- From: Brian McBride <bwm@hplb.hpl.hp.com>
- Date: Tue, 12 Jun 2001 11:54:48 +0100
- To: Jan Grant <Jan.Grant@bristol.ac.uk>
- CC: Aaron Swartz <aswartz@upclink.com>, rdf core <w3c-rdfcore-wg@w3.org>
Hi Jan, I understand the problem. If we can get a permanent directory structure for test cases then the problem will go away. How about for the moment we assume that all test cases have uri's of the following form: http://www.w3.org/2000/10/rdf-tests/rdfcore/[issue-id]/testnnn.rdf and results are http://www.w3.org/2000/10/rdf-tests/rdfcore/[issue-id]/testnnn.nt where [issue-id] is e.g. rdfms-not-id-and-resource-attr Brian Jan Grant wrote: > > OK, one more try. I'm pretty ambivalent about the xml:base issue. > However, because the value of rdf:id is syntactically an XML name > symbol (not a URI), which appears to get resolved into an absolute URI > dependant on the base URI of the RDF it appears in, I _can't_ post my > test cases and results to the list in the same email unless I have a > moment of prescience, wherever they use nonabsolute URIs. > > We agreed that ntriples should use absolute URIs. So while Aaron's > output is highly suggestive: > > > <#foobar> rdf:subject _:genid . > > the subject there looks illegal to me. > > That's why xml:base appears in the rdf:RDF element containing a couple > of test cases. If someone has an easy way to fix this, please let me > know. If that means extending ntriples to use this URI-base-unaware > notation, fine, but I'd like to see a definitive > draw-a-line-under-it-and-go-no-further ntriples definition. > > Cheers, > jan > > -- > jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/ > Tel +44(0)117 9287163 Fax +44 (0)117 9287112 RFC822 jan.grant@bris.ac.uk > __/\____/\_____/\____/|_____________________________________ flatline
Received on Tuesday, 12 June 2001 06:56:32 UTC