- From: Andrew Eisenberg <andrew.eisenberg@us.ibm.com>
- Date: Tue, 10 Apr 2007 20:45:17 -0400
- To: public-xhtml2@w3.org
- Cc: w3c-xml-cg@w3.org
- Message-ID: <OF98EBF0FD.9D81C07B-ON852572BA.0002D889-852572BA.00042880@us.ibm.com>
The XML CG has reviewed CURIE Syntax 1.0 [1] and offers the following
comment:
The Namespace specification defined QNames and allowed their use as
element names and attribute names. XML Schema defines xs:QName and
xs:anyURI types and allows these types to be used to define attributes and
elements. The XQuery 1.0 and XPath 2.0 Data Model (XDM) uses the types
defined in XML Schema.
The XML CG observes significant changes between the Working Draft of 7
March 2007 and the earlier draft that we previously saw. In general,
those changes ameliorated our earlier concerns, but we believe that the WD
should go a little further. In the (non-normative) sections 1
(Introduction) and 1.1 (Existing Use of CURIEs), misuse of well-defined
terms occurs several times. In particular, those sections incorrectly use
the terms "QName" and "URI". Even though the normative portions of the WD
no longer misuse those terms, we believe that the non-normative sections
should clearly state that it is inappropriate to use those terms for the
purposes addressed by this specification.
We remain concerned that CURIEs are being defined for use in XML grammars
without defining precisely where they can occur, without defining their
Infoset contributions, and without extending XML Schema. If these objects
are of significant importance in the XML community, it is appropriate to
propose that XML Schema define them for more general use (including in the
XQuery 1.0 and XPath 2.0 Data Model (XDM)).
-- Andrew
[1] CURIE Syntax 1.0
http://www.w3.org/TR/curie
--------------------
Andrew Eisenberg
IBM
4 Technology Park Drive
Westford, MA 01886
andrew.eisenberg@us.ibm.com
Phone: 978-399-5158 Fax: 978-399-7012
Received on Wednesday, 11 April 2007 00:45:27 UTC