- From: Tim Berners-Lee <timbl@w3.org>
- Date: Wed, 21 Mar 2001 21:26:39 -0500
- To: <jos.deroo.jd@belgium.agfa.com>, <mdean@bbn.com>
- Cc: <www-rdf-logic@w3.org>, <boley@informatik.uni-kl.de>
Jos, Thanks for converting this into N3. I had to make two changes to use it. In the rules, for cwm you have to state that the rule is true - the "forAll" doens't do that for you. The other thing is that you can't omit the colon when you use a locanname with a void namespace prefix. My versions are listed in http://www.3.org/2000/10/swap/Examples.html (where is the original of this from?) Tim ----- Original Message ----- From: <jos.deroo.jd@belgium.agfa.com> To: <mdean@bbn.com> Cc: <www-rdf-logic@w3.org>; <boley@informatik.uni-kl.de> Sent: Sunday, March 04, 2001 6:06 PM Subject: gedcom-relation example > > Mike, > > We have been playing a bit with your GEDCOM example. > We actually rewrote it in N3 and the relations are at > http://www.agfa.com/w3c/euler/gedcom-relations.n3 > To have a practical example, some facts are at > http://www.agfa.com/w3c/euler/gedcom-facts.n3 > We did all kinds of queries such as at > http://www.agfa.com/w3c/euler/gedcom-query.n3 > and found all kinds of proofs such as at > http://www.agfa.com/w3c/euler/gedcom-relations.n3 > We had to fix (actually simplify) cycle detection. > Recursions and mutual references seem to work ... > Maybe the proofs can give an idea about possible > simplifications of the relations ... > > -- > Jos De Roo, Agfa http://www.agfa.com/w3c/jdroo/ >
Received on Wednesday, 21 March 2001 21:27:00 UTC