W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > February 2002

Re: A collection of issue resolutions

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Wed, 13 Feb 2002 15:38:39 +0000
Message-Id: <>
To: Patrick Stickler <patrick.stickler@nokia.com>, RDF Core <w3c-rdfcore-wg@w3.org>
At 16:08 13/02/2002 +0200, Patrick Stickler wrote:

Based on this message, and your other followup, I think we may be talking 
at cross purposes here.  I'll try to clarify.

Consider the literal "Wuthering Heights":

The statements:

   "Wuthering Heights"  <rdf:type> <rdf:Literal> .
   "Wuthering Heights"  <ex:length> "17" . # ignoring datatyping

are reasonable and true since the resource "Wuthering Heights" denotes the 
literal "Wuthering Heights" which is both a literal and of length 17 chars 
(if I've counted correctly).

   "Wuthering Heights" <rdf:type> <ex:book> .

is a silly thing to say because "Wuthering Heights" is a 17 character 
string which is a bit too short to qualify as a book.  Under our present 
proposals, "Wuthering Heights" cannot denote the book Wuthering Heights.

This is what I think you may have meant when you said that Literals cannot 
denote 'interesting' resources.

Why allow literals to be subjects?  Well, some folks have said that 
literals meet all the criteria to be a resource, so its a bit strange to 
say that they can't be.

My working assumption is that literals remain tidy - loosely, they always 
denote themselves.  Would you agree that, with that limitation a future WG 
may choose to allow them to be subjects.

But also note, that we cannot tie the hands of a future WG.  Just as we are 
changing things in m&s, a future WG, for reasons of its own, may decide to 
allow literals to be untidy.  But that I think, is a matter best left to them.

With that clarification would you be happy with the proposed resolution?


>Not fine. There are very real reasons (now) why they should not.
>If literals become tidy, then literals cannot be subjects. Literals
>themselves do not denote resources. Literals, in conjunction with
>some context such as datatype or other qualification may participate
>in the denotation of a resource, but they themselves do not denote
>the resource (if they are tidy).
>If we want to allow the literal node to denote the
>resource, by hanging all those qualifications off the literal node
>so that the literal node becomes a literal-in-context, which denotes
>a resource, then the literal node also denotes the context/occurrence
>of that literal, and thus literals cannot be tidy.
>This was one of the key hot-issues in the recent tidy/untidy debates
>and I tried to point out the ramification that adopting tidy literals
>precluded literals as subjects (the P++ idiom).
>If this still isn't clear, let me know and I'll keep trying...
>Patrick Stickler              Phone: +358 50 483 9453
>Senior Research Scientist     Fax:   +358 7180 35409
>Nokia Research Center         Email: patrick.stickler@nokia.com
Received on Wednesday, 13 February 2002 10:39:56 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:53:55 UTC