Re: [Rhythmbox-devel] Re: usage question

[Ben Konrath <ben bagu org>]

I don't think that Christophe Fergea's response made it to the list, so
my comments are interleaved without chopping.

On Tue, 2004-06-08 at 14:15, Christophe Fergeau wrote:
> > Another reason for a plug-in system is that not all users want to use all
> > the features. For example, I wouldn't use a visualization system if such a
> > system were to ever make it into rhythmbox.
> 
> You didn't mention visualization in your examples, I was specifically
> talking about the list that was in your mail.

It wasn't me who wrote the original mail ... I was just butting in :)

> >  Also, the plug-in system could
> > potentially speed the development of plug-in especially if there was a
> > scripting interface as suggested earlier.
> 
> A scripting interface would probably entice some people into doing some
> cool stuff, but I don't think it would help at all with getting things
> like audio cd support or tag editing more quickly.
> 
> > 
> > Furthermore, I have noticed that Colin Walters has hinted a few times that
> > some features in rhythmbox are motived by a need to replace XMMS. If this
> > is the case, I believe that some sort of plug-in architecture would be
> > required to allow the "power user" to extend rhythmbox as they feel fit.
> > 
> 
> If rhythmbox UI doesn't fit people's need wrt an xmms replacement, I'm
> not sure a plugin architecture would really help them.

You're right in that a plug-in architecture probably wouldn't help the
UI too much. What I was really trying to say is that if XMMS replacement
is on the agenda, the UI model that XMMS implements isn't the only thing
that is going to be missed. The plethora of optional plug-ins available
for XMMS is something that I would personally would miss. 

As suggested in another thread, investigating if the bonobo interface
could be used as the plug-in interface might be worthwhile.

Cheers, Ben