W3C home > Mailing lists > Public > www-rdf-interest@w3.org > January 2000

RE: Followup 2: The Heart of RDF Darkness

From: Jeff Sussna <jeff.sussna@quokka.com>
Date: Tue, 4 Jan 2000 10:36:03 -0800
Message-ID: <E19A882C6CD5D211A8A70008C75B6AF40122CCCB@pcmail.quokka.com>
To: "'Eric Hellman'" <eric@openly.com>, Gabe Beged-Dov <begeddov@jfinity.com>
Cc: "'www-rdf-interest@w3.org'" <www-rdf-interest@w3.org>
Ugh! Yup, it looks like I will have lots of intermediate objects as well
(which isn't really a problem, except that one starts to wonder what one is
actually getting from the RDF layer itself).

Jeff

P.S. Gabe, looks like ID/resource mutual exclusivity wasn't an error. :-(

-----Original Message-----
From: Eric Hellman [mailto:eric@openly.com]
Sent: Friday, December 31, 1999 9:34 AM
To: Gabe Beged-Dov; Jeff Sussna
Cc: 'www-rdf-interest@w3.org'
Subject: Re: Followup 2: The Heart of RDF Darkness


I agree this is confusing, it was referred to in an earlier 
discussion as "striping" of syntax. Implicit in the RDF syntax is the 
notion that something is either a property or a value. The 
parseType="Resource" attribute leads to confusion because it implies 
the creation of an anonymous resource.
Having both rdf:ID and RDF:resource attributes is illegal because the 
parser can't decide whether it's a value or a property.

Another way to look at this is to asK if you have
>   <playsFor rdf:ID="44" rdf:resource="Pirates"/>
then what does the ID identify? the statement, or the resource? I 
agree it would be convenient to have a separate ID for the 
reification, but perhaps that's nasty.

Presented with a similar problem to Jeff Sussna's, we explicitly made 
an intermediate object.

so Mick has sponsorship x (x is a Sponsorship object)
x has startyear 1985
x has stopyear 1990
x has sponsorname pennzoil

So because RDF is property-oriented, we end up creating lots of objects.



At 3:40 PM -0800 12/30/99, Gabe Beged-Dov wrote:
>Jeff Sussna wrote:
>  >
>  > Given the notion of reification, it would be nice to be able to simply
say
>  > "property X defined within description Y has the timeframe 1987-1993".
As
>  > far as I can tell, however, this is impossible, since there is no way
to
>  > make a statement about a single member of a container (the 
>contained in this
>  > case being the Bag of property statements contained by the
description).
>
>Taken from the formal grammar section of the spec:
>
>    Within propertyElt (production [6.12]), the URI used
>in
>    a resource attribute identifies (after resolution)
>    the resource that is the object of the statement
>(i.e.,
>    the value of this property). The value of the ID
>    attribute, if specified, is the identifier for the
>    resource that represents the reification of the
>    statement.
>
>Note the second sentence. I don't think that this
>information is maintained by current parsers but it
>does allow you to identify the reification of a
>specific statement in a description.
>
>If it was supported, you could say:
>
><rdf:Description about="Jimmy">
>   <playsFor rdf:ID="44" rdf:resource="Pirates"/>
>   <playsFor rdf:ID="55" rdf:resource="Brigands"/>
></rdf:Description>
>
><rdf:Description about="#44">
>    <startDate>2/3/78</startDate>
>    <endDate>2/2/81</endDate>
></rdf:Description>
>
><rdf:Description about="#55">
>    <startDate>2/3/81</startDate>
>    <endDate>2/2/83</endDate>
></rdf:Description>
>
>Cordially from Corvallis,
>
>Gabe Beged-Dov

Eric Hellman
Openly Informatics, Inc.
http://www.openly.com/           21st Century Information Infrastructure
Received on Tuesday, 4 January 2000 13:35:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:42 GMT