Maemo + Moblin = MeeGo –> Join us!
BIG NEWS TODAY meego.com + video from MeeGo Steering Group + video from Nokia & Intel
This is a move that many had desired. On the paper it just made so much sense. However, not too many had really thought it would happen. Big companies have quite often big egos and it’s not easy for them to be humble and come together. I’m happy to work in a company that has been able to push and concede together with another company that has been also able to push and concede. 1 + 1 in this case will bring three and more. If these two megacorps could drop some of their priorities in order to reach a common ground, other players can do that too. Including you? (yes, you).
As a Nokia contact with the Maemo community and as an active community member myself, I expect this announcement to be big news for the maemo.org membership. Software freedom and more diversified devices have been two loud and consistent claims since the Maemo project was born in 2005. MeeGo brings that, and a lot more. In fact I think today is an historic day for the Linux and free software communities. Not only for seeing these two companies shaking hands in a Linux Foundation hosted project, but for the rest of handshakes expected to come after today’s launch. Now, where is your hand? 🙂
Intel and Nokia are two major investors and contributors in free software development. Both have big teams in house and collaborate with a wide variety of open source companies, projects and rock stars. Now they are combining strategies and resources in order to kickstart a free Linux mobile platform. Real code will come soon although plenty of it is already available in code repositories from Moblin, Maemo, Qt…
MeeGo is just like you would expect a free Linux OS to be: based on the standard Linux and Free Desktop technologies and developed publicly in a project open to all contributors. As a huge fan of the free software community at large, I’m just amazed by the huge amount of passionate ideas and work it pushes. Every time time someone attempts to encapsulate part of that energy and bring it aside for a mobile platform I think “Nah, you really want to be part of that storm, fuel that powerful entropy and be clever canalizing the energy to your platform and products.” You don’t know how happy I was the day I knew Nokia and Intel wanted to do just that with MeeGo.
Answers in the new community
You can find more details at meego.com . However, don’t look for all answers since there are so many missing. The reason is obvious: many of these answers rely today in the Maemo and Moblin communities. Also in the open source upstream projects feeding the MeeGo architecture. Also in the application developers that ultimately will make this platform successful. Also in the chipset vendors, device manufacturers and other users and stakeholders of this platform called to spin the whole mobile industry.
All these answers are somewhere: we just must find them. It will require the best of our brains. It will be deep, it will be fun. I just couldn’t wait today’s launch in order to start the bootstrapping of MeeGo. The whole thing is actually big and digesting it takes some time. If you want to be part of this I recommend you to go through the website, subscribe to the mailing list and decide the task or area you want to push first. In other words: choose your mission in the MeeGo project.
My preferred mission here and now is to contribute in the evolution of maemo.org and the Maemo community in the new MeeGo context. I have some ideas but, to be honest, I’m biting my tongue in order to let you go first. But still I couldn’t sit just quiet waiting for the launch, so I decided to start a wiki page describing and comparing the most interesting assets of the Maemo and Moblin communities. Please help improving it. It will be useful to figure out all what we could have if we are clever integrating and merging.
And now the personal anecdote
In Autumn 2006 I was a 770 user applying for a job at Nokia. There was this interview with Valtteri Halla (now MeeGo’s Technical Steering Group member) and Ari Jaaksi (now VP and head of Maemo Devices). I was explaining them how great it was working as self-employed or in small cooperatives and living in a lovely house in Andalusia. Ari asked me what were my motivations to leave all that, move to Helsinki and join a big corporation. Well, by that time my second child was born and I really needed a source of income. 🙂 But the reason not to hesitate taking such a chance was my explicit agenda of bringing Linux and free software to the real mainstream.
Today, three years after getting that job, I feel this agenda (pushed together with many others in the World: I love you all) is about half the way. Working at Nokia you really learn the meaning of the word “mainstream” and yes it includes your cousin, your neighbor and many people you don’t even think of. MeeGo is a platform to reach them and offer them something useful and exciting for their lives. Software freedom lovers: you know what I mean.
