Re: ISSUE 131 (OWL R Unification): Different semantics on syntactic fragment

On Aug 12, 2008, at 1:08 AM, Michael Schneider wrote:

> Hi Ian!
>
> Ian Horrocks wrote:
>
>> Michael,
>>
>> It has never been claimed that OWL R DL and OWL R Full are completely
>> equivalent on the syntactic fragment.
[snip]
> In the original issue, it was stated that
>
>   "The main benefit would be that we would not need  
> owl:intendedProfile"
>
> For me, this is at best a very minor nice-to-have benefit.

That's not the main benefit for *me*. The main benefits are:
	1) Profiles all (mostly) work the same (i.e., the core bit is a  
syntactic fragmetn)
	2) Extensibility on the semantics makes sense (i.e., tools aren't  
*forced* to be semantically restricted outside ethe core fragment)
	3) We don't have axioms which have semantically meaningful  
constructs only on one side.

For me 1 and 3 are *killers*.

> And now, we are near to close this "intended profile" issue,  
> anyway, by not having such a signaling URI at all. So, the  
> "unification" issue can even be regarded to be kind of moot.

No.

> But we are still talking about the unification, for which the price  
> to pay would be pretty high for the RDF side, which originally was  
> the only side that asked for such a rule-based language.

First, this isn't true. DLP was included in the original fragments  
document, IIRC. Please don't rewrite history esp. to make a debater's  
point.

Second, we have to consider the ecosystem, not just a single species.  
We also have to consider the range of users and the range of possible  
confusions.

We're adding a *lot* of real constructs to an RDF fragment...we  
should expect that people will be less indifferent to the entailments  
they get.
[snip]
>> Your example is a good illustration of why it would be *a very bad
>> idea* to define a 3rd semantics for OWL based on the OWL RL rules.
>> According to this semantics, it would NOT be the case that
>> owl:intersectionOf (C D) is a subClassOf D. Any reasoner finding this
>> entailment would be unsound and non-conformant w.r.t. this semantics.
>> This would, IMHO, be highly counter-intuitive.
>
> Now this was actually my counter example, so one can easily take it  
> as an example for an "unintuitively" missing entailment. But  
> equally well, from a rules perspective, one could also claim that  
> producing this entailment is counter intuitive.

What? Are you seriously coming from a user perspective *at all*? They  
aren't going to look at the rules. They are going to *write something  
down*.

> Actually, there are enough examples for derivations by applying the  
> rules, where there is no respective entailment by OWL R DL, simply  
> because it would fall outside OWL R DL's syntax (at least, an OWL R  
> DL reasoner wouldn't be required to produce it there).
>
> Consider that very asymmetric syntax of OWL R DL (e.g., unions, and  
> existential and universal restrictions may only occur on one side  
> of subclass axioms, respectively). Isn't this alone already  
> "counter-intuitive", at least to people who do not understand the  
> theoretic background behind this language? In comparison to this,  
> the OWL R Full ruleset looks pretty coherent to me.

Only by deception. You allow the syntactic freedom without the  
semantics. That's not coherent, that's *bad*.

At least by highlighting the asymmetries, you clearly indicate where  
what you say *has no effect*.

> Well, so we have claims about "counter-intuitive" reasoning results  
> on both sides. I would call that a draw! :)

I wouldn't. We don't have mere claims.

[snip]
> The point is that it would really not be a great marketing  
> statement to say: "We have a sound OWL RL reasoner!".

I don't see how you can simultaneously claim that users won't care  
about silent no-semantics axioms and yet care about calling it sound.  
Calling the rules complete for all graphs is *just* marketing and  
*bad* marketing because it relies on some fairly unpretty sophistry.

> Building sound-only reasoners is a trivial task (just take a "zero- 
> reasoner", which does not produce any inferences).

Oh man, you're clearly not in marketing land now.

> And a reasoner, which just implements the "official OWL R ruleset",  
> wouldn't be more than just an OWL R sound-only reasoner. While  
> there would be other reasoners around, which would really be sound  
> and complete w.r.t. the OWL R semantics. Would anyone buy the  
> former reasoner under these circumstances?

Sure, for scalability, for brand name, because they don't care about  
completeness, or for any of a  number of reasons.

(BTW, in bioinformatics circles, soundness is *much* less interesting  
than completeness (assuming it's not overnoised). Talking with folks  
like Robert Stevens,  I find that they would be *much* more perturbed  
by constructs that *didn't* have effects though they look like they  
should. I *imagine* HCLS is a field that would be highly interested  
in OWL-R (as a compilation target, and directly)).

I think it's a *very* bad idea to get too creative in profile land,  
in the standard. I'm not saying it's exactly analogous, but OWL Lite  
should be a cautionary tale.

Cheers,
Bijan.

Received on Tuesday, 12 August 2008 08:00:59 UTC