W3C home > Mailing lists > Public > public-swbp-wg@w3.org > March 2006

RE: [WNET] URIs as primitive queries?

From: Booth, David (HP Software - Boston) <dbooth@hp.com>
Date: Wed, 29 Mar 2006 17:04:50 -0500
Message-ID: <EBBD956B8A9002479B0C9CE9FE14A6C20B924E@tayexc19.americas.cpqcorp.net>
To: "Mark van Assem" <mark@cs.vu.nl>
Cc: "SWBPD list" <public-swbp-wg@w3.org>


> From: Mark van Assem [mailto:mark@cs.vu.nl]

> . . . There's one problem left: this approach would 
> give a URI clash for all the sub-types of wordsenses:
> 
>       http://wordnet.princeton.edu/wn/wordsense/noun/
> 
> can refer to all NounWordSenses, or to the WordSenses with a 
> Word with 
> the lexical label "noun". Some kind of prefix could be introduced to 
> solve this e.g. "type-noun".

I don't quite follow this.  Do you need to provide URIs for identifying
such things?  It sounds like you are trying to ascribe meaning to that
URI as an RDF node.  I was assuming that URI would not be defined as an
RDF node, because it is missing the sense number.  I.e., 

       http://wordnet.princeton.edu/wn/wordsense/noun/noun/1/

would be defined as a particular word sense of the word "noun", but

       http://wordnet.princeton.edu/wn/wordsense/noun/

would not be defined to mean anything (as an RDF node).  If you need a
URI for the set of all NounWordSenses, or for the set of all WordSenses
pertaining to the word "noun", then completely different URIs could be
minted for those purposes.  Are they needed?

> . . .
> > Also, the language may need to be indicated, so
> > perhaps something like the following would work better:
> > 
> >     http://wordnet.princeton.edu/wn/wordsense/en/bank/
> 
> Why is that necessary? We already know that WN is in "en-US". URI
> clashes with other WN's can be easily avoided by having 
> another base URI.

My mistake.  I saw the language being specified in the example SPARQL
query, and assumed that WN was going to span languages.

> 
> > 2. Starting from a particular URI that is an RDF node in a triple,
> > look up related information.  In this case, I don't think the 
> > application would (or should) know to deconstruct the URI 
> > in order to 
> > do a broader query, so I don't think the above mechanism would be 
> > appropriate for this usage.  (But please correct me if you 
> > think I'm wrong.)
> 
> Well, I think an application should not rely on this. But in practice
> it would probably be programmed to do so if it gets the job done. I 
> think this is slippery terrain, but I can't decide either way.

Well, if you're unsure, I'll be a little more forceful. :)  Don't do it!
It is not a practice that should be encouraged or endorsed.

> . . .
> OK, so the new proposal would be:
> 
> 	http://wordnet.princeton.edu/wn20/synset/bank/type-noun/1/
> 	http://wordnet.princeton.edu/wn20/wordsense/type-noun/bank/1/
> 	http://wordnet.princeton.edu/wn20/word/bank/
> 	http://wordnet.princeton.edu/wn20/schema/participleOf/

Much better!  But how about further simplifying the URIs to the
following:

> 	http://wordnet.princeton.edu/wn20/synset/bank/noun/1/
> 	http://wordnet.princeton.edu/wn20/word/bank/noun/1/
> 	http://wordnet.princeton.edu/wn20/word/bank/
> 	http://wordnet.princeton.edu/wn20/schema/participleOf/

This would provide even more consistency in the URIs.  Note that by
appending "noun/1/" to the end of the word URI, you get the URI for that
particular word sense -- the keyword "wordsense" is not needed.  

> 
> [1]http://www.w3.org/2001/sw/BestPractices/WNET/wn-conversion.html
> [2]http://lists.w3.org/Archives/Public/public-swbp-wg/2006Mar/0076

David Booth
Received on Wednesday, 29 March 2006 22:07:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:17:21 GMT