Re: Interoperability Framework and Vocabulary

Dear Oscar,

> [...]
>
> I guess it's clear that my interest is not to do research on music
> ontology alignment/mappings (well, any help is very welcome here!), but
> to do something more, err... practical (useful?). That is, for instance
> once the ID3->MO, OGG->MO and iTMS->MO its done, to create a nice set of
> instances based on the MO, so people can start doing/testing funny
> SPARQL queries (such as [5]). Does this sound reasonable?

It sounds really reasonable and this is *exactly* what I have in mind !!!
I think also that the power of this use case is to really have this bridge
between the various metadata that come from different communitities, namely the
contextual information (ID3 tags, OGG tags, iTunes categories, ... and in the
future, why not the rating or the recommendations of your friends) +++ the
features you can extract from the signal.
The challenges here are dealing with heterogeneous metadata in their very
nature: concepts versus numerical data !!! Having ontologies that unify that,
like you say MO, is the key thing !
In one word, just go on, I think this use case will be a killer one!
Best regards.

    Raphaël

--
Raphaël Troncy
CWI (Centre for Mathematics and Computer Science),
Kruislaan 413, 1098 SJ Amsterdam, The Netherlands
e-mail: raphael.troncy@cwi.nl & raphael.troncy@gmail.com
Tel: +31 (0)20 - 592 4093
Fax: +31 (0)20 - 592 4312
Web: http://www.cwi.nl/~troncy/

Received on Friday, 23 February 2007 12:42:51 UTC