GNOME Mobile: here it goes

02Oct07

A few balloons, by mortimer

Only 48h ago I was wondering GNOME Mobile: where are we? I got fast & precious feedback via Earth, Wind & Fire (can’t link to the improvised lunch shared with 6 GNOME+maemo hackers caught in the SMS corridors). Also some thoughts appeared in my laptop screen, surely pushed by the GNOME vivid source code itself.

I had my OpenMind talk already. About 70 (?) people in the room, a majority of them confessing being GNOME users and many even conneisseurs of the GNOME Mobile stuff itself. I was frank with them as I will be with you in this report written in the way back to Helsinki.

So, what is _really_ this GNOME Mobile initiative? And where are we.

This is about People. Different people than in the “normal” GNOME project? No, not really. New people? Hard to say, probably some nor more visible with the mobile agenda in place.

It is also about Code. Different/new code? Again, not really, or hard to say. The mobile stack is more or less a subset of the full GNOME stack. Again, some secondary players can de more essential here but the mystery is precisely to start from the same and known GNOME software stack.

User Experience, another key point. Are we delivering one? Euhm, no, not just one, and not even many in an official way. But there is some interesting stuff going on that you can grab and test even if it hasn’t got a “GNOME Mobile” label on it.

And finally Mobile, the critical aspect in the Mobile initiative. Do we have this in place? YES! Definitely. Thanks to this effort now everybody is aware of the mobile assets and agenda in GNOME. Only this makes worth all the rest. At this point.

What we have got so far

We presented the project last May and since then we could say that any new player in the mobile space can’t be unaware of this initiative. Good.

In that presentation we had three important things in place:

  1. Stack: the basic diagram showing agreed programming languages and components. A reference for alignment.
  2. Organizations: small and big companies, upstream projects, some individual contributors and the GNOME Foundation. Critical mass for real networking.
  3. Devices: initial collection of quite diverse hardware powered by GNOME Mobile software. Real products, much more than a proof of concept.

Then we also had this track of meetings and sessions with a mobile focus in GUADEC, showing a clear interest and willingness to move forward. Not clear conclusions, though. Nor much GNOME Mobile generic activity or discussions since then either, although the mobile agenda is penetrating in the discussion and work of the projects involved.

My guess is that we have been either on holidays, busy, waiting for somebody else to do a move or a combination of these factors. Things are clearly moving forward in several fronts, but I wonder if somebody is following all those and getting the wide GNOME-centric picture.

What we have learned so far

[The wonderful picture of the balloons goes in fact here. 🙂 ]

(Potential) Software freedom land is wider than we thought. Yes, we kind of knew but now we know for real. If liberating servers, PC boxes and laptops was seen as a huge task, by including the mobile computers (i.e. phones) the numbers grow almost exponentially. We are talking about the real mainstream, devices normal people use all the time.

We are talking mostly about the same stack (although perhaps this might change for certain components whose features/constraints are really different in mobile devices i.e. evolution-data-server instead of the whole Evolution pack)). We want to use the same stack as far it’s possible and makes sense.

We are talking mostly about the same benefits. Bugfixes, new features and other improvements in the GNOME “desktop” components benefit the GNOME mobile platform. The other way round works as well (something that was stressed in GUADEC a couple of times).

But it is evident that there are different priorities, and this is where most frictions might come up. Touchscreen interaction, bling-bling or power consumption are seen differently from a desktop or mobile point of view and in many cases they require different people and different compromises between those managing patches & commits.

The inclusion of the mobile space also changes the context, the players and some of the well assumed principles of the desktop+server perspective. For instance, Microsoft is not in a monopolistic position and Linux is not an almost automatic driver of software freedom. New paradigms, new opportunities… and probably new strategies too. More than one big cake, now we have a panorama of different cakes that are somehow combined and differentiated. The whole mass of cake is much bigger, though.

Another lesson is that brand matters. Before “GNOME Mobile and Embedded initiative” it was hard to fight in some arenas in order to get attention and serious attraction. We made a move, we created a meta-brand supported by known brands (from Nokia or Intel to GTK+ or Gstreamer) and now it is just easier to get approvals and decisions in the directions we point.

