Re: [TF-ENT] URIs for entailment regimes in service descriptions

(I explicitly cc Axel here to put on his RIF WG member hat...)

Hm. I must admit I did not really look into this, I simply took the
terms used in the RDF Semantics document; more exactly, took over the
URI-s RIF already uses. And you are right, this is not clear....

Bijan Parsia wrote:
> On 1 Nov 2009, at 10:57, Ivan Herman wrote:
> 
>> Birte,
>>
>> I was not at the call, sorry about that.
>>
>> What I try to propose to the SW Coordination Group is the following set
>> of URI-s
>>
>> http://www.w3.org/ns/entailment/Simple
>> http://www.w3.org/ns/entailment/RDF
>> http://www.w3.org/ns/entailment/RDFS
>> http://www.w3.org/ns/entailment/D
> 
> I'm sorry that I seem to be missing something, but what does D indicate
> exactly? The ambiguity that concerns me is that it could indicate that
> the system respects the semantics of "datatypes in general" or of a
> specific set of datatypes.
> 
> I presume it's the latter and the requisite datatype map is from:
>     http://www.w3.org/TR/rdf-mt/#dtype_interp
> ?
> 
> But that's not very well specified. It would be better to use the OWL
> 2/RIF specs, I think, or some subset thereof. (E.g., we shouldn't leave
> open whether float and integer are disjoint.)
> 

Sigh:-)

I would probably take the intersection of the OWL and RIF. Ie, leave out
owl:float and rif:local or rif:iri. Alternatively, we may restrict
ourselves to what SPARQL defines as operand data types (11.1 in the
current spec).

Looking at the RIF document

http://www.w3.org/TR/rif-rdf-owl/

RIF uses this URI to identify the common RIF-D model which... seems to
leave the concrete datatype map open.

> (I wonder whether using "D" is the best thing to do here. The term
> "D-entailment" is pretty obscure as far as I can tell. And, in this
> case, would it also entail RDF semantics? RDFS? Do we really need RDF?)
> 

The RIF document says:

[[[
The profiles are ordered as follows, where '<' reads "is lower than":

Simple < RDF < RDFS < D < OWL Full

OWL DL < OWL Full
]]]

where 'profiles' means (just to muddy the waters:-) the RIF profiles.

The ordering makes sense but it is more than what the RDF Semantics
seems to say for D. Axel, can you try to remember the reasoning behind this?

If we want some sort of a compatibility (that is how the whole
discussion on having separate and general URI-s for these started!) then
we might want to take that over. It makes sense.

Whether SPARQL really needs a separate RDF entailment here is a
different question... I am not sure (but the URI would still exist)

Cheers

I.




> Cheers,
> Bijan.

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf