Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: SUO: Re: Starter KB V2 Question #8




Mike wrote:
> 
> Chris Menzel says:
> 
> Note this is, again, not ambiguity -- for a term to be ambiguous is for
> it to have more than one meaning.  To be indefinite is to have a single
> meaning which can be extended in more than one way. 
> ---
> 
> This is a subtle, technical distinction. If I look at a term and its
> axioms and I can think of two ways to extend it and I'm not sure which
> one the axiom-writer meant, or indeed whether he meant what he said
> (i.e. it entials both cnpts) then from an informal point of view -- me
> looking at and trying to understand this term, I think of it as being
> ambiguous.  I am not sure what the author meant.
> 
> I'm not suggesting your comments are incorrect, -- only that there is
> a common sense way of thinking about this which makes them seem
> incorrect. 

Well, then the only thing to do here is to clear up that
misapprehension, Mike!  ;-)  Within an ontology, the meaning of a term T
is its axiomatization (to speak aphoristically).  Even if that
axiomatization is incomplete, so that T is indefinite, T is still wholly
unambiguous -- it has a single meaning.  However, the indefiniteness of
T relative to its ontology might well point to an ambiguity in the
ordinary language term that T is intended to represent.  In that case,
we either try to discern an single intended meaning and extend the
axiomatization of T appropriately, or we canonize the ambiguity and
introduce two new terms that share T's axiomatization but extend that
axiomatization in different ways.  The only ambiguity in the
neighborhood is in ordinary language -- no surprise!  It is eliminated
(or better, managed) in the ontology.

Regards,

-chris