Re: Datatyping literals: question and test cases

[Patrick Stickler, Nokia/Finland, (+358 40) 801 9690, patrick.stickler@nokia.com]

> >"foo"-"" is not an xsd:string.
> 
> Oh, yuk. I would vastly prefer that the blank lang tag is not in the 
> denotation. This seems to be shooting ourselves in the foot; with 
> this system, it is *impossible* to write an xsd string as a literal.

Er, what about

   "foo"^^xsd:string

Looks like an xsd:string literal to me...

If you mean that inlined literals (non-explicitly datatyped) would
be interpreted to be xsd:string's, then that's not allowed anyway.

I agree, though, that is very odd for the lang tag to participate
in the L2V mapping. It's really a scoping issue, not a characteristic
of the lexical representation.

Patrick

Received on Thursday, 31 October 2002 02:16:30 UTC