Re: [Playlist] XSPF -> RDF - using GRDDL

[Apologies if you get this twice. The W3C mailing lists don't like my  
usual e-mail address, so I'm resending it with a different address.]

I've done some work with RDF and playlists, but only tangentially  
related to the topic at hand.

Cognition <http://buzzword.org.uk/cognition/> is able to parse hAudio  
<http://microformats.org/wiki/haudio> into RDF using Manu Sporny's  
Audio RDF vocabulary <http://purl.org/media/audio#>. It can then  
convert data from Audio RDF to M3U playlists. Cognition does support  
GRDDL (though see the proviso in the next paragraph), so  
theoretically if the XSPF GRDDL transformation provided its output as  
Audio RDF, Cognition could be used to convert XSPF to M3U.

The proviso is that Cognition currently doesn't support all of GRDDL  
- it only supports rel=transformation and profileTransformation, not  
XML namespace transformations. This is because Cognition is really an  
(X)HTML parser, so doesn't support generic XML is an input - I am  
looking to change that in the next couple of versions though, and  
support non-(X)HTML input. So until Cognition supports non-(X)HTML  
input, and the rest of GRDDL, it won't actually be able to convert  
XSPF to M3U.

I've got a couple of pages the demonstrate Cognition's M3U output.  
One uses hAudio; the other uses Audio RDF embedded in XHTML using RDFa:

http://buzzword.org.uk/2008/audio/

-- 
Toby A Inkster
<mailto:mail@tobyinkster.co.uk>
<http://tobyinkster.co.uk>

Received on Saturday, 11 October 2008 16:35:13 UTC