From: <herman.ter.horst@philips.com>

Date: Fri, 10 Oct 2003 13:36:14 +0200

To: pat hayes <phayes@ihmc.us>

Cc: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>, w3c-rdfcore-wg@w3.org

Message-ID: <OFE9C0A9BD.C370C622-ONC1256DBB.003EBB2C-C1256DBB.003FD437@diamond.philips.com>

Date: Fri, 10 Oct 2003 13:36:14 +0200

To: pat hayes <phayes@ihmc.us>

Cc: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>, w3c-rdfcore-wg@w3.org

Message-ID: <OFE9C0A9BD.C370C622-ONC1256DBB.003EBB2C-C1256DBB.003FD437@diamond.philips.com>

>Adding these URIs to IP ... >is harmless for the other (elementary) Herbrand >results, so I have restored this condition. I do not believe that this is harmless for the Herbrand lemma, in particular the only if side: If I satisfies G then H(G) << I In order to prove this only if statement, some mapping A : blank(G) -> IRI is chosen such that I+A satisfies G. Then, k : IRH u IPH -> IRI u IPI is defined to be the restriction of I+A to IRH u IPH. For a URI v such that v type Property is in G but with v not being the property of a triple in G, the proof that k(v) is in IPI seems to need the first condition on RDF interpretations. However, I is an arbitrary simple interpretation. HermanReceived on Friday, 10 October 2003 07:37:18 UTC

*
This archive was generated by hypermail 2.3.1
: Wednesday, 7 January 2015 14:54:08 UTC
*