Re: "Including" other RDF and RDFS files

>pat hayes wrote:
>>I have mixed feelings about this. It IS a neat idea and is widely 
>>used. On the other hand, if it gets used too cleverly then it will 
>>violate the RDF spec, since it can easily produce a completely 
>>different logic which doesnt mix with the standard RDF inference 
>>machinery. Well, OK, so let 10|3 flowers bloom, is one reaction. 
>>BUt speaking as one of the standards-writers its hard for me to 
>>live with that without complaining.
>My proposal [1] has nothing to do with logic.  A semref from one RDF 
>document to another is just like a like an <a href> from one web 
>page to another.  There are no logical entailments implied or wanted.

Oh, OK. BUt then that sounds very like SeeAlso to me, and we have that.

>  If people want logical entailments from their references to graphs 
>in other documents, they should use a logical language which implies 
>what they want.  Perhaps something like owl:imports [2].
>By your own admission, "There are clearly syntactic and operational 
>boundaries between graphs."  Well point is we need some standard 
>arcs in our graphs that relate to those syntactic boundaries so that 
>our programs will have the facts to operate on.  The WG ignoring 
>that  need, and semingly telling us that they are only concerned 
>with logic programming, is not helping  the interoperablity of 
>applications that are not necessarily based on logical inference. 
>Last time I looked that was about 95 % of the RDF which was being 
>used for practical matters.  
>>Sure, it kind of makes sense, but not everyone else uses it that way.
>Well that feels very much like "Let them eat cake".    We *do* want 
>people to use the same property to refer from one RDF graph to 

But wait. What do you mean here exactly? The references (if that is 
the right term) TO the RDF graph will be the same, since that graph 
is in a document (probably in XML) that will have a URL. The 
references INSIDE the graph will also be the same, since they are 
urirefs and if you use the same uriref in your graph, then its the 
same uriref (eg if you use rdf:type in your graph, its still 
rdf:type); that doesn't require any semrefing or importing, its just 
part of the global WWW URI stricture that is common to all web pages. 
And you don't want any new kinds of inference. So I must be missing 
something here: what is that you want, that isn't already there?


>  and for that to happen the WG should (imho) put semref in RDFS.  
>Brian, if there is some more official way for me to ask for this 
>(whether it is taken up by the current WG or posponed to some future 
>rework) could you let me know ?
>Seth Russell

IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola              			(850)202 4440   fax
FL 32501           				(850)291 0667    cell	

Received on Monday, 30 September 2002 14:06:42 UTC