Questions that (perhaps) will need an answer

This is a tricky part, These are the most relevant I have collected.

One size fits all? Can or should we find out how to make the whole GNOME stack and project progress for the benefit of the classical desktop and the new mobile interfaces? Is this combined effort a strenght or an extra weight making difficult real progress in both areas?

A problem here for the typical free software developer is that hardware (and its manufacturers) is still too tied to the software platforms and even the frameworks for developing applications. You can buy a PC, install your preferred free OS and start developing, but you can’t do just the same on a mobile device as for today. Therefore we are more tied to hardware, and also tied to the efficient and transparent collaboration between the companies producing or distributing that hardware and the operating system components needed to work with it.

Push or fork? Specially for the companies having to deliver according to internal plans, following the full GNOME processes (still quite desktop-centric) might be a pain, and even a reason to just use something else. Or fork to grant an own path based on own requirements. Hopefully at this point we have seen enough examples about how expensive forking might result, but still in some cases might be the most feasible option. Sometimes the fork might consist in using component B instead of A, keeping at least each component mostly clean. In some cases the decision might be to hack component A before waiting upstream acceptance. The GNOME Mobile initiative should be a space to discuss and find the best solution for each case.

The push/fork dynamics apply also to the way the GNOME Mobile agenda is discussed and agreed. Do we need to have (much) discussion specifically about GNOME Mobile or is it the best way to just include the mobile agenda in the GNOME projects? In theory the second option is preferably and preferred. However, sometimes the project maintainers and the mobile pushers might find a harder time agreeing on things, and those not in controls of the source repositories might be tempted finding a shortcut somewhere else.

I want to add my own big question. All this refers mostly to software, but the success of the GNOME Mobile platform is just relatively conditioned to good software. The mobile thingy brings also a new generation of services and communities (call it social networks or customer groups, if you prefer). A great free operating system not being able to support these services offered to these communities will fail as a mainstream option and will be relegated again to niche products and cult computer usage.

The good thing is that the GNOME Online Desktop discussion/project goes precisely in this very same direction. Mobile devices are called to be permanently online some day, as they are called to squeeze the equation between user time, geographic space and impulse. Another case of common stack and benefits across the GNOME project that we should take from a wide perspective even if the priorities of desktop & mobile might not be the same.

(end of presentation)

Question 1: what does the software stack diagram really mean?
Answer: it is basically a marketing tool to tell that these are the components and languages we are focusing and recommend. This doesn’t mean that all the organizations or products involved use them all or in the same way, but at least we have a picture for common alignment.

Question 2: where is the progress happening now?
Answer: hard to tell, since the discussion is happening mostly inside each project. At least I don’t have a vision of what is really going on since I not following all the projects. If you look only at the GNOME Mobile lists you will find almost no movement. I guess we are busy these days.

Comment: in some industries silence one year sometimes means many new products next year.
Answer (more or less): I’m not worried about the GNOME Mobile related activity in projects or even collaboration between organizations, community and companies. Lots of things are happening hre and there. It is only that with my GNOME Foundation shirt and a selfish GNOME-centric point of view I don’t see clearly what is going on or where are we heading. I’m not sure whether this is good or bad, but overall I’m not worried. The initiative is good, has given some fruits and I expect more to come.

EOF



3 Responses to “GNOME Mobile: here it goes”

  1. i LOVE your pictures, man!🙂

  2. I find it interesting that in the photo of the presentation, the bubbles with the programming language names are shaped like guitar picks. Is that how you see programming languages: devices that enable you to rock out?

  3. 3 qgil

    Thanks, andre. It takes me some time to find the pictures since almost none of them are mine. I go to flickr performing searches filtered by CreativeCommons licenses. It’s my humble way to help featuring all the visual creativity and beauty available out there.

    fraggle, Tuomas Kuosmanen did that graphic. I have heard the guitar pick comment before but I don’t know if that’s the purpose. Jeff Waugh says “rock&roll!” quite frequently, so you might have a point. 😉



%d bloggers like this: