Re: Entailment Doc ready for review

On 19 September 2010 16:57, Andy Seaborne <andy.seaborne@epimorphics.com> wrote:
> One comment:
>
> What is an "Entailment Regime"?
>
> Someone asked me recently and I couldn't find a succinct definition
> anywhere, in the WG doc or even with Google.
>
>        Andy

I cc the group, assuming that this was not intentionally off-list.

Thinking about it, I was wondering myself where the term comes from
since it was already around when I joined the group. It actually comes
from the Query spec, where we have (Sec. 16.6):

<quote>
An entailment regime specifies

   1. a subset of RDF graphs called well-formed for the regime
   2. an entailment relation between subsets of well-formed graphs and
well-formed graphs.

Examples of entailment regimes include simple entailment [RDF-MT], RDF
entailment [RDF-MT], RDFS entailment [RDF-MT], D-entailment [RDF-MT]
and OWL Direct and RDF-Based Semantics entailment [Ref: OWL2
semantics]. Of these, only OWL Direct Semantics (OWL-DL) entailment
restricts the set of well-formed graphs. If E is an entailment regime
then we will refer to E-entailment, E-consistency, etc, following this
naming convention.
</quote>

>From that it seems to be pretty much what an entailment relation is,
just that it is RDF specific and the entailment relation much be
between RDF graphs. The references to RDF entailment, RDFS entailment,
... actually imply that entailment regimes is used as synonym for
entailment relation and it might actually be a misnomer.

The Query doc goes on to say "A SPARQL extension to E-entailment must
satisfy the following conditions. " and then gives the four
conditions. Now the Entailment Regimes document basically takes the
common entailment relations (RDF, RDFS, ...) and defines a SPARQL
extension for each of the entailment relations such that conditions
are satisfied and such that the answers to a SPARQL query take the
entailed solutions into account.

First of all, I think Query spec uses entailment regime where
entailment relation is meant, but to be consistent, I stick with that.
The introduction of the Entailment Regimes doc then starts with:
The SPARQL 1.1 Query specification [SPARQL 1.1 Query]  defines only
simple entailment, which allows for finding answers by matching the
triple pattern of the query onto the RDF graph of the queried data.
Other entailment regimes, such as RDFS entailment, allow for finding
answers to a query that are not directly specified in the queried
graph, but can be inferred using a set of inference rules. In this
document, we specify how SPARQL can be used with some other entailment
regimes beyond simple entailment.

Is that not the kind of thing you were looking for Andy?

It would have been nice if normal SPARQL was actually defined in terms
of simple entailment because then one would see that we really just
use a different entailment relation. If you take the RDF regime, for
example, and swap RDF entailment for simple entailment and don't se
the special RDF vocabulary, you would actually get the answers as with
subgraph matching.

Should we make this more explicit?

We could also say that within the document an entailment regimes is
actually understood as taking an entailment relation and define
condiions on its use so that the SPARQL conditions on extensions are
met.

Birte


-- 
Dr. Birte Glimm, Room 309
Computing Laboratory
Parks Road
Oxford
OX1 3QD
United Kingdom
+44 (0)1865 283520

Received on Monday, 20 September 2010 19:30:20 UTC