Re: help wanted: RDF issue rdfms-assertion

>At 11:55 PM 6/11/02 -0400, patrick hayes wrote:
>>.... The implication I draw is that I'm not hearing what you are 
>>meaning to say, that we aren't communicating properly. (Since I 
>>know that quite a lot of people are understanding this kind of talk 
>>in roughly the same way I am, I would also suggest that there might 
>>be a wider breakdown of communication.)
>
>Following this debate, it seems clear to me that there *is* miscommunication.
>
>I think there are at least three notions of "meaning" being kicked 
>around here:
>
>- social meaning - the default English interpretation.  Roughly, 
>what Quine describes as truth of a statement that "a native is 
>prepared to assent to".
>
>- RDF logical meaning - truths of RDF statements that are true by 
>virtue of the specified semantics of RDF:  truths that can be 
>calculated by machines using only the rules set out in the RDF 
>formal semantics specification.  Again appealing to Quine:  RDF 
>statements that are true by virtue of their RDF-grammatical 
>structure.
>
>- RDF received meaning - truths that are expressible in RDF, but 
>that are grounded in socially normalized intended interpretations. 
>If certain such truths are accepted as axioms, then the logical 
>machinery of RDF allows other such truths to be deduced.  I think 
>this is the notion of meaning that Tim is trying to promote as "RDF 
>meaning".
>
>#g
>
>(Hoping I've not merely added to the miscommunication.)

No, I think this is useful. Would you agree that 'received meaning' 
can be characterized as the social meaning of any logical 
consequences? That gives a clean characterization which I think is 
what Tim is getting at. If you publish a graph G and G  entails G', 
and we interpret G' using the same social conventions that everyone 
agrees could be reasonably used to interpret  G, then you are 
asserting that content of G' as well.  Tim wants to prevent human 
publishers of RDF content from wriggling out of their 
mechanically-inferred social obligations; I want to be clear that the 
machines doing the inference aren't expected to know what all this 
human stuff is about.

Pat
-- 
---------------------------------------------------------------------
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 13:03:08 UTC