- From: Phil Dawes <pdawes@users.sourceforge.net>
- Date: Tue, 21 Sep 2004 09:20:05 +0000
- To: Harry Halpin <hhalpin@ibiblio.org>
- Cc: Jon Hanna <jon@hackcraft.net>, www-rdf-interest@w3.org
Hi Harry, Harry Halpin writes: > > <http://www.hackcraft.net/foaf/> > > <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> > > <http://xmlns.com/foaf/0.1/Document> . > > <http://www.hackcraft.net/foaf/> <http://xmlns.com/foaf/0.1/maker> > > <http://www.hackcraft.net/jon/> . > > <http://www.hackcraft.net/jon/> > > <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> > > <http://xmlns.com/foaf/0.1/Person> . > > Now make the concept of "www.hackcraft.net/jon/" interoperable with > someone, say a me, who is using "www.ibiblio.org/hhalpin/jon" to refer to > you. And pretend we don't know each other :) How is a machine going to > discover that sameAs unless someone tells it that? > Using an IFP - e.g. foaf:mbox. E.g. Since Jon has a relatively small number of personal email addresses, you'll most likely use the same ones when describing him. Any knowledge setup worth its salt will be able to identify that you're refering to the same person by leveraging the foaf schema to note that foaf:mbox is inverseFunctional. To be honest I'm a bit lost as to why this thread is generating so many posts - these are non-problems as far as I can see. They are solved in the same way as they are solved in real life - using context. e.g. <http://example.com/pd> foaf:surname "Dawes" clearly indicates to anyone (man or machine) that understands the foaf:surname predicate that http://example.com/pd identifies a foaf:Person. Not a web page, audio file or anything else not compatible with foaf:Person (although a representation may be retrieved by HTTP GETting the URI, but that's a seperate matter). similarly: <http://example.com/pd> foaf:img <http://example.com/phildawes> indicates that <http://example.com/pd> identifies a foaf:Person, and <http://example.com/phildawes> identifies an foaf:Image. (the foaf prefix being shorthand for http://xmlns.com/foaf/0.1/ in all these examples) I can no more see why we need special URIs to identify certain types of 'thing' or 'Concept', than we need to put a special symbol at the start of a person's name to indicate that it identifies a person. Also, I don't think it will make much difference at this late stage. People (or rather 'agents') will continue to use http URIs to mean anything they like, as they have been for the past 8 years or so - you have to use context to attempt to deduce what they mean. Am I missing something? Cheers, Phil
Received on Tuesday, 21 September 2004 11:14:20 UTC