Re: [Council] State of Maemo, Q32010.2
 

Sorry qgil, if this sounds a bit rude, but you don't know
about what you are talking. Just because an application uses
Qt does not mean it is portable. Dialer, calendar, photo viewer,
mediaplayer, addressbook, all these apps are connected, together
and with the backgrond components. If the reference application
uses newer versions of tracker, contextkit another
calender backend, media/audio framework, you can not just
take one application to port it to maemo.
I follow the development on the meego-maillist, forum and bugtracker.
I know that there are huge changes.
1. I don't think so.
2. maemo5 applicatios are what we have now and
MeeGo applications are far from useable.

If Nokia doesn't want to open their apllicatio, ok!
But I had just a simple question. Does this
"... ideas opening various parts of maemo .."
means anything new? Are there new ideas?
Or why did they talk about this ideas regarding the state of maemo.

nicolai

Re: [Council] State of Maemo, Q32010.2
 

Though this is mostly meaningless for owners of current/past Maemo devices and will remain so until we can actually buy a MeeGo device.

This isn't a terribly clear message in the context of this discussion. I get that Maemo is now "legacy" and if Nokia isn't willing to open anything up that's fine. If they are willing that's even better of course, but tell us if it's worth our time to continue pursuing licence change requests or not. I mean, even when the licence change request queue was active there was only one component opened out of it.

A completely open OS is a worthwhile goal in general, but I think we all realise by now that Nokia isn't Qi, Lemote or Openmoko. In practical terms there are two main reasons why people want some components opened:

1. Running non-Maemo distributions on Nokia hardware: this is usually about low-level core components like bme and libcal without which "porting" say Debian to an N810 is pointless.
2. Fixing bugs: this usually concerns high-level UI components like apps, control panel applets and the like. I suspect most people wouldn't care about the licence if Nokia fixed the bugs (some of which are extremely triviai) themselves.

(there is some overlap of course,, closed-source core components can be buggy too).

If you take a slightly more high-level view, these can be summarised in one reason: people don't like it when the hardware they bought becomes obsolete due to stale software. The best way to counter that is to make sure that the Harmattan device(s) will also be able to run "real" MeeGo 1.2 once it's out, as well as at least all future 1.x versions.

Re: [Council] State of Maemo, Q32010.2
 

Then why was the subject brought up again? I doubt bme will ever be OSS, mce is quite the same in terms of licensing, if the "user experience layer" can't be open sourced, and the underlying components are WONTFIX'd, what use is opening bug reports about closed components?

And what if I have a media player with a killer UI, that doesn't work with the current pulseaudio (just as an example), chances are if I tamper with pulseaudio, I'll break maemo's haptic feedback, maemo's phone application, and pretty much any other apps that are closed and can't be compiled against new libs.

Re: [Council] State of Maemo, Q32010.2
 

Going back to the comment about the Media Player and Calendar. Both could certainly be replaced, but the question is really whether it's worth investing the time and effort to do this as they will presumably be completely replaced in the next Harmattan/Meego release.

Having source available would make this significantly easier and cheaper (in terms of time) to do. OTOH I understand the point about the UI being Nokia's differentiator, though now that Maemo is an old platform, I'd have thought it's less important.

Certainly my view on this conundrum (stuff needs fixing, but who wants to commit to doing that when their work will not last all that long) is to simply wait for a working and fully featured (not in terms of sw, but certainly hw and the sw components that will still be closed) Meego installation for the N900 and just move over completely and then work on Calendars, Media Players, etc., etc. knowing that my work might be useful for more than a month or two.

Re: [Council] State of Maemo, Q32010.2
 

As far as I can see MeeGo and the n900 implementation of it will not be stable for some time - and that is before we have any meaningful apps for it (e.g. a media player!)

Maemo is now hamstrung because the improvements needed for some of the apps can never be applied (e.g. closed-source media player).

So where do we go from here?

At present, meego and maemo are mutually exclusive. We only have a "bridge" (QT libs) that will allow limited portability.

The "in-built" n900 maemo apps are showing their age a bit now and need some kind of revision - which is highly unlikely to happen.

