Re: GNOME: lack of strategic roadmap

[Andrew Savory <andrew savory limofoundation org>]

Hi,

I'd like to jump in here with some data points. Rudely cc'ing GNOME  
Mobile list so the right guys see this.

On 22 Feb 2010, at 22:52, Philip Van Hoof wrote:

> On Mon, 2010-02-22 at 20:27 +0100, Dave Neary wrote:
>
>> Juanjo Marin wrote:
>>> * GTK is losing popularity. It is perceived by a lot of people as  
>>> old
>>> and difficult. I think we need any kind of action on this area  
>>> because
>>> is a cornerstone issue. Less programmers means less applications and
>>> contributions. We need to care of our platform users in the same  
>>> way we
>>> care of our desktop users. Some people has pointed this in the  
>>> past, eg
>>> [1]
>>
>> Perhaps the fact that GTK+ is seen as a cornerstone issue is a
>> cornerstone issue... there's no specific reason why GTK+, FLTK or EFL
>> would do the job just as well of providing a toolkit.
>
> I agree.

If any of you were at Mobile World Congress last week, you'd have seen  
a Samsung demo on the LiMo stand: a phone running EFL.

You'll also have seen, amongst others, the Vodafone 360 H1 and M1  
running GTK.

Meanwhile, GTK remains a part of our latest approved specifications.  
This means that GTK is expected to ship on all R2/R3 LiMo-compliant  
devices. (See http://developer.limofoundation.org/images/stories/pdf/100106_PUB_LiMo_Platform_Architecture_FINAL.pdf 
  for a somewhat dated architecture diagram).

So: just because one mobile platform has switched to Qt, doesn't mean  
they all will. Though I think it's fair to say this is under  
continuous evaluation, like any part of the stack, so GNOME cannot  
rest on its laurels.

> That's why I made my earlier comment that our community itself isn't
> negative or hostile towards commercial mobile ecosystems.

No more than they are hostile to any other commercial ecosystem :-)

The simple truth is large corporations STILL struggle to understand  
how to engage with open source projects, and with the commercial  
services companies around those projects. Patience and understanding  
is needed on both sides.

Shameless plug - my talk from FOSDEM this year outlines some of these  
issues (though the slides possibly don't stand on their own without a  
transcript): http://www.slideshare.net/asavory/limo-platform-and-mobile-linux

>>> * It seems we have lost the mobile battle. Can we do something  
>>> about it
>>> or simply retreat?. I like the idea of creating more components  
>>> and some
>>> of this components can be added to the GNOME mobile platform.
>>
>> Have we lost the mobile battle? It certainly appears that GTK+ has  
>> lost
>> the mobile battle, but all of the hard work that GNOME hackers have  
>> put
>> into the middleware platform and components like Gstreamer, Dbus,
>> Telepathy and Pulseaudio are now cornerstone parts of both the free
>> desktop and the mobile platform.
>
> In mobile we're doing pretty well at the middleware segment.
>
> But indeed ...

It would be great to see more focus on the full mobile stack, such as  
the work of emdebian or hackable:1 or freesmartphone. Seeing optimised  
versions of the GNOME Mobile stack running on reference devices would  
be awesome.

> We are seeing a lot of innovation in middleware, though. I am very,  
> very
> pleased that the GUADEC organizers have put a focus on metadata in  
> their
> call for papers.

Perhaps we should reach out to the mobile and embedded community and  
ask them to contribute e.g. "how to get GTK running on a smartphone"?  
Getting a few of those guys over to GUADEC might stimulate some  
interesting conversations ...


Andrew.
Open Source Manager
LiMo Foundation