RE: Interpreting RDF as representation of a resource?

> -----Original Message-----
> From:
> []On Behalf Of ext Jan 
> Algermissen
> Sent: 20 November, 2004 22:19
> To:
> Subject: Interpreting RDF as representation of a resource?
> Hi,
> if one receives some RDF in response to an HTTP GET request to
> an HTTP resource, the RDF can be interpreted as a representation of
> the particlurar resource.
> What about the other way round?
> Suppose I am sending some RDF via HTTP POST to some 
> processing resource
> (e.g. a store)
> POST /foo/MyStore HTTP/1.0
> Content-Type: application/rdf+xml
> <RDF body 
> goes here>
> Is there any way for the processing resource to interprete 
> the received
> RDF as a representation of a resource of a certain *type*?
> (E.g. to store the data in the right table if the store is an 
> HTTP wrapped
> RDBMS or further dispatching based on the type)
> Maybe that sounds awkward, but I'd be thankful for any thoughts.

It seems to me that there is no presently defined way to use POST
to submit a representation to a web authority, such that the web
authority would know based on the HTTP request that the content
body of the request consists of a representation.

After all, that's what PUT is for.

Of course, with PUT, you'd have to utilize conneg in some fashion
if you wish to have an RDF/XML representation in addition to some
other representation.

And of course, PUT could not be used to augment/extend an existing
RDF/XML representation with additional knowledge. The client would
have to first GET the existing RDF/XML representation, extend it,
and then PUT it back.

Using POST though would surely require proprietary, application
specific features to treat the input as a representation, either
in the service/server itself, and/or in the client submitting
the request.


> Jan
> -- 
> Jan Algermissen
> Consultant & Programmer	

Received on Monday, 22 November 2004 10:17:05 UTC