Re: "entailment regime"?

On Jul 8, 2007, at 11:55 PM, Sandro Hawke wrote:

> [This is kind of far afield from mainstream RIF, but in my head it's
> deeply linked to the extensibility design, so I'm keeping it on-list.]

Ok. I trust the chairs will send a signal when, in their judgement,  
this is noise rather than signal. But I'll also cut it short.

[snip]

> Unfortunately, no W3C Recommendation gives us a standard for doing
> either Option 1 or Option 2, so we're stuck in the land of Option 3.

Yes. We had loads of talk about this in DAWG but never got anywhere,  
alas.

[snip]

> On the other hand, as a user, I'd hate to have to guess which  
> entailment
> regime is the right one. There's a terrible failure mode, where I'm
> trying to learn something -- maybe the room number for the room where
> I'm supposed to lecture tonight -- and when I ask the room-reservation
> sparql endpoint, I have to guess which entailment regime it should  
> use.

I don't find this remotely realistic. If I'm using a SPARQL based  
*end user* application, then I expect to have set things up  
correctly. I may do so simply by my choice of reasoner/query engine  
(which is typically not user configurable, for an end user application)!

Users of entailment regime switching are either sophisticated (e.g.,  
enough to be able to switch between well-formedness and validaty  
according to a variety of schemas in XML) or switching is built into  
the application in a user sensible way. Anyone who forced you to use  
sparql directly in order to find a room is running one of the  
SUCKIEST information services in the known universe -- they should  
learn HTML and give you a web page you can google :)

(Note that Dave and I have fairly extensive experience of users of  
RDF through OWL...and our experience seems to largely coincide. It's  
not dispositive, but it does shift the burden of proof, I think. I'd  
typically expect real case studies or user studies, not just made up  
-- clearly implausible -- use cases, to feel the shift back.)

[snip]
> The
> problem I see is that the graph with that triple RDF-entails the graph
> without that triple, so in some cases (maybe not this example) I think
> you'd have incoherent combined semantics.   The vocabulary-in-use
> approach doesn't seem to have that problem.

I didn't understand this at all.

Cheers,
Bijan "that was *fairly short* right?" Parsia.

Received on Monday, 9 July 2007 00:00:52 UTC