Re: CURIEs

Hi there Norm,

Although your mail wasn't a formal last call comment to us, it got pointed  
out to us, and we discussed it anyway at our currently ongoing FtF.

> Of particular concern:
> - They open the door to a host-language defined prefix. So "foo" may
>   have one binding as a CURIE and a different binding as a QName and
>   may be used where a QName could be used.

Well, we define a data type for CURIEs, and say that there is a prefix,  
and the method of defining that prefix is host-language dependent. That is  
for the obvious reason that CURIEs may well be used in non-XML languages  
(which includes HTML in its non XML forms). It is not the intention per se  
to allow CURIEs and QNames to live in different worlds.

> - They say: "When CURIES are used in an XML-based host language,
>   prefix values MUST be able to be defined using the 'xmlns:' syntax
>   specified in [XMLNAMES]. Such host languages MAY also provide
>   additional prefix mapping definition mechanisms."
>  So x:foo as a QName might have a different binding than x:foo as a
>   CURIE, even in an XML document.

We have resolved to add wording that says that should a markup language  
provide more than one method of defining prefixes, that authors should not  
use them to bind different URIs to the same prefix.

Hope this helps.

Best wishes,

Steven Pemberton
For the XHTML2 WG

Received on Tuesday, 17 June 2008 13:59:45 UTC