9304d400   
   From: walter@bytecraft.com   
      
   Bill Silvert wrote:   
      
   > On Apr 11, 1:39 pm, Walter Banks wrote (in part):   
   > > Bill,   
   > >   
   > > Linguistic variables are the stuff we as humans understand best. Many   
   > > of the fuzzy logic problems are the manipulation of linguistic variables.   
   > > You could code your bicycle wheel adjustment as a series of fuzzy rules.   
   > >   
   > > old thermometers have scales calibrated in linguistic references like   
   > > "freezing", "cold" , "hot" and "very hot" which leads one to wonder if   
   > > the natural form of measurement may be linguistic and not crisp.   
   >   
   > Walter, I think we all know this. My point was that one uses fuzzy   
   > logic to address fuzzy issues, not crisp definitions. Using fuzzy   
   > logic to draw a mathematically exact circle makes no sense. Using   
   > fuzzy logic to draw a reasonable circle does. We can use fuzzy logic   
   > (fuzzy control theory) to make a bicycle wheel round, and we can use   
   > fuzzy logic to heat our food to the right temperature for serving.   
      
   Bill,   
      
      My comments were not aimed at you specifically and I really   
      missed making my point. Your Bicycle example is a classic non   
      linear problem that can be solved simply with a few fuzzy rules.   
      It is a good example where the fuzzy implementation is almost   
      trivial.   
      
      I think we are on the same page. I have focussed on linguistic   
      variables in talks on  fuzzy logic more than the math behind   
      their manipulation.   
      
   w..   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)