Re: regrets & skolem predicates

On Apr 21, 2011, at 6:41 AM, William Waites wrote:

> * [2011-04-20 16:10:23 -0500] Pat Hayes <phayes@ihmc.us> écrit:
> 
> ] 
> ] On Apr 20, 2011, at 9:20 AM, William Waites wrote:
> ] 
> ] > Unfortunately I won't be able to make the call today due to a
> ] > conflicting commitment.
> ] > 
> ] > However, a thought just occurred to me - what happens when someone
> ] > uses a skolem constant - now a URI - as a predicate?
> ] 
> ] Nothing happens. What did you think might happen?
> 
> Oh well, the world might end?

:-)

> 
> There is nothing really new here, people have always been free to mint
> predicates as they like. But so far I think we have seen a fair amount
> of conservatism here. A special URI space for skolem constants might
> encourage people to make many ad-hoc predicates - there are some
> problems for which this may make sense (some kinds of treatments of
> similarity and relevance for example, teasing apart "incorrect" uses
> of owl:sameAs come to mind).

Maybe. Seems like a stretch to me, though. 
> 
> This doesn't pose a theoretical problem but may pose a practical one
> since at least some stores make an assumption that the number of
> distinct predicates will be relatively small as an optimisation.
> 
> That, and the implications of a store trying to turn these skolems
> back into bnodes - if :s _:b1234 :o is illegal but :s <blank:b1234> :o
> is not, what does that do to equisatisfiability?

If you were just round-tripping, you would never have got a skolem name in that position, so this RDF must have been generated externally using the skolem name. And sure, if you do that then you might not be able to get back to where you were originally. Similarly if you were to use it as a dataype name inside a typed literal, for example. 

Pat

> 
> Cheers,
> -w
> -- 
> William Waites                <mailto:ww@styx.org>
> http://river.styx.org/ww/        <sip:ww@styx.org>
> F4B3 39BF E775 CF42 0BAB  3DF0 BE40 A6DF B06F FD45
> 
> 

------------------------------------------------------------
IHMC                                     (850)434 8903 or (650)494 3973   
40 South Alcaniz St.           (850)202 4416   office
Pensacola                            (850)202 4440   fax
FL 32502                              (850)291 0667   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes

Received on Thursday, 21 April 2011 23:35:17 UTC