From: fateman@cs.berkeley.edu   
      
   On 7/12/2017 3:35 PM, bursejan@gmail.com wrote:   
   > Look for yourself, I cant do all your homework,   
   > what am I here your nanny or what?   
      
   You can't even do your own, self-assigned, homework.   
   >   
     you even don't know a single paper   
   > that compares any GCD with GB for GCD.   
      
   Right.  So far as I know, no one has ever written a paper that compares any   
   plausible GCD algorithm with the (implausible) use of GB.   
      
   If a paper has been written, it might not be publishable for   
   the reason that it is an implausible, bad idea.  You could   
   write a paper, except that your results so far show that   
   a GB algorithm written in Prolog is something like 3000x   
   slower than the competition on trivial (degree 2, 2 variable)   
   inputs.   
      
   ....   
   In the paper by Zippel, which you found in M. Monagan's course readings,   
      
      
   you quote  "The algorithms EZ, mod, red all get easily killed!"   
      
   Sure;   
   Zippel chose particular bad cases to show that SPMOD (the default   
   in Maxima) wins big in those cases. He says "This example was carefully   
   designed   
   so that all the GCDS were "bad zero" problems..."   
      
      The actual probability of those cases showing   
   up in practice may be quite low.  However, Monagan seems to think   
   the paper is worth reading for his course.   
      
      You complain that you have to run it through OCR. You think that   
   even with perfect copy you would get Omnipage to recognize   
   subscripts and exponents and products?   
      
   The first 3 or 4 smaller tests are not much to type in.  polynomials   
   of degree 2 in 2 or 3 variables with single-digit integer coefficients...   
      
   Your credible dips to new lows.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)