Re: ACTION-268

On 30 Jan 2009, at 16:51, Bijan Parsia wrote:

>
> On 21 Jan 2009, at 19:31, Bijan Parsia wrote:
>
>> Done with:
>> 	 http://www.w3.org/2007/OWL/wiki/index.php? 
>> title=Syntax&oldid=17348 [Bijan Parsia]
>>
>> To:
>> 	http://www.w3.org/2007/OWL/wiki/Syntax#Keys
>
> In working on the draft reply, I was wondering if my change was  
> enough, or if it should read:
>
> """A key axiom of the form HasKey( owl:Thing OPE ) is similar to  
> the axiom InverseFunctionalProperty( OPE ), the main differences  
> being that the former axiom is applicable only to individuals that  
> are explicitly named in an ontology, while the latter axiom is also  
> applicable to individuals whose existence is implied by existential  
> quantification. Furthermore, an inverse-functional property is  
> inverse-functional for all assertions using that property, whereas  
> keys can be scoped to assertions involving individuals of a certain  
> class.""""
>
> Add: """In this way, classes can represent distinct tables with the  
> same key name or different versions of the same table directly."""

OK for me -- and it may help people (like Jim) coming from DB  
backgrounds to understand the rationale.

>
> If this is ok, I'll add it.
>
> Also, I think this is misleading:
>
> """This makes key axioms equivalent to a variant of DL-safe rules  
> [DL-Safe]. Thus, key axioms will typically not affect class-based  
> inferences such as the computation of the subsumption hierarchy,  
> but they will play a role in answering queries about individuals.  
> This choice has been made in order to keep the language decidable."""
>
> DL safe rules require DL safety to be decidable (to a first  
> approximation). Keys *do not*. We restrict keys for implementation  
> considerations.

I agree -- the last sentence is simply wrong. We could simply delete  
this sentence: I don't think that we want or need to justify all  
design choices in this document, and it might be difficult to provide  
a short yet convincing justification for this one.

Ian


>
> Cheers,
> Bijan.
>
>