- From: Graham Klyne <GK@NineByNine.org>
- Date: Fri, 17 Jan 2003 12:23:53 +0000
- To: Brian McBride <bwm@hplb.hpl.hp.com>
- Cc: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>, "Jos De_Roo" <jos.deroo@agfa.com>, <pfps@research.bell-labs.com>, <w3c-rdfcore-wg@w3.org>
At 11:31 AM 1/17/03 +0000, Brian McBride wrote: >At 11:59 17/01/2003 +0100, Jeremy Carroll wrote: > >[...] > > >>3: the semantics document should say clearly (I don't know if it does) that >>the denotation of an untyped literal without a language identifier is the >>unicode string. >>(Dan felt strongly about this, and no one else objected) > > > http://www.w3.org/2001/sw/RDFCore/TR/WD-rdf-mt-20030117/#urisandlit > >[[he meaning of a literal is principally determined by its character >string: it either refers to the value mapped from the string by the >associated datatype, or if no datatype is provided then it refers to the >literal itself, which is either a unicode character string or a pair of a >string with a language tag.]] At risk of repeating myself, I still think there's a lack of clarify here, when all relevant documents are read together. The above text can be read that a character string may be some part of a literal, so a literal is not actually a character string. Later, the same document says that for a plain literal, I(E)=E, so whatever a plain literal may be in the abstract syntax is also its denotation. In view of Jeremy's comment, I think the WG intent is that a plain literal without a language tag denotes a Unicode character string. In light of I(E)=E, that means said literal in the abstract syntax must be a Unicode character string. I don't think that's clear. It wouldn't take much to make it so. I'm not yet fully understanding what fix Peter Patel-Schneider is requesting, but I suspect that whatever that is may also clarify this matter in some way. #g ------------------- Graham Klyne <GK@NineByNine.org>
Received on Friday, 17 January 2003 07:21:22 UTC