W3C home > Mailing lists > Public > public-rdf-wg@w3.org > November 2012

Re: Factoring of entailment regimes (was: Re: Ill-typed vs. inconsistent?)

From: Richard Cyganiak <richard@cyganiak.de>
Date: Fri, 16 Nov 2012 15:35:35 +0000
Cc: Pat Hayes <phayes@ihmc.us>, Pierre-Antoine Champin <pierre-antoine.champin@liris.cnrs.fr>, Antoine Zimmermann <antoine.zimmermann@emse.fr>, RDF Working Group WG <public-rdf-wg@w3.org>
Message-Id: <37862F18-2FEE-4420-9F50-04BB7B69F9DF@cyganiak.de>
To: Ivan Herman <ivan@w3.org>
Ivan,

On 16 Nov 2012, at 14:59, Ivan Herman wrote:
> but, at least the current version of the Schema document defines terms both in the RDF and the RDFS namespaces (eg, rdf:Property, or the collections), ie, the differentiation may not be that clean...

I'm pretty confident that there is no problem, and that the differentiation is clean.

“RDF” is just the data model.

“RDF Schema” is *all* the vocabulary. Some of the vocabulary is in a namespace conventionally abbreviated as rdf:, and other parts are in rdfs:. There's no system behind that, it's by historical accident.

The proposed Simple Entailment and RDF-With-Literals Entailment cover all the data model. Blank nodes and literals, basically.

The proposed RDFS Entailment covers all the vocabulary, including both namespaces.

The rules for the different entailment regimes can be cleanly separated. Those for Simple and RDF-With-Literals go into Concepts. Those for RDFS go into Schema.

That matches *exactly* the normative content that's already in these documents. Concepts has the data model but doesn't mention vocabulary. Schema describes all the vocabulary, including both namespaces.

(The datatypes (rdf:XMLLiteral, rdf:HTML, rdf:langString) are a slight wrinkle to the clean division of content between Concepts and Schema, as they can be considered vocabulary, but are defined in Concepts. But they won't be mentioned in the rules, so that's not an issue for the discussion here.)

Best,
Richard




> 
> Anyway. I am not opposed to this, but I am not convinced either.
> 
> Ivan
> 
> 
> 
> On Nov 16, 2012, at 09:50 , Richard Cyganiak wrote:
> 
>> On 16 Nov 2012, at 14:32, Ivan Herman wrote:
>>> I thought we discussed having all the rules in a separate document (as a Note); what is wrong with that?
>> 
>> The Note informatively describes the semantics of RDF and RDFS. If the Note can be cleanly split in two, and the two parts included as informative material in the documents that define RDF and RDFS, isn't that better in every way?
>> 
>> SKOS, for example, defines the language and any equivalence/consistency constraints in the same document, and splitting them off into a separate document would seem like a bad idea. The same logic applies to RDFS, IMO.
>> 
>> FWIW, this would add perhaps one page of content to Concepts and two to Schema, or double that with examples. Would it really be worth having a new document for 3-6 pages of content?
>> 
>> Best,
>> Richard
> 
> 
> ----
> Ivan Herman, W3C Semantic Web Activity Lead
> Home: http://www.w3.org/People/Ivan/
> mobile: +31-641044153
> FOAF: http://www.ivan-herman.net/foaf.rdf
> 
> 
> 
> 
> 
> 
Received on Friday, 16 November 2012 15:36:05 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:53 GMT