, Mar 1, 2005

Mark points to, and agrees with, a post by Hugh Winkler:

When you sit down to write a description language for REST services (a IDL or WSDL for REST), you discover that doing so is unnecessary.

It seems to me that it’s not the REST part, but rather the RDF idea that would make service descriptions unnecessary. I can’t see why describing the schemas resource representations are expected to conform to — if such an assumption is made — would make no sense. On the contrary, REST + Relax NG + Schematron would seem a smart choice to me.

(Don’t get me wrong, though; I’ve recently started reading up on Semantic Web topics, and actually like the concepts a lot — this is likely going to be a topic of future posts.)