8 ms·

> Elop was right about needing to change horses. What was wrong with MeeGo? The memo just states: > at this rate, by the end of 2011, we might have only one

> Elop was right about needing to change horses. What was wrong with MeeGo? The memo just states: > at this rate, by the end of 2011, we might have only one MeeGo product in the market. But that's not soo bad considering that they started the platform in 2010. It's not like changing to a different platformbat this point would make things faster anyway.

oriolid 4y ago
> What was wrong with MeeGo? The way I remember it was that at the time the management-forced merge of Maemo and Moblin to Meego and the sudden switch from Gtk to Qt were thought to be risky moves and it was well known that Symbian supporters are doing anything they can to sabotage other platforms. Later I was told that N9 was incredibly polished but Meego wasn't that great behind the scenes and would have required massive rewrites if development had continued (which isn't that different from stories about early iOS).
- seba_dos1 4y ago
  > N9 was incredibly polished but Meego wasn't that great behind the scenes and would have required massive rewrites if development had continued That was probably because N9 didn't really use MeeGo - technically it was still a Maemo based system that was only disguised under MeeGo brand.
  - oriolid 4y ago
    The version of the story I was told that the Maemo/Moblin mashup was just as bad as the idea of combining two different frameworks that do essentially the same thing sounds like, nowhere near production ready, and the Harmattan aka MeeGo-branded Maemo with Qt was the only way it could be made to work at all. I haven't heard anyone say that genuine MeeGo would make things better.
    - seba_dos1 4y ago
      That was the sentiment of my comment as well, I guess I weren't clear enough - Nokia N9 was made to work great on what technically was still Maemo, but since they were supposed to go with MeeGo afterwards it would still require a lot of work to actually do that.
      - oriolid 4y ago
        Ok, I misunderstood. I thought it was clear at that point that Meego was a dead end, but Harmattan was missing things like app sandboxing and some app store and payment related features, the package management was a mess, etc.
        panick21_ 4y ago
        Its funny that things like running videos while sliding window out of screen and live task switch all worked. But somehow package management was the problem. When I developed for iOS I was shocked how non of that existed. Seems to me they had something really nice to go forward with.
        oriolid 4y ago
        Priorities tend to shift when your whole team has already been laid off, the product line cancelled and you still have the ability to make something flashy to put on your CV (again, I wasn't there myself). And probably there were a few demosceners in the team. But to be honest, the smoothness of the UI was really impressive, and even more when you know how underpowered the CPU was compared to Android phones at the time.
  - panick21_ 4y ago
    Can somebody explain what Maemo and MeeGo means here? I remember the N900 coming out and wanting Maemo. Then years later when I didn't care anymore I MeeGo peaked my interest, but I didn't really care anymore. What other then GTK vs QT is different. And why was it needed, I never understood why they simple didn't go forward with Maemo.
    - toyg 4y ago
      Maemo was debian-based and built around GTK. Its device target was effectively the N900. Nokia then rebuilt the interface (at fantastic dev speed, it has to be said) in QT (which had been blessed as the UI framework in order to facilitate onboarding of Symbian developers, who had been told to use Qt for new Symbian apps) and shipped it in the N9. By then, however, management had struck an agreement with Intel to join forces over a Linux OS for devices. Intel had its own Linux distro, Moblin, which was based on RedHat. Moblin and Maemo were meant to merge into "MeeGo", a distro based on RedHat but with a QT interface. The project started fairly quickly but, by then, the N9 was basically ready, so Nokia effectively shipped what they had beforehand and just called it MeeGo. Beyond the UI, iirc, the differences between Maemo and Moblin/MeeGo were the packaging system and some service daemons. The most annoying part, really, was that third-party apps built for Maemo would have had to be repackaged and retested for MeeGo, effectively throwing away all community efforts made over several years. Despite the best efforts by Nokia to placate folks, the community they had built around Maemo was completely pissed off and largely gave up, focusing on the actually-profitable systems. And then the burning platform memo happened.
      - panick21_ 4y ago
        Thanks! This really reminds me of the Unix wars. Constantly companies announcing partnerships and then there developers spending time merging proprietary systems. And before they are ready another partnership bringing in some other thing that has to be merged. Not sure what exactly they gained with this Intel deal. I guess they should have just continued with Maemo and attacked other software developers in general, rather then rewriting the whole stack just to attract Symbian developers. A year after the N900 they should have been an N1000 by 2010. By the time late this memo happened the N1100 should have been ready to drop.

