Re: NAF v. SNAF - where is this being addressed?

Gerd Wagner wrote:

>     But isn't that the case with inferencing in general: it is scoped
>     either implicitly or explicity to some KB and never to the "whole
>     Web" (whatever that means)? For instance, if I use Racer with
>     Protege/OWL, the scope is implicitly the loaded ontology (including
>     potential imports). 

I think the point is that RDFS and OWL are designed to be monotonic so that 
whatever inferences you draw from your scoped use of Racer remain valid 
even when you take into account additional, as yet unseen, assertions.

NAF violates this, hence its use in the semantic web being ... a cause for 
debate.

Dave

Received on Thursday, 30 June 2005 13:19:13 UTC