W3C home > Mailing lists > Public > www-tag@w3.org > October 2002

RE: Potential TAG issue in re consistency, Schema, etc.

From: Dare Obasanjo <dareo@microsoft.com>
Date: Sun, 13 Oct 2002 22:23:23 -0700
Message-ID: <B885BEDCB3664E4AB1C72F1D85CB29F8040EF647@RED-MSG-10.redmond.corp.microsoft.com>
To: "Tim Bray" <tbray@textuality.com>, "Tim Berners-Lee" <timbl@w3.org>
Cc: "WWW-Tag" <www-tag@w3.org>, "Ann Navarro" <ann@webgeek.com>

Jeni Tennison as usual makes informed and valid comments about a W3C specification based on a thorough understanding of the issues. Unfortunately I can't say the same for your points. 
 
1.) XQuery has conformance levels built into the spec (DOM Levels 1, 2 and 3 anyone?) and the first level of conformance is not overly onerous nor does it require in depth knowledge of W3C XML Schema. 
 
2.) This isn't a concrete or detailed enough comment to be able to provide an informed response. 
 
3.) W3C XML Schema is here. A strongly typed XML query language needs a basis for its type system. W3C XML Schema fits the bill and it adds consistency to the W3C XML architecture if W3C XML Schema is used in XQuery. 
 
If you have concrete opposition or a concrete suggestion for the W3C XML Query working group it would be best for all parties involvled if you just stated your case instead of trying to wield the TAG like a stick to get whatever it is you want from the W3C XML Schema working group. 

	-----Original Message----- 
	From: Tim Bray [mailto:tbray@textuality.com] 
	Sent: Sun 10/13/2002 8:37 PM 
	To: Tim Berners-Lee 
	Cc: WWW-Tag; Ann Navarro 
	Subject: Re: Potential TAG issue in re consistency, Schema, etc.
	
	


	Tim Berners-Lee wrote:
	> I think Tim Bray is arguing here against XQ engines actually needing to be
	> xml schema engines -- as opposed to the XQ spec being written using xml
	> schema.
	> Unless I have misunderstood.
	
	I'm saying three different things about XQuery, two of which may have
	architectural import:
	
	1. XQuery is far too large and complex for a V1.0 spec, with a severe
	cost to interoperability.
	2. XQuery extends way too far into the domain of abstract-type-based
	querying and run-time schema wrangling.
	3. If XQuery is going to require schema machinery, it should not
	hard-wire itself to W3C XML Schemas to the exclusion of all
	alternatives. [With the exception of using Schema's basic atomic data
	types, which lots of specs including competing schema languages now do.]
	
	In #1 and #3, you could s/XQuery/current W3C specifications/ and be left
	with a valid statement of concern, which suggests that they are issues
	of W3C-global import and potential TAG (or AB, or AC) fodder.  #2 seems
	like the kind of issue that should get worked out in the normal W3C
	process. -Tim
	
	
	
Received on Monday, 14 October 2002 01:23:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:12 GMT