Filed under: maemo | 75 Comments
Tags: community, development, freedesktop.org, GNU/Linux, Intel, maemo, MeeGo, Moblin, Nokia, open source
otoh
flors 
One of the main reasons I like Maemo is that it’s based on Debian. What will this new merged OS be like?
Digesting right now 😉
But the linked wiki page at meego.com is password protected.
Great news!
Just the wiki page you link to requires authentication… :-}
Hi Quim, congratulations! Just a minor note: The wiki page is currently password protected. Any chance on getting that fixed?
This could be interesting, but did they have to call it “MeeGo”?
This is very interesting news. I’d been getting concerned that maemo might disappear under a flood of Android, likewise moblin — a great shame, because having a full linux device like the n900 is so much more powerful than just android.
I’m living in Canada though, the n900 isn’t suited to the frequency bands used here 😦
Hopefully, this merger will mean more devices available + maybe even one we can use here in the frozen north.
Seriously, the moblin comic people look much better than the MeeGo ones 😉
With this, you don’t win against the cute Android.
VERY good blog and great news indeed.
Greetings from Oslo!
Nokia, speak better English !
English spoken with Finnish accent do not sound very good. It’s utterly horrible.
Can’t you just hire a NATIVE ENGLISH speaking teacher/trainer whom you can exercise with before shooting the final video?
@Alexander: Please kindly shut your ignorant, idiocy-spewing mouth.
If it matters, nokia and intel already worked on this telephony stack :
http://www.newlc.com/en/ofono-open-source-telephony-solution
Do I understand it correctly that this means Intel is ditching Clutter in favor of Qt?
@12 Markus: interested in the same matter.
Clutter is an Intel Technology, as well as QT is for Nokia… It would be great if the new platform just give us the choice. 🙂
I don’t have finished to read all the announcements in the links published on my Twitter stream but I already feel so excited by this great news!
I think this will be a very good opportunity both for Maemo and free software to spread all around the world! Go Maemo… ops… MeeGO! 😀
Why in the world would anyone think it is positive to re-write the world based on that heap-of-crap toolkit: Qt ? why would everyone want to depend on a library completely owned, and controlled by one company ?
Since when are gtk+ or clutter ‘legacy’ toolkits ?
“The MeeGo UI toolkit is the primary toolkit for developing MeeGo applications and is based on Qt with specific enhancements and additions. GTK and Clutter are also included for application compatibility.”
http://meego.com/developers/meego-architecture
Another roaring defeat for my toolkit of choice, Gtk+.
One quality I have traditionally enjoyed about the open source community is that they were honest about things and didn’t hide hard facts. Could someone relevant please comment on Gtk+’s second defeat in the mobile arena without trying to sugar coat it?
@Skeptic
MeeGo is inclusive by default, unlike traditional proprietary mobile systems. Want to write Gtk+/Clutter apps? Go ahead!
At the same time, I think it’s important to acknowledge that there is a huge untapped developer pool out there, for whom both Maemo and Moblin are way too low-level at the moment. Gtk+ and Clutter are themselves fairly low level, and the SDKs of the two platforms immediately loose a huge amount of potential app developers by being Linux exclusive. Qt has the potential to change that, with great tools and documentation, enabling fun and easy development on the platform of your choice. And more apps is always good, right?
With GTK+ being second citizen of this platform, I can only say, good luck with that.
will meego run on N900 ??
Anonymous, the MeeGo site says it will be based on rpm, not deb.
Qt makes sense since Nokia owns it and Qt is designed to be platform independent (http://qt.nokia.com/products) meaning for example that Nokia or someone else can develop an app that runs on meego and symbian very easily. GTK+ just doesn’t offer that kind of robustness and flexibility at the moment.
Well the fact that Nokia owns Qt should be enough for anyone to understand the choice 😀
Will meego work on N900 if not what a farce I feel sorry for anyone who has bought one
Regretfully it is hard to interpret this as anything less than the death knell of Gtk+ on mobile devices and of GNOME Mobile in general. Farewell toolkit of choice, you were a good friend.
Joined up. Hope this is as open (or even more!) than maemo has been in the past.
