W3C home > Mailing lists > Public > www-tag@w3.org > June 2005

RE: More on distinguishing information resources from other resou rces

From: Bullard, Claude L (Len) <len.bullard@intergraph.com>
Date: Wed, 29 Jun 2005 08:43:28 -0500
Message-ID: <15725CF6AFE2F34DB8A5B4770B7334EE072070C2@hq1.pcmail.ingr.com>
To: 'Jonathan Borden' <jonathan@openhealth.org>
Cc: 'www-tag' <www-tag@w3.org>

Again, only the abstract or ROA of the airplane using the VIN 
is named.
 
An instance might have parts from different craft with mixed 
VINs.  That's common in the automotive industry and probably 
the airline industry.  The VIN gets you the assembly identity.

After another few rounds of beating on this, it may become 
clear that is isn't possible to completely identify any 
real world (by that I mean dynamic, changing, multi-part) 
object with URIs with 100% reliability.  This system is 
approximate at its very best.  It names documents more 
reliably than real world objects, so the system is defined 
in terms of itself (the representations).  Identity is a 
maintained value; not static.

Humans have no problems with this.  Machines may if you expect 
them to deliver 100% unambiguous results 100% of the time. The 
web is not capable of that in Trad or Sem form.  Human language 
isn't capable of that.

Roy is right.  It doesn't matter.

len

From: www-tag-request@w3.org [mailto:www-tag-request@w3.org]On Behalf Of
Jonathan Borden

By such definition, although Mark Baker cannot currently be completely
defined in OWL, such things as a Car, an airplane and/or various colors
could be (e.g. the documentation for the Boeing 707 along with a particular
"VIN").

Jonathan
Received on Wednesday, 29 June 2005 13:43:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:36 GMT