Re: QName URI Scheme Re-Visited, Revised, and Revealing

> If you're suggesting that each spec that uses QNames
> should say how to map them to URIs, I heartily agree!

Ugh, I hadn't considered that problem. For a while there, I naturally
assumed that if an XML language doesn't declare QNames that it uses to
anything "special" outside of the XML NS appendix on namespace partitions,
then it used that partioning mechanism... but it's non-normative, and they
don't (not "and therefore", just "and").

If there could be some consistent way to denote how QNames are used
per-XML-application, that would be pretty neat. Otherwise, it appears as if
my original suggestion of representing the QNames by modelling them using a
set of XML application specific propeties is more attractive than defining
a new QN URI scheme. It would probably be most useful if XML application
specifications would define properties for any new QName usage idioms.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .

Received on Thursday, 23 August 2001 11:07:58 UTC