At Nokia World when presenting the N900 they said it was step 4 of 5.
I've also seen it quoted many times in interviews with nokians since.
If people missed this which I don't know how they could b/c every mobile blog on the net quoted it and still are when writing about n900 then you must of done lack of research about the product before hitting purchase.

Well, I for one don't think it's that simple as was discussed on that thread. The governance is only one thing... another thing is that the difference between even the Android and Maemo base platform may not be that great if you count in the Google Apps which do provide some pretty basic functionality as well.

Device manufacturers can add even more non-free software on top of that and/or modify the base Android system without releasing the modified versions anywhere because of the Apache license used for all of it.

Also, if you look at the rest of the apps installable for the platform, most of Maemo's apps seem to be open source while Android has about 10 free software apps available.

Time to weigh in on this thread as maemo.org distmaster and as someone who has been trying to put Fremantle on N8x0 along with lessons learnt from this experience and why things didn't pan out well for Mer and why I think the situation is different than with the N8x0.

There's a load of different questions in this thread, which can be summarised as:

1) Technical: Will the Harmattan 'distro' run on N900 hardware?

Let's start with a comparison. The key difference between a N8x0 and a N900 is the fact that the N900 has a GLES2.0 capable chip and a N8x0 has a GLES1.1 capable chip. The centerpiece of the N900, the desktop, requires a GLES2.0 capable chip to run smoothly and properly. Another problem is the fact that the SDK of Maemo5 compiles exclusively for ARMv7 and up, while N8x0 has ARMv6.

From a distribution point of view, it is simply later versions of most of the same packages that made up Diablo on your N8x0s with some added functionality, new APIs, new kernel. But the desktop is the dealbreaker as well as the application binary incompatibility.

As someone said earlier in the thread from a summit slide 'Harmattan / Maemo 6, including OMAP3, OpenGL ES, capacitive screen support and multitouch UI'

From a purely distribution (speaking of source packages) point of view and SDK point of view, I'd like to hope that the distribution would run. Since they're both OMAP3, the applications would probably be compatible too.

Why do I think it probably would? Because we're already running many of the Harmattan versions of packages open sourced at http://maemo.gitorious.org in Mer. On the N900.

2) Usability: Will the Harmattan UI be usable on N900 hardware?

The general question is if UI relying on multitouch would be usable. Who knows. But the argument I keep repeating to myself (as this matters to me in Mer and it's future) is that a developer will have to run their SDK on a single-touch desktop machine and hence there will be a way to deal with single-touch in the UI.

3) Support: Who will maintain Maemo 6 on N900? Who do I complain to when something goes wrong?

Would it be a Nokia released OS (with all the bells, whistles and support), or something maintained by community or by a third party.

Is community capable of providing it if they had all the assistance needed including relicensing of some components?

From my side, I'd really hope we would be capable of this in the community.

The foundation for this is already in Mer and while Mer isn't that functional yet (bad choices and GLES turning out to be 1.1 on N8x0), for Harmattan, we wouldn't have to work for a year to get to this point again - we have something we can drop packages into and see how well it works. Hopefully we can start doing this in alpha SDK, just like we did with Fremantle pre-alpha and alpha SDK.

Someone once said: "Mer is also a proof point of the openness of the Nokia devices designed for Maemo. For me one of the beauties of Maemo is that you can get rid of it. It is not imposed".

We are capable of taking matters into our own hands and help make the Nokia devices do our bidding and -replace- our phone OS with our -OWN- OS image. And even have a collaborative relationship with the original OS makers while doing it, working out licensing terms etc instead of every day living with a C&D risk.

Instead of looking at Nokia only providing support for a certain time after device release, look at how they're trying to help us take matters into our own hands. They're empowering us.

And they're already now taking steps to put Harmattan pieces out in the open on http://maemo.gitorious.org and http://qt.gitorious.org and developing them openly. They already take patches, too!

So you're fine with receiving no more OS bug fixes/enhancements, except for major issues? Let's see if you still agree with that statement once you're affected by something that won't be fixed without spending another $600.

None of that has to do with actual opennes of the operating environment though. All of that has to do with the ecosystem.

Take for example NITDroid. That is a truly free Android port. Note, many found it "useless" because it does not, and cannot (unless they got around it) include the "Google Market" which is a proprietary app.. however - Android is not the Market, Android is the Operating system. All apps that are *on* the market should run on the NITDroid, but getting them is problematic because google has locked down the Market itself.

However, other market systems *do* work for it.. that also work on an android device.

Now your turn - wheres the Maemo-port done by the android community for their phone?

ETA: Also, there are more than 10. I only use Free software and I have quite a bit on my G1 phone.

Hi all, I'm new to all of this "forum" stuff, so please excuse me if I make an error. I've just finished an 18 month pay monthly contract with O2 (I live in the UK) and I'm due a handset upgrade (some are free, some aren't depending on the cost of the handset). My relationship with O2 is ending as they won't touch the Nokia N900 with a bargepole. It seems they are too up Apple's backside with the iPhone (most of the people that I know who have one, reckon that it's the dog's.........) but I've decided NOT to run with the pack and take my valuable custom elsewhere and go with Vodaphone for 24 months on the understanding that I get a N900. I've heard such good things about this phone - the OS, for one thing, which is Linux based, if I'm not mistaken? However, it seems that no matter how many good reports I've read, these have been really shaken by some of the things that you guys have been putting on the last nine pages of this thread. I'm not even going to try and quote some of the things, but basically, what I've read is that it will not be supported by Nokia, OS wise, and maemo 5 will be more or less out of date in the next few weeks/months? Am I right? Do I bother dragging myself into Vodaphone tomorrow to get my holy grail, or do I wait? Or god forbid, do I get an iPhone? Sorry about typos - I get quite emotional

Go on a store and play with it. If you like it and it does what you need it to do, then purchase it.

Also, it can be a great device if everyone contributes instead of creating more whine posts.

Thank you STskeeps for weighing in and for this post.

So in all honestly then... the entire maemo base (any version) was available and open for you to recompile and use for mer? Or did you have to replace components with your own software due to licensing/proprietary concerns??

(not including drivers).

This is more for my own curiosity than for relevance to the actual thread.

I don't really see any reason why Maemo 6 cannot be supported by the N900, except to drive the sales of the M6 device...

A public statement of M6 support would do WONDER for the N900 and the Maemo platform. As a potential buyer of the N900, I'm pretty much sold, until I read about M6 and its uncertain future on the N900.

People should remember this announcement which said:

"This early release comes with an invitation to build variants based on Maemo 5 compatible with existing hardware like the N800 and N810. Maemo SW can't promise commercial quality for such configurations but through maemo.org we are able to collaborate at a community level with technical support, license changes and code."

Our hybris was also that we wanted to start out with a fully open source base instead of cutting corners with closed source things. Also, we went in directions that made us incompatible with Maemo5 for some things as well. So in some ways, in all the organisation build-up and hacking day after day, we forgot to remember about the goal - having a usable OS by the user. In practice, we didn't ask for that much relicensing and those we asked for, we got.

But the invitation still stands, I'd say. If we needed recompiles, I'm sure it would be possible to work it out somehow - especially with my role as distmaster now.