- From: Frank Manola <fmanola@mitre.org>
- Date: Mon, 11 Aug 2003 08:31:37 -0400
- To: Patrick Stickler <patrick.stickler@nokia.com>
- CC: w3c-rdfcore-wg@w3.org
Patrick Stickler wrote: > Question: > > Can I specify rdfs:XMLLiteral as the value of rdf:datatype? I believe it's rdf:XMLLiteral, rather than rdfs:XMLLiteral (if not, both Primer and Concepts need to be revised, among other documents). > > Answer: > > Yes. Though it should be done with caution, particularly if > writing RDF/XML by hand, to ensure that the specified lexical > form is fully valid, as RDF parsers are not required to > check explicitly specified lexical forms. > Patrick-- Could you clarify "as RDF parsers are not required to check explicitly specified lexical forms"? Our general pitch on datatypes is that the proper lexical forms for a datatype are defined by the datatype. So while RDF can't determine whether a datatype URIref actually refers to a datatype, and can't determine whether a lexical form is valid with respect to a particular datatype, software designed to process that particular datatype can (and presumably must). In the case of rdf:XMLLiteral, the datatype in question is defined as part of RDF, and if we follow the same general model as for other datatypes, it seems to me that RDF software must be capable of both determining whether the datatype URIref rdf:XMLLiteral actually refers to a datatype (it does), and determine whether a lexical form is valid with respect to that datatype (that is, assuming we allow explicitly writing an rdf:XMLLiteral value as a typed literal, rather than using the parseType). If we're making an exception in the case of rdf:XMLLiteral, then it seems to me that the next FAQ needs to be "why" (I'm not saying there may not be good reasons, but this seems to break the general model)? --Frank -- Frank Manola The MITRE Corporation 202 Burlington Road, MS A345 Bedford, MA 01730-1420 mailto:fmanola@mitre.org voice: 781-271-8147 FAX: 781-271-875
Received on Monday, 11 August 2003 08:07:01 UTC