Bangarang Active?

[Andrew Lake]

Sebastian Kügler wrote:
> ... I've had a quick look at Bangarang a few weeks ago, in
> order to evaluate its usefulness for touchscreens. I think the main issues
> I've found were too small controls, and too little gesture support (swipe-
> scrollling in playlists, for example). There is, from the top of my head also
> quite some interesting stuff in tooltips, which is close to impossible to
> sensibly use on touchscreens.
>
> Overall, it didn't feel miles away from a good touch-screen application, the
> applications look & feel is already quite OK. It's a really nice application,
> congrats to that.
Oh nice, thanks much for taking the time to test! It'll help me
prioritize a few areas to tackle.

Sebastian Kügler wrote:
> And with "yes it is" (which is the wrong answer to "is Nepomuk out of scope?"
> Marco means: Nepomuk is there on Plasma Active. :-) Contour uses it as well,
> and I'm working on what once will become a set of data-driven "widgets"
> (Plasma Quick components), that are backed by Nepomuk data. I've just
> yesterday committed a metadata Plasma::DataEngine to kde:plasma-
> mobile/dataengines/metadata. An bare bones applet (which is kinda buggy) can
> be found in applets/resourcelist, just to give you an idea how this kind of
> stuff could look like in QML / Plasma Quick.
Ah, that clears things up, thanks.  I'll take a look at the repo since
it'll help me get up to speed with QML/Plasma Quick.

Sebastian Kügler wrote:
>> Andrew Lake wrote:
>> > - I was hoping for the 2.1 release to focus on activity-aware
>> > playlists and media semantics (highest rated/frequently
>> > played/recently played). Would something like this fit into the
>> > current vision for Plasma Active?
>
> Absolutely. This might also tie in nicely with the Contour recommendation
> system Sebastian Trüg is working on. You guys should talk =)
>
> [Marco Martin wrote:]
>> activities are pretty important in plasma active, as the name suggests
Cool!  I have a few ideas about context-aware media semantics.  It'll
be helpful to see where it'll make sense to implement app-specific
solutions vs solutions lower in the stack.

Sebastian Kügler wrote:
>> Andrew Lake wrote:
>> > - There are some specific interaction mechanics that I'm not entirely
>> > clear about regarding availability from current/proposed Active
>> > libraries: Multiple selection in a list, Touch and hold, Pop-up
>> > menu/selection list.
>
> We're still at a very early stage with respect to standard interaction
> schemes, unfortunately. I'll bring this topic up at the upcoming UX sprint,
> and we will sit together during Tokamak to identify these UI design patterns
> for which we need standard, high-level solutions for. You raise a very good
> point. :)
Ok then I'll try to see if there are interaction mechanics that might
be useful for the Bangarang class of use-cases. :-)

Sebastian Kügler wrote:
>> Andrew Lake wrote:
>> > - I may need a little guidance on whether to modify the existing C++
>> > UI code or just develop a new UI using QML. No worries either way.
>>
> [Marco Martin wrote:]
>> I would say a QML ui with plasma components would be the way to go..
>> for media apps there is still the problem of video widgets in qgraphicsview
>> that are still.. meh :/
>
> Worst case: We'll use a window atom to place a windows on top of the canvas,
> for fullscreen, we'll want that anyway. But yeah, meh.
Funny, I encountered that QGraphicsView challenge back when I started
working on Bangarang two years back. No worries though, I'll do the
best I can with whatever's available. Ok, I'll go get caught up on QML
now. :-)

Thanks much Sebastian and Marco for taking the time to respond.

peace and much respect,
Andrew