[Playlist] xspf pagination

[Paul B Lamere]

Another use case for this is to implement a 'radio mode' like last.fm or 
Pandora provide.  For these types of services the playlist can be 
essentially infinite in length, akin to a radio station  If we had a way 
to link xspf playslist as Chris is suggesting, then we could easily 
implement infinite playlists with xspf. 

Note also that the XSPF flash player MusicPlayer does this via a 
parameter to the player called 'radio_mode'.  If this parameter is set 
to true, then it assumes that the last track of a playlist a really a 
link to the next playlist location.

Paul


Chris Anderson wrote:

>On 9/1/07, Ivo Emanuel Gonçalves <justivo at gmail.com> wrote:
>  
>
>>On 9/1/07, Chris Anderson <jchris at mfdz.com> wrote:
>>    
>>
>>>I'm thinking that perhaps using the link attribute,
>>>with rel="next" would be the best way to do it within the spec.
>>>      
>>>
>>That makes no sense, in my opinion.  It looks like a hack to go around
>>a certain program, and thus it is a bad idea.
>>    
>>
>
>The real world situation that suggests this is Grabb.it's search
>service, which returns the top twenty matches, and currently has no
>facility for requesting additional matches (other than refining search
>terms).
>
>My options in this case seem to be:
>
>(1 - the bad one) just have the Grabb.it client software know how to
>paginate through result sets, instead of following uris to get them.
>
>(2 - better) follow a uri in the xspf document to get the next page of
>results. the rel="next" method is valid XSPF and falls inline with the
>best practices in other documents and feed types that Lucas pointed
>out.
>
>I was hoping that some other XSPF client programmers might see the
>utility of being able to link between playlists in a reliable way.
>This is a small step in the direction of creating an ecosystem of
>playlists.
>
>A large step would be offering an OPML listing of a set of playlists.
>I'm happy to see some movement in this direction by sites like
>StreamRipper. Hopefully we'll see more of it, but I'm afraid the
>Myspace-flash-embed crowd might be not see the incentive. Walled
>gardens vs ecosystems, I suppose...
>
>  
>