Re: gnome integration plan

["Zeeshan Ali (Khattak)" <zeenix gmail com>]

Hi,

>> We don't have a plan written in stone, but for sure we have one in our
>> mind ;)

  Yeah, I don't think you can afford to have it in stone if they
depend on lazy developers like me. :P

>> Our plan regarding that issue is:
>>
>> a) Finish rygel-grilo, which will expose Grilo content in DBus, so other
>> applications can access it (see [1]). As you know, this program
>> implements a spec[2] that have been changed quite a lot during last
>> times, so everytime I was almost to finish the implementation, spec
>> changed. Hope that now it is more stable.

  Well change is not always a bad thing. :) If you recall the first
big set of changes came by your (legitimate) requests. Believe me when
I say that I want this to become stable as much as you do.

>> b) Rygel-grilo will come with a small library that will help to
>> implement both producers and consumers following spec[2]. Actually,
>> rygel-grilo is using this library to produce content. Next step will be
>> to write a 'consumer' plugin for Totem.

  I also suggest you change the name of this package as by the time
other apps start using it for using grilo's plugins, there will be no
Rygel involved and having rygel in the name gives a very wrong
impression and might hurt your.. er.. our world domination plan. :)

>> c) Unfortunately, we do not have too much resources to try to write
>> plugins for other applications. We prefer to focus more in improving
>> Grilo and left these plugins for contributors.

  I totally understand but keep in mind that we went the same path
with gupnp and I later realized that it was more important to
implement application plugins rather than perfecting gupnp itself.

>> As you know, we proposed
>> a couple of them in GNOME GSoC, which unfortunately nobody took.

  Yes i recall because I suggested you do that. :) Thanks for trying
it but I must tell you that this happens to me each year as well. Not
many people in GNOME are interested in multimedia stuff and there are
retards who think adding facebook integration into banshee is more
important than a nice multimedia framework. :(

> Yeah, the idea is to finish first a) & b) and then write the plugin for Totem.

  It is entirely up to you in which order you do things but I must
point our that b) doesn't necessarily *completely* depend on a). With
the current spec (for which you have most of the implementation IIRC?)
you can create a and then b and then the totem plugin. If the spec
changes in between, just ignore it until you have everything ready.
Then you look at the spec for any changes and change implementation of
a). You can afford to break Rygel compatibility in between, thats just
fine. :)

> As for other apps, Joaquim told me he would be interested in writing the Rhythmbox
> plugin, but first he needs an implementation of b) that's more or less stable,
> that was some time ago though I hope he still wants to do it :). Also, he mentioned
> we would prefer to write it in Python but for that we should improve our introspection
> support and see how that works for Python.

  Fair enough. :) I really wish I had some time to help you guys out
but I don't see that happening anytime soon. The least I can do is try
to get the latest changes to the spec implemented asap for me.

-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124