- From: Graham Klyne <gk@ninebynine.org>
- Date: Mon, 12 Jul 2004 13:32:39 +0100
- To: RDF comments <www-rdf-comments@w3.org>
(This comment is being presented for possible inclusion in the RDF errata page http://www.w3.org/2001/sw/RDFCore/errata, and also as feedback to any future working group that may be chartered to do further work on the RDF specification.) With reference to: http://www.w3.org/TR/2004/REC-rdf-syntax-grammar-20040210/ Section 7.2.11 contains this text: [[ If there is an attribute a in propertyAttr with a.URI == rdf:type then u:=uri(identifier:=resolve(a.string-value)) and the following tiple is added to the graph: ... ]] but in section 6.3.3, resolve is defined as: [[ resolve(e, s) A string created by interpreting string s as a relative URI reference to the ·base-uri· accessor of e as defined in Section 5.3 Resolving URIs. The resulting string represents an RDF URI reference. ]] i.e. resolve is a 2-argument function. I assume that the section 7.2.11 text should read as: [[ If there is an attribute a in propertyAttr with a.URI == rdf:type then u:=uri(identifier:=resolve(e,a.string-value)) and the following tiple is added to the graph: ... ]] where e is defined previously as the element event to which the rdf:type attribute is attached. ... Also, looking at production 7.2.11 and 7.2.25, it seems that it would be more consistent if the rdf:type attribute were defined by a separate production, rather than being singled out as a specific value in section 7.2.11. #g ------------ Graham Klyne For email: http://www.ninebynine.org/#Contact
Received on Monday, 12 July 2004 18:26:29 UTC