View source for KDEEdu/Sprint2012/MobileNotes

<pre>
 * KDE Edu on Mobile
  - What projects should be ported?
   - The one who codes decides
    - Ones that don't make much sense
     - ktouch
     - rocs
     - step
     - cantor (though mitght make sense for tablet)
      - maybe problem with dependencies (octave, maxima, ...)
     - ???


  - What's the differences to "regular linux"
   - Plasma Active is "regular linux"
   - N9 was close to "regular linux"
   - Android is weirdish
   - Playbook is not linux, has a POSIX layer


  - Development platform
   - We should assume we have Qt5/KF5, focus on porting the app, not on porting the "OS layer"


  - Do we want different versions for the Mobile version?
   - Integrate with the platform, if you want to integrate you might need abstractions
   - It might be easier to "focus" on one platform to get something running
   - What did Aleix do with KAlgebra?
    - Have QML with the same API and install the files that belong to that platform


  - Shipping kdelibs
   - Necessitas said there should be no problem in bundling kdelibs in the "necessitas installer"


  - Should desktop version also be QML?
   - Might make sense for appplications that want to go mobile, this way even if there's
     different "frontend UI" for the desktop one


  - Should they be available for free on the platform markets?
   - Also discussed yesterday
   - Who is the one that gets the money?
    - Maintainer?
    - eV?
     - Might be a problem given the statutes
    - kdeedu group?
   - People could "upload" the same app for free
    - Should not be a problem to get the appstore to kick the other one
   - Not force anyone to make it free or paid


  - Localization issues?
   - Translators are not someimes aware that the kvtml files should be translated (desktop affected too)
   - ktouch next
    - Keyboard layouts: No need to translate them since it's autodetected
    - Courses: Names could not make much sense to translate since it's "keyboard/language" specific
               and if you want to learn german keymap, you probably understand german anyway, what's
               the point on getting the title in French?


  - How to improve app-in purchase like a new translation? Porting GHNS or something else?
   - KNS is abstracted so you could use a "android backend" for it


  - Standard for flashcards programs?
   - Parley wouldn't probably ported to it there's not that much manpower
   - Maybe easier to "extend" KVTML


  - How to promote KDE Edu better in mobile software?
   - Not mobile specific
   - On the mobile store the "KDE EDU" brand gets a bit more dilutted
    - Mention "KDE EDU" on the appstore description
    - Put the "KDE EDU" logo somewhere in the app (main page if possible)
    - Mention "KDE EDU" in About dialog
    - Add a list of "other KDE EDU" apps?


  - How to get devices or simulators up and running easily?
   - Lower entry barrier
   - It's platform responsability to provide the raw SDK
   - We should provide the tools/documetantation on top of the raw SDK
   - Approach companies and communicate we have cool stuff → get hw
    - Lazslo is doing it for RIM (21 people signed up)


  - Common feedback from the appstore reviews?
   - Quite valuable feedback from the appstore comments
    - Users wanted localization for language apps


  - Adding screenshots and few words on the website?
   - Only desktop apps are on the website
   - There should be a link to the market


  - Evaluation: what metrics make a platform more potential than another for KDE Edu?
   - HW available?
   - Qt working?
   - kdelibs/KF working?
   - users


  - How to make the build simple for various platforms (i.e. helping developer, packagers, and so forth)?
   - We should aim for the same code base in master with a different cmake option
    - helps translations, jenkins, etc


  - Mobile edu logo (less detailed? Getting svgs so that is easier to handle forth and back?) 
   - Lazslo removed the "Grow your knowledge" text from the logo
    - Makes sense for removing clutter, "Grow your knowledge" might be too small to read properly
</pre>