oriolid 4y ago

> What was wrong with MeeGo? The way I remember it was that at the time the management-forced merge of Maemo and Moblin to Meego and the sudden switch from Gtk to Qt were thought to be risky moves and it was well known that Symbian supporters are doing anything they can to sabotage other platforms. Later I was told that N9 was incredibly polished but Meego wasn't that great behind the scenes and would have required massive rewrites if development had continued (which isn't that different from stories about early iOS).

seba_dos1 4y ago
> N9 was incredibly polished but Meego wasn't that great behind the scenes and would have required massive rewrites if development had continued That was probably because N9 didn't really use MeeGo - technically it was still a Maemo based system that was only disguised under MeeGo brand.
- oriolid 4y ago
  The version of the story I was told that the Maemo/Moblin mashup was just as bad as the idea of combining two different frameworks that do essentially the same thing sounds like, nowhere near production ready, and the Harmattan aka MeeGo-branded Maemo with Qt was the only way it could be made to work at all. I haven't heard anyone say that genuine MeeGo would make things better.
  - seba_dos1 4y ago
    That was the sentiment of my comment as well, I guess I weren't clear enough - Nokia N9 was made to work great on what technically was still Maemo, but since they were supposed to go with MeeGo afterwards it would still require a lot of work to actually do that.
    - oriolid 4y ago
      Ok, I misunderstood. I thought it was clear at that point that Meego was a dead end, but Harmattan was missing things like app sandboxing and some app store and payment related features, the package management was a mess, etc.
      - panick21_ 4y ago
        Its funny that things like running videos while sliding window out of screen and live task switch all worked. But somehow package management was the problem. When I developed for iOS I was shocked how non of that existed. Seems to me they had something really nice to go forward with.
        oriolid 4y ago
        Priorities tend to shift when your whole team has already been laid off, the product line cancelled and you still have the ability to make something flashy to put on your CV (again, I wasn't there myself). And probably there were a few demosceners in the team. But to be honest, the smoothness of the UI was really impressive, and even more when you know how underpowered the CPU was compared to Android phones at the time.
- panick21_ 4y ago
  Can somebody explain what Maemo and MeeGo means here? I remember the N900 coming out and wanting Maemo. Then years later when I didn't care anymore I MeeGo peaked my interest, but I didn't really care anymore. What other then GTK vs QT is different. And why was it needed, I never understood why they simple didn't go forward with Maemo.
  - toyg 4y ago
    Maemo was debian-based and built around GTK. Its device target was effectively the N900. Nokia then rebuilt the interface (at fantastic dev speed, it has to be said) in QT (which had been blessed as the UI framework in order to facilitate onboarding of Symbian developers, who had been told to use Qt for new Symbian apps) and shipped it in the N9. By then, however, management had struck an agreement with Intel to join forces over a Linux OS for devices. Intel had its own Linux distro, Moblin, which was based on RedHat. Moblin and Maemo were meant to merge into "MeeGo", a distro based on RedHat but with a QT interface. The project started fairly quickly but, by then, the N9 was basically ready, so Nokia effectively shipped what they had beforehand and just called it MeeGo. Beyond the UI, iirc, the differences between Maemo and Moblin/MeeGo were the packaging system and some service daemons. The most annoying part, really, was that third-party apps built for Maemo would have had to be repackaged and retested for MeeGo, effectively throwing away all community efforts made over several years. Despite the best efforts by Nokia to placate folks, the community they had built around Maemo was completely pissed off and largely gave up, focusing on the actually-profitable systems. And then the burning platform memo happened.
    - panick21_ 4y ago
      Thanks! This really reminds me of the Unix wars. Constantly companies announcing partnerships and then there developers spending time merging proprietary systems. And before they are ready another partnership bringing in some other thing that has to be merged. Not sure what exactly they gained with this Intel deal. I guess they should have just continued with Maemo and attacked other software developers in general, rather then rewriting the whole stack just to attract Symbian developers. A year after the N900 they should have been an N1000 by 2010. By the time late this memo happened the N1100 should have been ready to drop.