RE: Potential new issue: PSVI considered harmful

-----Original Message----- 
From: Tim Bray [mailto:tbray@xxxxxxxxxxxxxx] 
Sent: Thu 6/13/2002 11:06 AM 
To: www-tag@xxxxxx 
Cc: 
Subject: Re: Potential new issue: PSVI considered harmful

Simon St.Laurent wrote:

>>> 4. Work on XQuery and other things that require a Type-Augmented
>>> Infoset must not depend on schema processing, and should not have
>>> normative linkages to any schema language specifications.
>>
>> Efforts to impose the PSVI as part of
>> the XML core are not very welcome in a lot of places.
>
> The problem
>that worries me here is where typing information comes from, and the
>supposition that types exist only as a side-effect of validation
>processing. 

Currently most XML typing mechanisms involve some degree of validation 
(defining constraints on an XML document and confirming that the predefined 
constraints are observed by the document). What ever, TAI you come up with will 
have to involve something akin to validation to be generally useful or else it 
devolves into using xsi:type with with the built-in primitives. Of course, 
you'd have to come up with a mechanism for annotating attributes in the 
instance document which would probably lead some sort of gross hack. 
 
As for XQuery: 
Considering that the XQuery data model already provides mappings to it from the 
PSVI[0] and similar mappings for DTDs are already in the works. I don't think 
that it is unreasonable to expect XQuery to support other constraint systems if 
they can be mapped to the XQuery data model. However, again I suggest bringing 
this up with the XML Query WG instead of trying to mandate this via the TAG.
 
[0] http://www.w3.org/TR/2002/WD-query-datamodel-20020430/#IDASOVR