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

Re: Comments on Architectural Principles of the World Wide Web

From: Joseph Reagle <reagle@w3.org>
Date: Mon, 9 Sep 2002 12:44:36 -0400
To: "Tim Berners-Lee" <timbl@w3.org>, "Ian Jacobs" <ij@w3.org>
Cc: <www-tag@w3.org>
Message-Id: <200209091244.36974.reagle@w3.org>

On Friday 06 September 2002 02:31 pm, Tim Berners-Lee wrote:
> > http://www.w3.org/TR/2002/WD-webarch-20020830/
> This is an example of the confusion between the identifier and a
> reference. This confusion is really exacerbated by this awful term
> "absolute URI reference" for
> the identifier!
>
> A QName is a perfectly good reference to something which is
> identified by a URI.  The QName isn't itself a URI, just as a relative
> reference a not a URI.

Ah, this makes a lot of sense to me when you put it that way. But it's not 
exactly what the TAG finding says... Let's consider the following bit of 
XML and the QName:
o. <xsd:restriction base="xsd:string">
o. The QName xsd:string is a reference to "something" (a type)
o. The TAG finding says this is a "concise, unambiguous name for the 
URI/local-name pair" {" http://www.w3.org/2001/XMLSchema", "string"}
o You are saying it is a reference to an identifier, and the TAG finding 
does speak to this too, "Other specifications use QNames as shortcuts for 
unique identifiers derived from a URI/local-name pair that have no 
relationship to element or attribute types." However, in this case could 
you please indicate the actual identifier? Or has Schema WG not yet 
provided it? Should they? If WGs use QNames as identifiers, do they still 
have an obligation to create the identifier under some QName -> Identifier 
translation?
Received on Monday, 9 September 2002 12:44:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:11 GMT