@John Stowers:
I think you’re right, sadly. Very, very disappointing, esp. considering Intel’s original investment in GTK/Clutter. Although, the worst part for me personally is that it’s not going to stop me from doggedly insisting on GTK for as long as I can get away with it; I’d never seen Qt Creator before viewing this screenshot today, and it made me cry: http://meego.com/sites/all/files/users/u13/qt16.png
In fact, that screenshot really sums up MeeGo how development looks to me overall right now: Qt, Qt Creator, and C++ — functional, but ugly, painful, and loathsome. Moving to Android (and thus Java) almost looks fun by comparison.
Indeed that’s good news, but like many other I can only regret the adoption of Qt as a common platform.
Why don’t you guys just join an existing distribution like Fedora, Debian, Gentoo?
Thank you for this wonderful discussion with deep respect
Jens Knutson: Please, give Qt and Qt Creator a try. I’m sure you’ll be productive very soon with Qt if you don’t play the I-am-biased game.
harryf,
I don’t think you understand why Jens wants to cry. If looking at the screen shot doesn’t explain why, then you just don’t understand what GNOME users look for in an interface. That interface is a mess. It tries to push so many features in the UI at once without thought to how the features will be used. That’s one key reason why people stick with GNOME even though featurewise KDE has the edge.
@Anil: The comparison with Java and Android doesn’t make sense. Look at a standard Eclipse screenshot and count the amount of items. Creator is very, very clean compared to that. Even beats the GNOME IDE, Anjuta.
This is very interesting, Quim. It’s pretty fascinating to see how ‘household’ open source is finally beginning to become. And this time without the unnerving feeling of Google looking over your shoulder.
I only hope you will make it as well-supported and easy to use as Android. And that you will make listening to developers a top priority. But what company could be more qualified to live up to that challenge than the largest company in the home country of Linus Torvalds?
I guess I’ve found my next phone – this is going to be big 🙂
Just when Moblin started to get usable and I was seriously thinking about getting a NetBook…
Nokia, wtf? First destroying Maemo by forcing Qt into it and triggering a rewrite of the whole thing. Now doing the same for Moblin AND disrupting Maemo a second time in under a year by changing much of the low level stuff.
A mobile platform needs to be stable over years. You don’t even manage to keep it remotely the same for 6 month. There’s a solution… Android.
@harryf: Anil has it right.
As for “bias”, you’re damn right I’m biased. GTK is the toolkit I know, it works great, it gets more slick and enables more intelligent, useful, usable features with each release, and it fits in /fully/ with my desktop of choice. Most importantly though, the GTK hacker culture values clean, usable interfaces… the GTK/GNOME hacker community generally seem to understand and value the “human” side of software much better than many other FOSS cultures.
Re: the comparison with Java and Android, it does make sense if you understand the context. I’m saying that, overall, coding in Java for Android in Eclipse (or NetBeans, or gedit or ) looks a lot less dreadfully ugly and painful than Qt/C++. Simple as that. If Qt and C++ are what really turn your crank, hey, good for you, I’m envious!
Thanks for the explanation, Quim, and especially for the anecdote at the end. You had never told me about the story behind you taking your position. Speaking of which, what does this mean for YOU?
Randy (Texrat)
To expand upon a couple of other comments here, I was shocked and more than a little disappointed at the decision to go with a non-Debian-based system. The Debian-based system was actually one of the major points that got me to give the n900 a chance and actually start exploring development (in addition to purchase, of course).
Can anybody clarify the technical reasons for the decision? A change like this means that a whole lot of developer expectations about the systems will change; and it also means the loss of the very vibrant ecosystem connected to the Debian base.
And to reply to my own on this, if switching to a non-Debian-based system hasn’t been set in stone yet, who can we contact to add our input? To me, this goes much deeper than the difference between GTK+ and Qt.
And here I was, so happy/excited that Nokia was making a Debian-based alternative to Android/iPhone. I’ve been quite happy with my N900, and before today’s announcement, I was planning on buying its successor as well. Now, I really don’t know. RIP Maemo.
I second ConcernedDev’s question: WTF made you guys decide to drop Debian?!
