Re: are all strings in a sequence valid potential QNam

On Wed, 2010-02-03 at 21:25 -0500, ac wrote:
> Hi,
> 
> Complaining may not be so constructive but it may seem that expressing 
> opinions and viewpoints with logic, experience, and clear sight may be.  
> The objective should not be determining who is right but rather what is 
> best for most in the long term.

The on-topic part here is that XSLT 2, XQuery and XPath 2 have support
(optional) for XML 1.1, and to deal with that you have exactly the same
issues as for XML 5th edition in terms of qnames.

In addition, there's no way in an XPath expression to enquire about the
version of XML in the XML declaration of target documents, so it seems
to me the best choice is to allow the widest set, and rely on the XML
parser to complain if appropriate.  It does have an effect on
tokenization, of course, but the XML 1.2 suggestions would have had
exactly the same effect if implemented as XML 1.1 and XML 1.0 5e in
this regard. 

It's my full-time job to try & promote interoperability in XML (OK, it's
_part_ of my job), and that includes trying to encourage developers to
write code that will accept the full range of XML documents...

Best,

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org www.advogato.org