.

It's easy to convince a builder that he needs to have good raw
materials, and it's easy for him to justify paying for them. It's not so
easy convincing him that he should be investing R&D dollars in making
better bricks or cement. He just expects a certain quality of product,
and is ready to pay a fair price for it.

So what we have is a mindset issue, where the people holding the
purse-strings are more used to thinking of application platforms and
libraries in this way, but what we want them to do is invest resources
in making the raw materials better and then get an endless supply of
them for free (I told you I'd massacre the analogy), rather than just
buying what they need.

But there's no shame in being raw materials, even if manufacturing
bricks isn't as sexy as building houses. What we need is the structure
and form which allows those builders to co-ordinate their raw materials
R&D in a co-ordinated way.

In the corporate world, this typically happens by setting up a
non-profit or university research department where everyone is a member
or benefactor and pays a lot of money every year, and that body then
hires the people to do the work. That doesn't map easily onto our
problem space of distributed community development, but perhaps we need
to think of some way to structure investment in the GNOME platform that
does make sense to our community?

Cheers,
Dave.

-- 
Dave Neary
GNOME Foundation member
dneary gnome org