Ampache service, was: Re: extragear/multimedia/amarok/src

[Ian Monroe]

On Wed, Apr 2, 2008 at 8:46 AM, Nikolaj Hald Nielsen
<nhnfreespirit at gmail.com> wrote:

>  In that case, you will have to get someone to improve the Ampache API
>  and the Ampache service. I do not know if the student who originally
>  proposed to work on the Ampache service is interested in this as that
>  is pretty far from what he applied for.

The student has been flexible so far. It's still the same basic idea:
add stuff to Ampache to support new features in Amarok. As someone who
has been a student twice, my concern (outside of getting the GSoC to
begin with) was that I have a well-defined task, not so much what the
task was. Eg I never needed to use DAAP (my summer 06 GSoC), and
certainly not with iTunes, but it was a very well-defined GSoC so I
was satisfied with that.

>  My big problem now is, what about services where it is simply not
>  possible to make the QueryMaker flexible enough to fulfill the
>  requirements or use in Smart/Dynamic playlists? Because of the way
>  some of them work, or the API's they are based on ( that we cannot
>  change in the same way as the Ampache one ) reaching this level of
>  integration will simply not be possible for many services, and will
>  not make sense for others. Should we discard these completely? If not,
>  how do we explain this difference to the user?

Well when we get around to having Smart Playlists, the user would have
some way to pick what collections to draw from and it simply wouldn't
be there. Most of the services that won't show up you wouldn't expect
to show up. Searched-based services, shoutcast. We'll probably get a
bug report for Mp3tunes not showing up, but that's justified, as
that'd be a good goal to work with mp3tunes to do that in the future.

>  I cannot help but feel that my ideas and vision which has carried the
>  service framework and most of the services this far, are being pretty
>  lightly brushed aside at the moment as everyone all of a sudden has an
>  opinion on what is most important for the services... :-(

This isn't all of a sudden, like Mark said that was his vision for 2 (and mine).

>  As I have written before, I am in no way opposed to making a project
>  that improves the Ampache API and the service enough that it can be
>  used as a first class collection in Amarok 2, but I am finding it very
>  hard to accept that the general consensus seems to be that this has to
>  come at the cost of my ideas, and the project that a student has
>  applied for to implement these.

It probably does mean ampache uploading will be put off until later,
but it doesn't seem so dramatic.

>  - Nikolaj

Ian