Is this valid use?

I was wondering if we can use RDF to describe the input and output
parameters of a method.

For example,

<mycompany:MethodInput>
      <rdf:Bag>
	<rdf:li resource="http://www.mycompany.com/businesslogic#Item"/>
	<rdf:li resource="http://www.mycompany.com/businesslogic#Supplier"/>
	<rdf:li resource="http://www.mycompany.com/businesslogic#ShipTo"/>
	<rdf:li
resource="http://www.mycompany.com/businesslogic#DigitalSig"/>
      </rdf:Bag>
</mycompany:MethodInput>

<mycompany:MethodOutput>
      <rdf:Bag>
	<rdf:li resource="http://www.mycompany.com/businesslogic#OrderNo"/>
      </rdf:Bag>
</mycompany:MethodOutput>

The idea here is to define the input parameters and output parameters as a
resource. One could argue why not use schema to define your types, but I am
more interested in machine discoverable semantics.

For example, the resource Supplier could have the following statements about
it:

Supplier should have registered with mycompany.
Supplier should contain a digital signature.

So, when the consumer of Method calls, his intelligent agent will ensure
that these rules are met for the input supplier. 

Does any of you see this as a useful exercise to do or I am just talking
trash? :-).

I am also interested in knowing if any of you have started applying RDF to
Web services.

Best Regards,
Sateesh

Received on Tuesday, 24 April 2001 18:32:24 UTC