Can we map IANA mimetypes to URIs?

Henry has started a thread on the semantic-web mailing lists on the subject. Henry and I were brainstorming how to simplify the Atom RDF content construct in the Atom-OWL project. Currently we have the following:
[] :content [ a :Content;
               :type "text/html";
               :value "<b>hello</b>" ] .
But we would like to do this instead (leaving the xml:lang issue aside):
[] :content :value "<b>hello</b>"^^ianatext:html .
where ianatext:html would resolve to http://www.iana.org/assignments/media-types/text/html.

There are many unanswered issues with the approach:

  • Not all of the media types return an HTTP 200, some are 404.
  • Can we make the assertion that ianatext:html is a rdfs:Datatype in an Ontology? Peter says we can.
  • Should ianatext:html be a sub-property of rdfs:XMLLiteral?
  • Will we have the need for wildcards? ianatext:*
  • How do we solve the xml:lang for Atom plaintext and HTML because literals can’t be both datatyped and have a language?

Any opinions/suggestions are gladly welcome since building an ontology is not trivial even when we have a very clear specification to start. In the meantime, I’ll keep focusing on my Abdera/RDF Atom endpoint implementation, coming soon!

Spoiler: It can post,get,put,delete. Automatic collection creation, supports introspection, collections and entries (both edit and alternate), JSON/RDF support, SPARQL endpoing, ETags and last night you can use the GData client to interact with it (get/insert, update gave me an error but I have to go look for a minivan).

  • co.mments
  • connotea
  • del.icio.us
  • digg
  • Ma.gnolia
  • Reddit
  • scuttle
  • Technorati
  • YahooMyWeb

About this entry