I've been using mplayer for many of my video files because I find that the maemo media player does not play them smoothly (Mplayer also supports the most audio formats too). I often switch between symphonie and the n900 media player to play all my media files on my n900.

Luckily I am a software developer and I am toying with the idea of writing my own little frontend for mplayer - something I can use on my n900 and on my Linux PC at home- I know there are others out there but da**it I want to write my own!

What I'm trying to say is that maybe we should take things in a different direction?

Maybe we can replace stuff with our own qt-based apps?

I have no desire to install meego on my n900 because I don't see it having full hardware support in the forseable future.

Not to mention all the kernel and software hacks possible with the n900 and maemo at the moment.

I also do not like the "DRM" layer aspect of meego either!

Maybe we should accept that meego is still a very long way off and I don't think a stable version on a handset will be ready until mid-end of 2011 at the earliest. God knows when for the n900.

Maybe it is time for the community to start replacing the nokia closed-sourced components with our own. When meego is eventually ready we can then "port" these new apps across?

Re: [Council] State of Maemo, Q32010.2
 

MeeGo's core demonstration/reference apps aren't going to be as full-featured or mature as Nokia's own apps, presumably. And Nokia's apps will (almost certainly) be closed source.

So to get those working requires a level of ABI compatibility (and package fakery). Open MeeGo running (some) Harmattan apps might be doable - but I'm struggling to believe it'll be a polished experience, to be honest; and what if Harmattan's apps/SSU repo requires a valid Harmattan device IMEI? Downloading and using them on another MeeGo instance, even the N900, might be a license violation.

Agreed.

Re: [Council] State of Maemo, Q32010.2
 

IMHO, separating MeeGo and Maemo 6 will certainly make the situation similar to android/symbian, "Custom/Cooked ROMs" as they are called with have Maemo 6 apps on the N900, with the license completely ignored by those making the images.

Oh and as for replacing Maemo apps with our own ones, I discussed 1:1 copies of the apps (as qwerty12 has done for one of the applets) on #maemo, but it might be a waste of work/time when the apps are already there, but the source is closed.

Re: [Council] State of Maemo, Q32010.2
 

/me nods.

Yeah; and - no disrespect to Graham & GPE - it took Fremantle's (closed source) calendar for us to have a calendar I wanted to use on a daily basis on Maemo. That took 4 years and ended up with Nokia having to do it.

There are so many little things necessary to get a single good polished app, which doesn't eat battery, integrates nicely, and has a decent(ish) UX that you'd have to have something so utterly broken and a lot of committed developers to focus on that one thing to replace it, IMHO.

Re: [Council] State of Maemo, Q32010.2
 

Some people in community are working to improve Maemo.
Some people in community are working to create MeeGo and the future.

Both groups compete for attention and resources. This is unavoidable, but let's not allow that competition to draw us away from productive contributions.

Time is short enough. Season's greetings.

Re: [Council] State of Maemo, Q32010.2
 

I quite agree that the current Nokia apps will almost certainly look nicer in the short term, and it may well be that it's simply not possible to use the Harmattan/Nokia-Meego UI + apps for whatever reason which means we'll need to write/port our own replacements if we wish to use Meego (or even a more up to date Maemo).

I would hope that with a common goal of creating e.g. a media player, people could create (or more likely port across) something that's usable and able to be fixed by those with the skill (though it would be good to have a standard app in mind to avoid fragmentation).

Unfortunately it's generally quite hard to get people to focus on prettiness and efficiency (unless that's their job or their passion) when there are other cool features to be added or applications to add. Conversely, once something works, then the drive to do anything more often evaporates. C'est la vie with OS and having access to the existing source code might make at least some of this less difficult (prettiness and efficiency have presumably already been implemented for the most part, so we could just add features.)

Whether or not this will actually happen I don't know - if a new Nokia phone running Harmattan is released, devs will naturally gravitate towards it, which will reduce the number of people trying to write replacements for the N900 + plain Meego (for instance).

I suppose the best course of action would be for us to start writing apps now for the N900 running Maemo + Qt (i.e. portable to Meego once it's day-to-day usable), but then the question of which libraries to use, compatibility between versions, etc. raises its head.