RE: Rationalizing the term URI

Hi Norm,

> -----Original Message-----
> From: Norman Walsh [mailto:Norman.Walsh@Sun.COM]
> Sent: Thursday, January 23, 2003 4:08 PM
> To: www-tag@w3.org
> Subject: Re: Rationalizing the term URI
> 
> / "Tim Berners-Lee" <timbl@w3.org> was heard to say:
> | URI reference - a string used in a language to specify a URI, for which
> | relative form may be used where a base exists. ((This is not the only
way of
> | specifying the value of a URI - one can use various
> | character sets, namespace prefixes, etc))
> 
> Until the TAG solves rdfmsQnameUriMapping-6, I'd want to be very
> careful in what we say about namespace prefixes specifying URI
> references.

Yes...I feel equally cautious.

> Namespace prefixes are part of a syntactic device for constructing a
> QName which is a (URI, local-name) tuple, but they aren't URIs.

Some one will probably beat me up for this... to be very clear... in the
tuple above are you assuming the potential for 'URI' to include a (possibly
null) fragment id... ie the tuple is (URIReference, local-name) and the
(URIRef, local-name) -> URIRef mapping might get bit ugly if the LH URIRef
carries a fragment.

>                                         Be seeing you,
>                                           norm

Cheers,

Stuart