W3C home > Mailing lists > Public > www-tag@w3.org > January 2003

Re: Clarifying what a URL identifies (Four Uses of a URL)

From: Tim Bray <tbray@textuality.com>
Date: Tue, 21 Jan 2003 13:50:01 -0800
Message-ID: <3E2DC089.8070400@textuality.com>
To: Dare Obasanjo <dareo@microsoft.com>
Cc: Sandro Hawke <sandro@w3.org>, Michael Mealling <michael@neonym.net>, David Booth <dbooth@w3.org>, www-tag@w3.org, "Roy T. Fielding" <fielding@apache.org>, Dan Connolly <connolly@w3.org>

Dare Obasanjo wrote:

>>On the other hand, using the same URI to mean different 
>>things is a Bad Thing and leads to confusion and misbehavior 
>>not only at the Semantic Web level but in terms of general 
>>human utility.  
> 
> No matter how forcefully you state it you cannot get around the fact
> that people will use both http://www.25hoursaday.com to refer to me as a
> person or to whatever representation is returned by Apache when an HTTP
> GET is performed. It seems that you are implying that a Semantic Web
> based on URIs is broken [as designed?]. 

I agree that despite the fact that amgibuity in the meaning of URIs is 
bad, it happens all the time, just like 404s are bad but happen all the 
time.   I'm a bit nervous that the SemWeb machinery isn't being designed 
with enough care to survive the inevitable occurrences of ambiguity in 
the way that HTTP is designed to survive occurrences of 404.  If the 
Semantic Web is based on an assumption that a URI will always be used 
consistently, then it is indeed broken as designed.

>>It's a formalism.  The Web Architecture has a formalism 
>>called a "Resource" which is the one thing that corresponds 
>>to each URI. 
> 
> This statement is meaningless and yet W3C TAG members keep repeating it.
> What is the one resource that the URI "http://www.w3.org/Consortium/"
> identifies? 

The architecture doesn't have a way to talk about what a resource "is". 
  All it does is allow you to interchange representations, robustly and 
scalably.  This doesn't seem to cause a problem in practice outside of 
the domain of KR applications.  -Tim
Received on Tuesday, 21 January 2003 16:50:35 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:15 GMT