Five Ubuntu Touch Facts

A few days ago we announced Ubuntu for Tablets; the next piece on our wider Ubuntu convergence story. The tablet joins the Phone, TV, Ubuntu for Android, and the Desktop. See an excellent hands-on video review of the current developer build from Engadget.

Today the source and images for Ubuntu for Phones and Tablets (collectively known as Ubuntu Touch) was released.

I know there is some anticipation regarding this release and I just wanted to share a few facts to ensure we are all on the same page:

  1. Both Phone and Tablet code and images are available – today we are releasing two things for both the phone and the tablet. Firstly, if you simply want to run the software on a spare device, you can install the images on your device without caring about the code. If on the other hand you want to see the code (and contribute to it) we are also making this available too so that you can build, explore, and hack on it.
  2. This is unfinished and in-development software – it is important to remember that this is in-development software and as such is not finished yet. You are going to find that some features and applications are missing, and you will likely find bugs. We wanted to release the code and images early so that our community can try the software, provide feedback, and be able to join the development effort. With this goal to get the content out early we just want to ensure everyone fully understands that this is not yet a final product. I strongly recommend you only install the code/images on a spare handset/tablet and not your main phone/tablet due to the fact it is in-development code.
  3. A limited set of devices are supported – the images are only available for the Galaxy Nexus, Nexus 4, Nexus 7, and Nexus 10; these are the devices that our development team has been working towards. We appreciate that you may have a different phone or tablet, but unfortunately support for other devices is not currently planned. We will however be kicking off an outreach campaign soon to encourage and support our community in porting the code to other devices. Stay tuned for more!
  4. A new SDK is available also – in addition to the release of the code and images we have also released a new version of the SDK which includes a number of new features, most usefully the ability to deploy a QML app to a device so you can run it!
    • Ubuntu SDK application templates and wizard
    • QML2 UI designer
    • Templates for testing framework and internationalization
    • Deploy QML applications on an Ubuntu Phone/Tablet device
    • Basic terminal (ssh, adb) connectivity tools to the device
  5. Know where to find help – if you have questions or queries you should post your questions to Ask Ubuntu by clicking here.

I am sure you are now chomping at the bit to grab the images, check out the code, and get the new SDK release! Go and find all the details here.

  • J.C&TheGang

    Wonderful news! Thank you so much! I’m looking forward to tinkering around with this beautiful piece of engineering.

  • Fabio Sasso

    Can I install it in futureon my cheap 99€ mediacom tablet? or support is only for Nexus, samsung, sony etc…?

  • Corey Quilliam

    Edited post, scratch that.

  • http://twitter.com/jspaleta Jef Spaleta

    Jono, It’s nice to see the SDK make an appearance. Key part of app ecosystem building.

  • the Humble 1

    I did not try or particularly know about the SDK or the new APIs (if any) is being developed for Ubuntu Touch, but as far as I understand the app framework is based on HTML5 and the like.

    I think that’s not what’s to be done to write the convergence story. Instead of creating a new ecosystem and supporting SDKs and APIs, I’d rather think devoting engineering resources to make existing libraries used by apps on the desktop, i.e. GTK+, Qt, etc., support touch and adapts the UI according to the device as does some JS libraries for web apps. That way existing Linux apps will be available for use on phablets. Such powerful desktop apps are exactly what is sorely missed on mobile devices, in my humble personal opinion.

    And finalize whatever you’re going to use instead of X, port Unity and use the same thing across all the devices already. Plus web apps can still be supported as it is on desktop.

  • http://davidplanella.wordpress.com David Planella

    Note that we’re already doing that. The SDK is based on Qt Quick 2, from the Qt framework, which does indeed support touch. We support both native and HTML5 development.

    Apps developed using the Ubuntu SDK can run on the desktop and on mobile devices. In fact, it’s pretty awesome. You don’t even have to cross-compile: test your app on the desktop, flash it to your phone and test it there.

  • http://davidplanella.wordpress.com David Planella

    Currently only the devices listed here work with the images: https://wiki.ubuntu.com/Touch/Install

    Stay tuned for an upcoming community initiative to create ports for other devices!

  • the Mouthy 1

    In that case, wouldn’t it be better to port existing apps instead of creating ones from scratch as the CoreApps Wiki page suggests? Calculator, Calendar, Terminal, etc. already have great and already default desktop applications. Port/fork the GTK libraries to make them touch friendly. No need to create new apps.

    In the case of YouTube, Facebook and Twitter, Ubuntu can already pin the websites as web apps, why not make it work on Touch? The websites themselves will provide a touch UI, reading the user agent, offering a consistent experience, approaching them as a service provided by someone else, not an app. If you want to provide social networking application, there’s Gwibber.

    I understand that the SDK supports existing Qt tech. But Ubuntu Touch isn’t supporting existing apps, which is the point I am trying to drive home. I’ve extensively used Android, and it’s limitation lies not in the operating system, but the apps. Converging the operating system is a great idea, I love it and support it. But not including the existing applications that could be used in the said operating system defeats the point, don’t you think?

    I love Ubuntu, but if I install it on my phone and instead of finding what I have familiarized myself with on my desktops and laptops, it is nothing more than an OS with the same name and with yet another new app ecosystem.

