Re: [RIF] [UCR]: What is the RIF (revisited)

On Feb 12, 2006, at 10:55 AM, Dieter Fensel wrote:

> At 10:18 12.02.2006 +0100, Igor Mozetic wrote:
>
>> So, according to the above, is my understanding correct that Pellet
>> fully supports OWL-DL?
>> Or is it just a demo, that it *could* be implemented?
>>
>> Regards,
>> Igor
>
> Indeed more specifically, was it an OWL-DL or an OWL-Lite reasoner.
> That is, did it deal with nominals?
>
> (hopping this is not out of scope)

IIRC, we only claimed to be a complete OWL-Lite reasoner at CR. That 
was what we were specifically aiming for (namely to show that OWL Lite 
was implementable). I *believe* that at the time we had rather more 
support (N and probably O, but I'm not 100% sure anymore of the 
timeline; O may have come the following summer). All my attempts to 
update the record got lost :) (Part of this was an artifact of the 
confromance classes defined in the test suite.)

Nowadays, of course, Pellet is a complete OWL DL consistency checker 
(since Oct) and had support for nominals in various combinations long 
before that.

For more on the OWL CR period see:
	http://www.w3.org/2001/sw/WebOnt/impls
esp. the Exit Criteria:
	http://www.w3.org/2001/sw/WebOnt/impls#exit

As you can see, some of the sorting was a bit for appearances. (I.e., 
FaCT and Racer certainly were complete OWL Lite consistency checkers). 
And Pellet could have counted as a useful OWL Full reasoner, maybe, as 
we did some automated coercion of OWL Full ontologies into a DL form. 
(Indeed, we pass 80% of the full entailment tests, at that time, see 
<http://www.w3.org/2003/08/owl-systems/test-results-out>).

Cheers,
Bijan.

Received on Sunday, 12 February 2006 16:35:17 UTC