You guys should just give Qt a try. Qt has wery nice and clean API. I do use GNOME desktop on Debian and I don’t really like KDE so much. But I as a software developer DO like Qt wery much.
If you compare Qt and GTK+ then the difference between those is as big as between day and night. And Qt is the day GTK+ is the night. In my opinion GTK+ is ugly and total mess when compared to Qt. What’s even worse is that GTK+ has never had any good object oriented API for the C++ developers.
Qt is also crosspltform. It is easy to write applications which xan easily made compile on several platforms: Linux, Symbian, Windows, Mac OSX and Windows mobile. Good luck trying to do that using GTK+.
So go and read about Qt, I’m sure you will like it.You can find information from these links:
http://qt.nokia.com
http://forum.nokia.com
http://wiki.forum.nokia.com
19 Tomas: As if Qt/C++ is any easier than the (common) entry point in GNOME development: python with pygtk.
I personally am not thrilled by Qt/C++, and would prefer Android- or WebOS-style in order to attract MANY devs.
Or at least ship high-level bindings for C# (or Java fwiw^^).
After all, Qt as a toolkit is not that bad, as the devs on Maemo don’t clutter everything as much as the KDE desktop is, so I hope this goes on with Qt and it won’t inherit bad style from KDE.
As far as package formats go: IMHO both deb and rpm are both legacy. How about trying out something really lightweight and innovative like Arch Linux’s pacman?
Just to add on my comment:
When I see interfaces like this, I get scared: http://www.youtube.com/watch?v=0NK4zeAe5-Y
Built with the new “Qt quick”. That looks just horrible, and I can only agree with people saying, that they value the clean interfaces one is used from GTK apps.
I’m sorry but I don’t think that this merge will do anything for either Nokia or Intel. Two half-finished products does NOT mean one single finished one. If anything, it means 1/4 finished! And adding two independent teams to work together, is a management nightmare.
Personally, I see the whole thing dying eventually. Nokia might want to pull a “Windows Phone 7”-like rebooting and eventually go away from Symbian by using Meebo, but they will have a harder time than they think. It’ll take years to get there, if the project won’t die before hand already. That’s my take at least.
The mistake many people are making here is to substitute Qt to mean kde and gtk to mean gnome. Its a very simplistic way to look at it. Although KDE is based on Qt Its by no means an indication of what Qt is capable of. Some of the best and most used applications on Desktop computing are based on Qt (looking at Virtualbox, Opera, Skype, VLC ) One of the main advantages of Qt is its portability. It allows a developer to creates write an application once and be easily port same to just about any platform you can think of (megoo,symbian, windows mobile, windows, mac, Linux)
Another advantage of Qt over gtk its ability to integrate and adopt the native UI of any desktop shell its running on. Try using pidgin or any gtk app on KDE what you would get is something out of a windows 95 era. Its not same with Qt apps They all draw the native desktop widgets and hence integrate very well. Hence VLC, Skype and Virtual-box all retain the gnomish look when used in gnome. Also in term of development, Qt is the Android of Toolkit atm you just cant compare the pace of development and momentum behind Qt compared to GTK. Choosing Qt has the toolkit for meego is really a no brainer.
Should have retained Moblin name.
@Bigbrovar very well said. With the logic of some people here we should go back to MFC since a developer would rather gnaw his finger off than add a UI element, so the UI stays very, very clean 🙂
@Kadath,
Honestly speaking; someone has to tell them the truth.
FinnEnglish sounds really bad.
“Why in the world would anyone think it is positive to re-write the world based on that heap-of-crap toolkit: Qt ? why would everyone want to depend on a library completely owned, and controlled by one company ?
Since when are gtk+ or clutter ‘legacy’ toolkits ?”
QT is far more advanced and polished then Gtk. An only heap-of-crap toolkit here is Gtk – sluggish, lack of modern features like animations and its code is a complete mess! Idiocy like object oriented programing in C… Clutter provides animations, but only if you have OpenGL acceleration enabled and QT provides navtive animations with and without OpenGL. QT isn’t completely owned by a company and it’s GPL, so you can always fork it. I can say Gnome is owned by some foudation which is very MS friendly – they chosen OOXML instead of ODF etc. Linux world would be more advanced without Gtk.
