Announcement

Collapse
No announcement yet.

Plans For Ubuntu On Phones Are Laid

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Plans For Ubuntu On Phones Are Laid

    Phoronix: Plans For Ubuntu On Phones Are Laid

    Besides everything else that went on today at the Ubuntu Developer Summit in Orlando for the upcoming Ubuntu 12.04 LTS release, initial plans for Ubuntu on mobile smart phones were laid out...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Ubuntu 14.04 is way too far away. Don't you mean 12.04?

    If it's going to take two years then Ubuntu on phones or even Ubuntu on tablets, then it's a lost game. It would mean the general Linux front-line is on the wrong battle field.

    Comment


    • #3
      Obviously the Linux mainstream should rip Android code.

      Comment


      • #4
        Aside: Article should read jack of all trades, master of none.

        Canonical are deluded if they think they can do a phone OS. The kernel is relatively easy - just take Android's. But apps and data really matter. Existing apps that don't rely on hover or accurate mouse clicks will work fine in a tablet form factor since they have a large screen, lots of battery and RAM etc.

        But on the small screen you have to have completely different user interfaces. Who is going to refactor all the necessary apps adapting them to a mobile toolkit? If they want the apps to be memory and CPU efficient (reducing the bill of materials for the phones) then their internals will have to be restructured too, or completely new apps written. That is a lot of work. Then you have to convince other developers to make apps for your APIs and method of operation - after all people will be wanting to read their Yahoo Mail, view maps and play Angry Birds. How to handle multiple architectures (ARM alone has a few), compiled versus interpreted runtimes, DRM for sold apps etc are a lot of work.

        Canonical is a few hundred people total with already full plates - every other group that has tried this has been larger and taken several years. I've yet to see any indication as to what magic Canonical have that will somehow let them do so much more, quicker and better than previous efforts. And even if they succeed, what exactly is the point? They won't have the same volumes as the incumbents (Android, Apple, Microsoft) so they won't be able to make much licensing money. Phone companies wanting to be in control of their own destiny will develop something themselves (eg Samsung's Bada). Hardware companies being cheap will just use Android.

        Comment


        • #5
          Originally posted by grotgrot View Post
          Aside: Article should read jack of all trades, master of none.

          Canonical are deluded if they think they can do a phone OS. The kernel is relatively easy - just take Android's. But apps and data really matter. Existing apps that don't rely on hover or accurate mouse clicks will work fine in a tablet form factor since they have a large screen, lots of battery and RAM etc.

          But on the small screen you have to have completely different user interfaces. Who is going to refactor all the necessary apps adapting them to a mobile toolkit? If they want the apps to be memory and CPU efficient (reducing the bill of materials for the phones) then their internals will have to be restructured too, or completely new apps written. That is a lot of work. Then you have to convince other developers to make apps for your APIs and method of operation - after all people will be wanting to read their Yahoo Mail, view maps and play Angry Birds. How to handle multiple architectures (ARM alone has a few), compiled versus interpreted runtimes, DRM for sold apps etc are a lot of work.
          Simply categorise the apps. Tag which ones are phone suitable. Regarding programs that aren't suitable, I think people would innovate a means to control them though another app, for the smaller screen. Simulate a mouse or tab key to window-items.

          Also consider that phones of the future will blur the boundaries, as in a small phone may be able to plug into a keyboard/mouse and larger screen. This can be seen with devices getting HDMI ports for example. Tablets will also merge the boundaries between desktop and single hand held phones.

          Personally I prefer a tablet over a phone. I disagree with Jobs saying 7inch tablets are DOA.

          Anyway Linux needs to cross all fields, so abilities have to be added, as hardware demands.

          Comment


          • #6
            Originally posted by e8hffff View Post
            Simply categorise the apps. Tag which ones are phone suitable. Regarding programs that aren't suitable, I think people would innovate a means to control them though another app, for the smaller screen.
            Neither Evolution nor Thunderbird is suitable for a phone as an email program. You would have to a write a complete new user interface. Whether that then remote controls them or just talks to the non-UI side of their code doesn't alter how much work it would be to do the UI. And of course both use huge amounts of CPU and RAM. None of the existing web browsers would work either. You could probably do minesweeper by translating long clicks into right clicks but even that will take someone a few days to code and test. Programs that use networking would need to be adapted to ensure they can work well even in the face of "flaky" networking. And you'd have to write new programs for dialers, write/adapt something for contact management, IM etc. It would be trivial to come up with a half baked experience that would be universally panned and no one would use. To come up with something even comparable to the phones that will exist in 2014 is a heck of a lot of work. (Alternate possibility is that the engineers working at all the other companies that have done this before are unproductive dolts.)

