# Re: Propositions

sowa <sowa@turing.pacss.binghamton.edu>
Date: Wed, 16 Feb 94 06:04:57 EST
From: sowa <sowa@turing.pacss.binghamton.edu>
Message-id: <9402161104.AA15483@turing.pacss.binghamton.edu>
To: cg@cs.umn.edu, interlingua@ISI.EDU, sowa@turing.pacss.binghamton.edu
Subject: Re: Propositions

In an earlier note, I wrote
>>PS: Actually, that procedure I defined in my previous note
>>does not even depend on the predicate names and alphabetical order.
>>The primary purpose of sorting is to eliminate duplicates and
>>to ensure a unique normal form. .....
And Pat Hayes responded
> Why not define conjunctions to be finite sets of conjuncts? Then issues of
> ordering and duplication just don't arise. Finite sets can be considered a
> data type in computational discussions.
Yes, that is a fine way to state the definition. But I just wanted
to point out that this nonconstructive definition can be implemented
by an algorithm that takes (N log N) time. Furthermore, the definition
can be computed in different languages with different names and
character sets and still lead to equivalent results. The need for
a language-independent definition of "proposition" was the original
motivation for this discussion.
John