Re: [xml-dev] XHTML 2 Working Group won't be renewed?

["Pete Cordell" <petexmldev@codalogic.com>]

Original Message From: "Michael Kay"

>> - I would allow the use of the namespace directly, eg
>> <com.mycomp:elem> instead of forcing the definition of a
>> prefix <m:elem xmlns="com.mycomp">
>
> There's actually a debate going on in the XSLT/XQuery groups at the moment
> about allowing a syntax for QNames ...
> Clark notation {uri}local is one candidate, ...
> I've been proposing backticks: `uri`local or `uri`:local ...

Why do you need to wrap the URI in any characters at all?  Why not just do:

<http://www.mycompany.com/myschema:local/>

I know there are two : characters in there, but you just have to specify 
that it is the last occurence of the : character delimits the local part 
from the namespace.

When namespace prefixes come into play they are just treated them as aliases 
of the URI they represent.

A bit like in C# (and presumably Java) you can specify:

    private Microsoft.Office.Interop.Word.Application  word = null;

and there's no ambiguity about which is the local part.

And you can set up an alias as:

    using Word = Microsoft.Office.Interop.Word;

and do:

    private Word.Application  word = null;

There's also the issue that there is an abundance of / characters in the 
original example name, which may conflict with XML parsing of elements such 
as <empty/>.  But I assume that would be a problem for the {uri}local type 
schemes also.  And it doesn't seem too difficult to come up with rules that 
would allow the meanings of the / characters to be disambiguated.  (e.g. If 
there is a : any / before the last : is part of the URI, not part of the XML 
structure.)

Regards,

Pete Cordell
Codalogic Ltd
Interface XML to C++ the easy way using XML C++
data binding to convert XSD schemas to C++ classes.
Visit http://codalogic.com/lmx/ or http://www.xml2cpp.com
for more info
>