Re: rdf-text telecon agenda

From: Peter F.Patel-Schneider <pfps@research.bell-labs.com>
Subject: Re: rdf-text telecon agenda
Date: Wed, 27 May 2009 07:20:42 -0500

> From: Sandro Hawke <sandro@w3.org>
> Subject: rdf-text telecon agenda
> Date: Tue, 26 May 2009 17:35:44 -0500

[...]

>> 9.  How to meet the interoperability goal...?
>> 
>>   .. brainstorming, sharing ideas, etc ...
>>      
>>      * Pat's approach using RDF'

Not completely the right way to go, in my view.

>>     Status of Table 3?

Remove it, and the previous paragraph.  All the information therein is
irrelevant to the definition of a new datatype.

>>     What do we say specifically about SPARQL?  
>> 
>>       - it shouldn't be be in the queried graph (but this this
>>         isn't about SPARQL)
>>       - it shouldn't be in the BGP
>>       - it shouldn't be in a filter
>>                STR("foo@"^^rdf:PlainLiteral), LANG( ), DATATYPE( )
>>       - it shouldn't be in CONSTRUCT

As little as is possible.

My first preference towards meeting the interoperability goal would be
to say *nothing* about restricting rdf:text datatyped literals in RDF.
There are already many ways to have datatyped literals in RDF (and its
semantic extensions, such as RDF+owl:sameAs) whose value space has a
non-trivial intersection with the "value space" of plain RDF literals.
Given this, what use is it to prevent one more way?

My second preference would be to just change the OWL 2 mapping to RDF
graphs document to map rdf:text datatyped literal into plain RDF
literals.

My third and fourth preferences would be to say that applications (and
recommendations) that incorporate rdf:text may/should be nice to older
applications (and recommendatations) and therefore may/should not emit
rdf:text datatyped literals in RDF syntaxes by changing them to plain
literals.

My fifth preference would be to say that in *syntaxes* for RDF graphs,
e.g., RDF/XML and Turtle, (and related syntaxes, such as any syntaxes
for SPARQL basic graph patterns, I guess) the syntax for rdf:text
datatyped literals *is* the syntax for plain RDF literals.

My last preference would be to make statements where complete compliance
would require all RDF applications to change.  This is what the current
document says.

Peter F. Patel-Schneider
Alcatel-Lucent

Received on Wednesday, 27 May 2009 13:21:14 UTC