W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > January to March 2009

[Bug 6522] Please un-deprecate the the namespace http://www.w3.org/2001/XMLSchema-datatypes

From: <bugzilla@wiggum.w3.org>
Date: Wed, 04 Feb 2009 17:37:23 +0000
To: www-xml-schema-comments@w3.org
Message-Id: <E1LUlgh-0000nx-Ff@wiggum.w3.org>

http://www.w3.org/Bugs/Public/show_bug.cgi?id=6522





--- Comment #9 from C. M. Sperberg-McQueen <cmsmcq@blackmesatech.com>  2009-02-04 17:37:22 ---
Comment #8 appears to net out to an argument in favor of
interpretation (a) over (b), (c), or (d), if we wish to explain what
the names in the datatypes namespace denote.  

I think (a) is compatible with saying that XSD validators are not
expected to recognize names in the datatypes namespace, or do anything
special with them.  The datatypes namespace, in this story, is for the
convenience and use of those who wish to name the built-in datatypes
without naming the built-in type definitions which describe them; XSD
validators are not in that situation.

Since (a) appears to fit a little better with the one use of the
datatypes namespace we are now aware of, (a) seems just now like the
least-cost highest-value resolution to this issue.

On a side note, comment #8 also suggests that it was as mistake to
say, in the spec, that different simple type definitions can describe
the same datatype, since "the fact that they are the same has no
observable consequences".  You may be right, but I respectfully
disagree both with the conclusion and with the premise.

Any spec defines concepts for use as tools in talking and thinking
(and not solely for use in conformance checking); among those defined
by XSD 1.1 Datatypes are "datatype" and "simple type definition".
Since the two are intimately interconnected but distinct, identifying
many-to-one relations between them can help make clear how they
differ, even if the fact of the many-to-one relation were not
observable or had no consequences for validator conformance.

But in fact, I think it is observable that in the following example,
the simple type definitions s1, s2, and s3 describe the same datatype.

 <xsd:simpleType name="s1">
  <xsd:restriction base="xsd:integer">
   <xsd:minExclusive value="0"/>
   <xsd:maxInclusive value="3"/>
  </xsd:restriction>
 </xsd:simpleType>
 <xsd:simpleType name="s2">
  <xsd:restriction base="xsd:short">
   <xsd:minInclusive value="1"/>
   <xsd:maxInclusive value="3"/>
  </xsd:restriction>
 </xsd:simpleType>
 <xsd:simpleType name="s3">
  <xsd:restriction base="xsd:decimal">
   <xsd:enumeration value="1"/>
   <xsd:enumeration value="2"/>
   <xsd:enumeration value="3"/>
  </xsd:restriction>
 </xsd:simpleType>

So it seems to me that there are observable consequences to the fact
in question.


-- 
Configure bugmail: http://www.w3.org/Bugs/Public/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
Received on Wednesday, 4 February 2009 17:37:31 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:50:09 UTC