James Cridland wrote:
My team have produced another corker...
http://www.bbc.co.uk/music/beta is a lovely looking site, and contains
lots and lots of lovely APIs... more details at
http://www.bbc.co.uk/music/developers#RESTful
How splendid. Well done, chaps and chapesses.
This is just lovely! Nice work :)
And great to see MusicBrainz data making itself useful yet again.
Here's a thought I had about the per-artist pages. Maybe it doesn't
belong on the BBC site, but just to air the thought. In FOAF last year I
added a foaf:openid property, that allows you to represent which URIs
that are some agent's openid.
Now at the moment your data has this:
<mo:MusicGroup
rdf:about="/music/artists/cc197bad-dc9c-440d-a5b5-d52ba2e14234#artist">
<foaf:name>Coldplay</foaf:name>
<owl:sameAs rdf:resource="http://dbpedia.org/resource/Coldplay"/>
<mo:image
rdf:resource="/music/images/artists/7col_in/cc197bad-dc9c-440d-a5b5-d52ba2e14234.jpg"/>
<mo:musicbrainz
rdf:resource="http://musicbrainz.org/artist/cc197bad-dc9c-440d-a5b5-d52ba2e14234"/>
<mo:homepage rdf:resource="http://www.coldplay.com/"/>
<mo:myspace rdf:resource="http://www.myspace.com/coldplay"/>
...
</mo:MusicGroup>
What I'm thinking (about FOAF in general) is that rather than asking
users for their OpenID (since 0.1% of the population will understand
what one is), ... in some contexts we can probe for it automatically. So
here, we might look at the HTML for the homepage, myspace etc
properties, and determine if those documents also happen to be OpenIDs.
So here, we might look in http://www.coldplay.com to see if OpenID
markup is present (it's not). And we might look in
http://www.myspace.com/coldplay too. Nothing yet, however
http://www.techcrunch.com/2008/07/21/myspace-to-join-openid-bringing-total-enabled-accounts-to-over-a-half-billion/
...suggests that's on it's way. In which case (if the MusicBrainz data
is considered accurate enough), <foaf:openid
rdf:resource="http://www.myspace.com/coldplay"/> could also be emitted.
And the same for the other countless artists in there who have MySpace
profiles or other URLs that serve as OpenIDs.
What could this be useful for? Humm tricky. Let me be honest, I'm not
sure. Maybe nothing. But for example, if we have a nice way to look up
the openids for a specified artist, we have a way of creating Web pages
that only they (or their trusted agents) can access. This could be
interesting for the ecology of sites and Web developers around the BBC -
you could cheaply make a site that allowed BBC-featured artists to log
in for chat, messaging, etc connected to BBC content or services,
without a vast amount of central coordination. Or we could make use of
the information for tipjar purposes: some (eg. unsigned) artists might
be in a position to collect rewards from their audience in
unconventional ways. The OpenID would let the artist prove to any site
that they were indeed that artist. Or they might solicit interactions
from their fans without necessarily requiring all those fans to have
accounts on the giant Social Network sites (facebook, myspace etc). Or
the artists could use their openids to log in to musicky sites to offer
improved / corrected metadata about the artist, their works, gigs etc.
There's no particular reason for the BBC to offer the 'openid'
information rather than eg. MusicBrainz directly, but I think it might
be worth throwing a few braincells at potential apps...
cheers,
Dan
ps. related notes re a flickr-based scenario
http://danbri.org/words/2008/01/09/252
pps. the foaf:tipjar property came out of a conversation with Mike
Linksveyer of Creative Commons and Robert Kaye of MusicBrainz; at the
time we had nothing like OpenID. Having a way for artists to prove who
they are changes the possibilities rather interestingly...
http://rdfweb.org/mt/foaflog/archives/2004/02/12/20.07.32/
--
http://danbri.org/
-
Sent via the backstage.bbc.co.uk discussion group. To unsubscribe, please
visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html.
Unofficial list archive: http://www.mail-archive.com/[email protected]/
