W3C home > Mailing lists > Public > public-powderwg@w3.org > July 2008

Comment on Formal Semantics, status of GRDDL transform

From: Ivan Herman <ivan@w3.org>
Date: Mon, 14 Jul 2008 16:19:54 +0200
Message-ID: <487B608A.8040707@w3.org>
To: public-powderwg@w3.org
Comment on http://www.w3.org/TR/2008/WD-powder-formal-20080709/, Formal 

A small remark after the second example in section 3, namely:

The owl:intersectionOf of a singleton collection in both descriptor 
sets, although redundant, is a result of the GRDDL transformation.

triggered a question: what is _exactly_ the status of the GRDDL 
transformation? Is the GRDDL transformation normative? Or is the 
definition in this document the normative one, and the Working Group 
provides a GRDDL transformation implementing it for the good of the 
community? In other words, can I implement my own (GRDDL or otherwise) 
transformation that provides a _semantically equivalent_ but 
syntactically different graph?

There are obvious examples (order of constructions within the generated 
document) and also a bit trickier one. The note above refers to:

<owl:Class rdf:ID="descriptorset_2">
   <owl:intersectionOf rdf:parseType="Collection">
       <owl:onProperty rdf:resource="http://example.org/vocab#color"/>
       <owl:hasValue rdf:resource="http://example.org/vocab#red"/>

what if I generate the following OWL construction instead?

<owl:Class rdf:ID="descriptorset_2_ivan">
       <owl:onProperty rdf:resource="http://example.org/vocab#color"/>
       <owl:hasValue rdf:resource="http://example.org/vocab#red"/>

the generated RDF Graphs will be different (the whole construction for 
Lists will not be there) but, I believe, descriptorset_2_ivan and 
descriptorset_2 will be equivalent in the sense that the instances of 
those classes will be identical. Will I be POWDER compliant if I do that?



Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf

Received on Monday, 14 July 2008 14:32:27 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:06:04 UTC