W3C home > Mailing lists > Public > www-rdf-interest@w3.org > February 2004

RE: "Locally-Significant" Statements

From: Leo Sauermann <leo@gnowsis.com>
Date: Wed, 18 Feb 2004 11:38:46 +0100
To: "'Chris Wilper'" <cwilper@cs.cornell.edu>, "'Rhoads, Stephen'" <SRhoads@thrupoint.net>, <www-rdf-interest@w3.org>
Message-ID: <000501c3f60b$63462380$0501a8c0@ZION>

+1 to the wrapper.

The information about the cereal
http://www.generalmills.com/cereal/Cheerios is mainly from generalmills.
Package size, ingridients, promotion material, etc should be there.

The retailer publishes an "offer to buy" the product. The items are a
kind of "price list item", and not cheerios.

The advantage of the "wrapper" is also that I have a GOOD URL by the
retailer. I could then retrieve RDF triples about the offer by resolving
the URL and downloading something, so this may be good practice (if you
are member of the "club of URL lovers" - see uri crisis).

The instanceOf approach seems to have a bug, 
it should be "rdf:instanceOf", isn't it ????
and that is outdated, it was replaced to rdf:type or rdf:subClassOf
(check this google search : http://tinyurl.com/3ylsn )
or did I miss your intention ?
Anyhow, subclasses are surely good when you make some kind of "custom
promotion package" version,
f.e. 
<http://www.disney.com/promotions/MickeyMouseCheerios> <rdfs:subClassOf>
<http://www.generalmills.com/cereal/Cheerios>, 
<rdfs:comment> "Mickey Mouse promotion cherrios with little plastic
figurines".


The context approach ("We just make statements about the original URI of
the product and use context to disambiguate -- source RDF files today;
quadruples, quintiples, name graphs or whatever tomorrow") 
has the problem that after merging sources, the queries may get real
long to get data out.


( I am getting hungry right now, all this food discussions, have to go
get something to eat...)

regards
Leo

 
 -----Original Message-----
From: www-rdf-interest-request@w3.org
[mailto:www-rdf-interest-request@w3.org] On Behalf Of Chris Wilper
Sent: Wednesday, February 18, 2004 2:35 AM
To: Rhoads, Stephen; www-rdf-interest@w3.org
Subject: RE: "Locally-Significant" Statements


I like your "wrapper" approach best.  The idea that a "distribution of a
product" has a price (rather than the more abstract "product") seems
most logical.  The "instance of" approach seems awkward, and the
"context" approach makes it hard to answer questions across the
different graphs....but I'm curious to hear what other people think.

- Chris

-----Original Message-----
From:   Rhoads, Stephen [mailto:SRhoads@thrupoint.net]
Sent:   Tue 2/17/2004 7:12 PM
To:     www-rdf-interest@w3.org
Cc:    
Subject:        "Locally-Significant" Statements
Folks,

I would appreciate some guidance as to the best approach to the problem
of
what I will call (for lack of a better term) making locally significant
statements about a resource.  I have identified three approaches which I
am
calling "Wrapper", "InstanceOf", and "Context".

Take the example below.  General Mills manufactures an ex:Cereal called
(..snipped...) 
Received on Wednesday, 18 February 2004 05:37:40 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:58 UTC