    The above is my opinion as an end user and a power user. Of course, as a developer (which actually I am not) finally being able to create apps for different form factors without porting at all is incredibly cool. But in the end, as an Ubuntu user I can’t find the Ubuntu I know, and as someone looking for a phone it’s some novel new OS like HP’s Web OS.

    TL;DR Don’t abandon the repository of apps build by the love of Linux over the years, but use it to empower the platform and differentiate as one that can run the same apps we’ve always used.

    … I am quite chatty today for reason. I just ranted on some other blog’s comment (entirely different topic, mind you), but thanks for reading.

  • http://twitter.com/jspaleta Jef Spaleta

    you should read http://www.phoronix.com/scan.php?page=news_item&px=MTMwOTM

    “While Canonical is pulling in portions of Android/CyanogenMod, its the lower levels and not Dalvik for the Java Virtual Machine bits. So while there’s pieces of Android, it is not enough to Android applications on Ubuntu. Without a traditional display server like X.Org used by Ubuntu on desktop, standard Ubuntu applications also will not run from this mobile edition of Ubuntu. The Ubuntu Touch interface is written in Qt5/QML to run on Android.”

    It’s really important to understand what you are seeing in the Ubuntu Touch preview deliverable is not the traditional Ubuntu linux stack at the plumbing layer. This is not Ubuntu recompiled for ARM. This is most likely CyanogenMod stripped down and rebuilt to provide a variant of Ubuntu branded interface. Which is to say this is not the compiz plugin code you see on the Ubuntu desktop, its a new implementation. An implementation I expect to see replace the compiz approach on the desktop offering as well.

    Because these images are using Android derived hardware plumbing for the display subsystem, you cannot expect traditional linux graphics applications to automatically work. Nor can you expect alternative interfaces from the *buntu repositories to be able to work either, not without a lot of work.

    Applications have to be built with toolkits that understand how to talk to Android’s display server tech as well as X11 to be cross platform between U desktop and U tablet. Qt5 has support for both display servers, and it explains why qt5/qml has been chosen for the SDK.

    There are some important implications here with regard to the continued relevance the existing debian style U. repository universe. Canonical will have to bridge the gap between the android plumbing and the tranditional linux distro plumbing to make it possible for existing applications to work cross-platform. I’m not saying it not technically possible to do, nor am I saying this isn’t on the roadmap for them already. Hopefully they get the work done to make this seamless. As it stands I think it will be easier to run android apps on U tablet than it will be to run traditional U desktop X11 apps. Here’s to hoping I’m wrong about that.

    -jef

  • the Bamboozled 1

    Yes, I read that today (my last comment was posted last night in my time zone). I understand the the implication of using a display server other than X11, and therefore why there’s a new app framework.

    I am honestly baffled on the display server choice here, and if this the display server going forward, I am flummoxed further. It is incompatible with any existing OS. So much for the “convergence story”. Might was well re-brand it and market it as new OS.

    I understand that Wayland is not ready, but it already has stable APIs and such, and that’s where whole Linux ecosystem is headed, complete with X compatibility. I’m sure Qt works on Wayland as well, does it not? If Ubuntu Touch used Wayland to support the SDK, the other existing pieces would naturally fall into place, either with XWayland or the graphic libraries themselves being ported to run on Wayland (which is more probable in the longer run). I’d rather devote engineering resources to Wayland for touch compatibility than hoping to nurture a new ecosystem.

    Here’s hoping that SurfaceFlinger is a temporary measure to showcase the U Touch, and Wayland or, at the very least X, will be the way going forward with the convergence story.

  • http://twitter.com/jspaleta Jef Spaleta

    Jono has already been quoted elsewhere, that Canonical is working on a custom display server. Not X, Not Wayland. I’ve seen nothing yet that would suggest that the custom solution Canonical is working on will be SurfaceFlinger or SF fork or something new entirely. We’ll have to wait and see as Canonical releases more information about what they are working on.

    But using SF isn’t necessarily a stopgap measure. Just like it will be possible to run X as a client process to Wayland, I fully expect it is possible to run an Xserver layer over the top of the SF compositor as a way to get legacy desktop applications to work. In fact an open source Xserver as been available for Android since March of last year and is installable from google play. So that particular technological solution exists and Canonical could easily “embrace and extend” that project as part of final deliverable. Not that I have any knowledge of any plans in that area, as I’m pretty much the antithesis of a trusted community member who can gain access to Canonical’s skunkwork plans.

    But for sure, a lot of work is left to be done.

  • the Avenged 1

    Yeah, I know about the custom display server comment and, be it SF or otherwise, it won’t make a difference if it doesn’t have backwards compatibility with X or Canonical doesn’t plan to port existing X apps/libraries to the display server in question.

    Without being able to use existing apps, I can see no reason why I, as an Ubuntu user will go for Ubuntu Touch, or would any “Linux user”, if Canonical PR planned to appeal to any.

    Well, that’s all I wanted to say and thus I rest my case. All I’ve said were in vengeance of my expectations and finding what we did. Here’s hoping for a real Ubuntu on mobile devices.