Please stop this useless toolkit flames. With discussions at this level Linux and the free desktop would have achieved nothing.
Qt offers a complete cross-platform toolkit for native development and also web runtime development on top of QtWebKit plus a nice set of developer tools and documentation. Note that I’m not mentioning programming languages or paradigms here at all. If you are interested in MeeGo and have other preferences, great: push for them.
No need to argue.
This is very disappointing news. I fear months and years of development work will now go into integration work and the actual goal of producing a better mobile os will be lost.
Nokia seem to have become very whimsical and unpredictable in their decision making.
This is disappointing. Intel primarily funded the work on Clutter and Moblin. Now that Intel will be shifting focus to Qt, which has its own GL/scene-graph subsystem, Intel will have little incentive to put much funding and attention to Clutter, GObject, and GTK+. So, Clutter and GTK+ loses more dedicated programmers, and GNOME’s infrastructure becomes that much more difficult to maintain and move forward. GNOME’s future in the device space is looking increasingly grim, and, since it was the device space which was in great part driving the innovations for GNOME 3.x, I can only wonder what effect this announcement will have on future GNOME 3.x development. This is very disappointing indeed.
MeeGo is not a very good name. Sounds like the name you would give to a medicine for curing constipation than a software platform!
If the N900 successor is being planned to have an X86 Intel chip will it run Windows 7 so that I could dual boot either Windows 7 or MeeGo??
Well, Maemo and Moblin could die separately, or they could join together and learn from their mistakes. Neither Moblin or Maemo had access to the application pool needed to make such environments successful, nor were they going to get the developers needed to get them. Unfortunately, the development technology around Moblin and Maemo had quite a bit to do with that.
Developing with GTK+, Clutter and the huge panopoly of different libraries was just too damn hard and painful. It always has been. They are all too low level when developers get hold of them, they all have their own APIs that look different, they all work in their own ways with fairly loose integration and it’s just not what developers using platforms like Windows, OS X, Android or the iPhone want to want to see. Why is the HIG so important for GTK and Gnome? Because when you create a new application you don’t just inherit the layout, look and feel of the general environment. You start from scratch. That will be an unheard of shock for most developers. It’s just not acceptable. Given that many things just haven’t improved you can only look at that development stack as a very large money pit.
Lack of resolution independence, difficult to port to different devices and platforms, difficult to maintain and deploy on different platforms…………. Moblin and Maemo were DOA from a development point of view, so, yes, I would expect the change in tac to make a difference. Like it or lump it, development and what you use for it is all important. If you don’t see that………….well, you live on a different world.
As for that screenshot of Qt Creator:
1. Someone has obviously tried to compress the thing into a very small space.
2. What are developers supposed to use on the existing Maemo and Moblin environments and where is the Qt Creator IDE equivalent? That’s right. There isn’t one.
The development environment was always going to weigh ‘Gnome Mobile’ down unless it improved, and I don’t think it can be looked on as anything other than an embarrassing failure.
Bigbrovar: “The mistake many people are making here is to substitute Qt to mean kde and gtk to mean gnome.”
Thanks for pointing that out. The toolkit does not determine the interface, so the argument that gtk+ apps are cleaner is simply not true, its the developers choice.
This is why they chose Qt: It runs on almost every platform ant integrates well. Ever seen gtk-apps on kde or on windows? You just cant something like that. And this what its all about. They need developers who make thousands of apps to establish the OS next to android and iphone OS. The majority of these devs wants a simple, easy to use high level toolkit and thats Qt is meant to be.
@55 nobody: Ever seen an iPhone app on Windows? neither have I!
So that is all crap. Furthermore, with the right theme, they look just like XP (eXperience engine/theme), and I bet there is even an Aero style somewhere.
Just nobody is shipping that.
Also, it’s not about the looks, it about the feel.
And then write once, deploy everywhere doesn’t make it either.
Because a Qt app will never feel native on any system, be it Windows, OS X or GNOME.
It will just feel like a Qt app, and maybe the buttons will look similar, but that’s all.