            Originally posted by e8hffff View Post
            Personally I prefer a tablet over a phone.
            Tablets are considerably easier to adapt software to. Phones are a lot harder because of the small screen size, flaky networking and if you want to keep costs down then you need to constrain CPU and RAM usage.

            Originally posted by e8hffff View Post
            Anyway Linux needs to cross all fields, so abilities have to be added, as hardware demands.
            Linux already does. It is on every Android phone out there - all approximately 150 million of them and growing by 500,000 every day. That is certainly larger than the number of Linux desktops and servers. I have a shell, ls and vi on my phone.

            So this is really about the GUI. Everyone seems to go for an OpenGL display server and then platform specific widgets and layout managers. You need some sort of sandboxing/security system so that applications are restrained in what they can do. You need to decide if apps are entirely native code like iOS in which case you constrain which processors are supported, or have a bytecode solution (Android, RIM, Windows Phone) which means adopting Java or .Net and still need some native code support for games. I've yet to see any explanation of how Canonical can do this logistically, why any phone manufacturer would want the result, and why any carrier would want to get involved with phones that had it. To succeed you have to do at least one thing really well - what is that for Canonical? As much as they worship Unity, it isn't a striking improvement over Android or iOS. Heck you could implement a launcher for Android with the Unity look and feel and then not have to worry about all the other stuff needed for a phone.

            Comment


            • #7
              Wrong statements

              Originally posted by phoronix View Post
              Phoronix: Plans For Ubuntu On Phones Are Laid

              Besides everything else that went on today at the Ubuntu Developer Summit in Orlando for the upcoming Ubuntu 12.04 LTS release, initial plans for Ubuntu on mobile smart phones were laid out...

              http://www.phoronix.com/vr.php?view=MTAxMDY
              Bada has outsold Windows Phone this year. How can you say it has failed ?

              Comment


              • #8
                without hardware -or serious support from an mfg- is bound to fail


                doubt canonical has the money to built HW and ubuntu is not as strong of a brand in order for mfgs to put their money on it.

                Comment


                • #9
                  Originally posted by e8hffff View Post
                  Ubuntu 14.04 is way too far away. Don't you mean 12.04?

                  If it's going to take two years then Ubuntu on phones or even Ubuntu on tablets, then it's a lost game. It would mean the general Linux front-line is on the wrong battle field.
                  Ubuntu Mobile in 2014 is correct. How you equate Ubuntu with Linux in general puzzles me. KDE is putting Linux with Plasma Active (KPA) on tablets right now. KPA 2.0 will be released as soon as December or so.

                  open-slx and basysKom are working on Linux for mobile devices.
                  Ubuntu will take so long because of Canonical's NIH syndrome. If Canonical would just pick up MeeGo/Mer and improve on KPA (no huge task to write a few new QML files to get a new user experience on top of KPA technology) and Canonical could get a KPA-based Unity Mobile working within a few weeks.

                  Originally posted by grotgrot View Post
                  Who is going to refactor all the necessary apps adapting them to a mobile toolkit?
                  KDE ? at least if Canonical are not totally stupid, they'll use Kontact Touch, Calligra Active, etc.
                  It'll be all Qt-based anyway.

                  Another possibility would be to revive old Maemo/Hildon applications but that'll take up way more resources.

                  Comment


                  • #10
                    What a horribly bad idea. Take a lesson from Meego please. Linux on a smartphone is already in place and is living large, Android. Canonical/Ubuntu should continue on their path of refining the distro for general use on x86 as well as dumping tons of support into ARM based systems. This is where the next battleground is going to be. Those that want a FULL OS on tablets no a mobile OS. Microsoft has seen this and is developing Windows 8 and fully supporting ARM and Ubuntu should be doing the same.

                    Comment

                    Working...
                    X