Re: MISC: Internet Media Type registration: proposed TAG finding

>    [Jim Hendler]
>    I can't help thinking that our problem is rhetoric and not
>    technology.  ....
>
>It sure seems that way to this spectator of the conversation.
>To end this dispute about terminology, let's just cede "RDF agent" to
>the logicists, and use another term, say, "RDF-based agent," for an
>inference engine that uses RDF as its input and output notation.
>
>The objection that there are no a-priori limits on what an RDF-based
>agent can infer seems silly to me.  Each RDF-based agent is to be
>proposed in its own context and evaluated on whatever merits are
>claimed for it.

In which case, what is the point of even proposing to use RDF? What 
'context' can there be for a semantic web agent, other than the 
entire semantic web? And isn't the whole point of the exercise to 
define a machine-processable notion of 'meaning' which can be used 
across the whole web, so that the writer of the meaning cannot 
possibly know in advance the 'context' of its use by some other 
software agent? And isn't the point - indeed, ultimately, the only 
possible point - of the RDF (and DAML+OIL and OWL and ...) standards 
to provide a uniform specification for that machine-processable 
content? And what other uniform specification can there be, other 
than the formal semantics of the language (and any processing models 
which are conformant to that formal semantics, of course)? And 
wouldn't it then be the case that any agent which drew inferences 
that were not valid according to the RDF specifications would in fact 
NOT be conformant to those specifications?  I would certainly hope 
so. Don't get me wrong: such a nonconforming engine might be very 
useful, and I might use it myself. But I would like to know when it 
is using RDF and when it is just guessing.

Pat Hayes
-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes

Received on Wednesday, 12 June 2002 01:41:29 UTC