W3C home > Mailing lists > Public > xml-uri@w3.org > May 2000

RE: A little courtesy, please

From: Bill dehOra <Wdehora@cromwellmedia.co.uk>
Date: Wed, 24 May 2000 16:13:39 +0100
Message-ID: <43C2F98D8414D411865A00508BC22AB9064165@odin.cromwellmedia.co.uk>
To: "'John Cowan'" <cowan@locke.ccil.org>
Cc: xml-uri@w3.org

:Of course I can't.  But accessing is not the only operation that
:exists on the Web, particularly the semantic Web.  I can write
:perfectly correct metadata, thus:
:
:<rdf:Description about=
:	
:"brick://us/ny/nyc/13%20%E.%203rd%20St.?side=east&course=11&seq=25">
:		<color>red</color>
:</rdf:Description>

That's true, there is much to be stated for making statements about
resources as opposed to making resources of them. As for color being a
property (datum) of a brick, I'm aginnit, 'cos that might imply that
different bricks that are red have the same color (or even that they are
red). Maybe RDF2 will have built in tropes and metamers :-). Point taken
nonetheless.

Actually while we're on it, any thoughts about the synchronisation of the
description and the the thing itself? The brick moves, or is painted, or is
replaced (or I nicked it ;-). Then the metadata or description is out of the
sync with the facts that are the case with the brick.  I've heard it touted
as particular problem with RDF, but maybe it's a feature of indirection
inside computers that we have to live and thread carefully with, as opposed
to a flaw in resource/thing systems themselves.


:No, I think in the above case that I am providing a datum 
:about *the brick*,not about some representation of it.

I'll concede that: differentiating between the sense data of a thing would
be argument for arguments sake.

-Bill
Received on Wednesday, 24 May 2000 11:12:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:32:42 UTC