Another part of the criticism no-one seems to respond is, that for GTK/GNOME apps, people use all kinds of languages, while (all popular?) Qt apps are written in C++, which may not be the language of choice for many people.
http://www.kdedevelopers.org/node/4167 <- Writing Qt apps in Java, probably a small game changer to my own criticism, just got that link.
All the same old tired arguments………..
An iPhone app on Windows? Well, that would be rather silly wouldn’t it considering the platforms are completely different on completely different hardware and iPhone apps aren’t designed to be cross-platform? A silly example for cross-platform applications, but I’d expect nothing less. However, I’m sure Nokia would like to see a class of applications that run on Symbian and many of their Linux based devices as would Intel on theirs and resue the work. The fact that developers can use the same toolkit to develop on their desktops as they are porting to on mobile devices is yet another added bonus.
Qt is a cross-platform toolkit. It does its level best to integrate into the host system with look and feel without you having to do anything, but no, it isn’t going to integrate completely with it without some porting – which you can do. However, it does a far, far, better job than Java, Swing or SWT – which has a bug list the size of Brazil outside of Win32. The point is that the effort to port, should you do so, is hugely reduced. If you need to port to a few different platforms, as Nokia and Intel want to, then pick your poison. Why do people think the Firefox port is so damn poor in Linux versus the Windows version? You need to maximise your manpower by letting the software do the work for you.
Now, if I had a choice between having a lot of decent applications and being able to do something useful with my computer or mobile device and achieving some perceived zen of integration, then I know what I and most people will choose. Applications and functionality.
Bindings? Not that crap again please. Qt has a ton of bindings fom Java, Perl, Python to Ruby. It’s not easier or more difficult to create bindings for it. If that’s a selling point elsewhere then where are all the brilliant GTK applications written with Ruby, Java or Python? Very few and far between is the answer. Hell, people are rewriting apps in Mono back into C.
This is very interesting news
Quim,
Did you miss our questions or are you just ignoring them?
We’ve yet to hear any explanation for abandoning Debian. I take your silence to mean that there *aren’t* any technical reasons — this was purely a political move. Please, correct me if I’m wrong.
Sorry, I’m spending most of my trying trying to clarify things but the discussion is split in several places.
My personal opinion about rpm/deb: http://lists.meego.com/pipermail/meego-dev/2010-February/000072.html
The meego-dev list has a long thread about this. If anybody has anything new to say that is the right place.
This is indeed great and I am eagerly awaiting this new OS. Actually I was eagerly awaiting the Maemo 6 which would be Qt based (as a KDE user that is obvious 😉 ). I think one of the most critical aspects of the migration of the Moblin and Maemo communities to start co-habitating in the MeeGo space is an upgrade path of existing Moblin and Maemo devices, preferably even the older N*** Maemo devices and not only the N900 (where a migration to ofono could be done at the same time?). This is in the power of Intel and Nokia, respectively. Doing this at an early stage will encourage central individuals in the two developer communities to migrate to this new platform on their existing devices, which will inspire the rest to follow suit.
On top of that – the big “win” for Nokia in offering an upgrade path for its existing Maemo devices will be that they can get the reputation of shipping “future-proof” devices. This in it self is very valuable since it means that less people will wait out the next hardware version rather than buying the current one (a common problem in technology, a habit that has been cultured due to consumers being burnt by primarily the upgrade paths of closed, proprietary systems).
It sounds like some decisions, such as picking rpm over apt, were made more on political than technical grounds. The Debian package-management system has always been the best. I have tried rpm-based systems over the decades from time to time, and always given up on them because they were based on inferior package management.
The fact that Meego is a stupid name may not kill it: Debian is just as stupid, and they survived.
Still, I will hope that Openmoko wins in the end. Their name is sort of dumb too, but at least they are Debian-based.
I’m always happy to hear about a new iniciative on Free Software, so I wish best luck to this project and I will collaborate in any way that it is possible. Anyway I have to admit that I hoped you did not give up on Debian, but it’s ok, I understand that “users” make the final choice and they want a lot of applications and always an easier way to use them.
I kind of like the idea. Meego!