On 01-13-98 William Elliot wrote to John Boone...
Hello William and thanks for writing,
WE> The notions of fuzzy set A and B, A or B can be used for
WE> multi valued propositional calculus. To wit: p,q the true
WE> values of P,Q, 0<=p,q<=1. Truth value of not P is 1-p, of
WE> P & Q is min(p,q), of P or Q is max(p,q). Note that P or Q
WE> equivalent not(notP & notQ) and P & Q equivalent not(notP
WE> or notQ). So does it fly? Where to? Can we take P
WE> implies Q, P -> Q, to be not(P & notQ) or equivalently notP
WE> or Q?
WE> First off lets look to the finite case. The truth values T =
WE> {0.1/n,2/n,...,(n-1)/n,1}. Well this is just too much for
WE> my mind so I will chose n = 3 and denote three values
WE> false, maybe, true, f,m,t for 0,1/2,1. Does it fly? Where
WE> do we want to fly with it?
>
WE> P Q notP P&Q PorQ P->Q Q->P P=Q
WE> f f t f f t t t
WE> f m f m t m m
WE> f t f t t f f
WE> m f m f m m t m
WE> m m m m m m m
WE> m t m t t m m
WE> t f f f t f t f
WE> t m m t m t m
WE> t t t t t t t
WE> Most of this seems to hang except for the last column P->Q
WE> & Q->P, P equivalent Q. There's a problem here that P and
WE> Q can have the same truth value, 'maybe', without P=Q being
WE> 'true'. The problem is that two statements aren't
WE> equivalent just because they always have the same truth
WE> values, they're 'maybe' equivalent. Guess I've
WE> demonstrated my earlier claim that no workable multi value
WE> propositional calculus can been devised.
I don't have time to respond (as I need to finish a book
I have been reading by Schering, for work which is more
pressing at the present time), but I am not ignoring this;
so, I printed it out to ponder when I have time.
I did have a chance of "look at" your other response
I printed, and yes it is interesting.
Take care,
John
___
* OFFLINE 1.54
--- Maximus 3.01
---------------
* Origin: Strawberry Fields (1:116/5)
|