From: ted.dunning@gmail.com   
      
   The rule of experience is that AI systems do not live or die based on   
   their theoretical foundations, but rather on how many man-centuries are   
   invested in creating them.   
      
   There are some notable exceptions where limited learning techniques   
   such as data-mining are appropriate.  In those situations, using   
   symbolic AI of almost any sort chains you to an unmaintainable morass   
   of a system (except in the special case of a limited set of heuristics   
   for variable creation and a few business rules for output   
   interpretation).   
      
   Category theory is very unlikely to actually provide much insight into   
   building real systems precisely because it is so meta-theoretical.  The   
   basic problem is that cognition doesn't appear to be all that much like   
   physics.   
      
   [ comp.ai is moderated.  To submit, just post and be patient, or if ]   
   [ that fails mail your article to , and ]   
   [ ask your news administrator to fix the problems with your system. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)