- From: Graham Klyne <Graham.Klyne@MIMEsweeper.com>
- Date: Wed, 10 Oct 2001 16:58:59 +0100
- To: Dan Connolly <connolly@w3.org>
- Cc: Pat Hayes <phayes@ai.uwf.edu>, w3c-rdfcore-wg@w3.org
At 08:32 AM 10/10/01 -0500, Dan Connolly wrote: >Perhaps I overstated the case... the evidence gathered by Art[1] >that RDF 1.0 tools are all over the board on literals puts >very little weight on any argument of the form "RDF 1.0 implementors >have understood the spec to say XYZ...". > >Meanwhile, I've gotten a clarification of how DAML+OIL treats >literals, and I've seen claims that it's been implemented. > >So I think we're not very much constrained by history. >In some ways, I lean toward saying we should strike >parseType="Literal" and xml:lang from the RDF 1.0 spec >on the basis that there's insufficient implementation experience, >just like aboutEachPrefix, and come back to richer >literals and datatypes in RDF 1.x. > > >[1] http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0095.html >Mon, 8 Oct 2001 16:29:11 -0400 Concerning xml:lang, I agree that would be cleaner. Concerning parseType="Literal", I think there's still a difficulty: <rdf:RDF ...> <eg:Node> <eg:literalProperty parseType="Literal"> <eg:NotANode> <eg:notAProperty>some text</eg:notAProperty> <eg:NotANode> </eg:literalProperty> </eg:Node> </rdf:RDF How would this be expressed in absence of parseType="Literal"? #g ------------------------------------------------------------ Graham Klyne MIMEsweeper Group Strategic Research <http://www.mimesweeper.com> <Graham.Klyne@MIMEsweeper.com> ------------------------------------------------------------
Received on Wednesday, 10 October 2001 12:25:48 UTC