- From: Patrick Stickler <patrick.stickler@nokia.com>
- Date: Wed, 30 Oct 2002 10:26:18 +0300
- To: w3c-rdfcore-wg@w3.org, "ext Jeremy Carroll" <jjc@hpl.hp.com>, Patrick S tickler <patrick.stickler@nokia.com>
_____________Original message ____________ Subject: RE: rdfs:StringLiteral Sender: ext Jeremy Carroll <jjc@hpl.hp.com> Date: Wed, 30 Oct 2002 10:01:05 +0300 thanks Patrick I misspoke. Should all datatypes be subclass of rdfs:Literal; No. As that would mean that their value spaces are proper subsets of rdfs:Literal and e.g. an integer value is not itself a literal. In reality, the rdfs:Literal class is IMO a bug as it reflects a syntactic quality of an object, not it's semantics. But since in the case of inlined literals the semantics directly reflects the syntactic representation of the literal, we could simply shorten rdfs:StringLiteral to rdfs:Literal which I think would reflect the original intent regarding its meaning -- that all inlined literals are members of rdfs:Literal, and are self denoting. Then, rdfs:Literal is an rdfs:Datatype, but has no other datatype subclasses. Thus, rdfs:Literal, rdsf:XMLLiteral, and xsd:string are all unrelated datatypes. Patrick
Received on Wednesday, 30 October 2002 03:29:16 UTC