From: fateman@cs.berkeley.edu   
      
   On 7/7/2017 8:34 AM, bursejan@gmail.com wrote:   
   > I still dont see some observational data of yours   
   > concering your R. Fateman conjecture, could you be   
   > more specific. So far there was only some handwaving,   
   >   
   > a lot of refrences to these and that GCD algorithms.   
   > We have also learnt, that for univariate polynomials   
   > Euclid GCD = GB,   
      
   I don't know what you mean by "=".   The result might   
   be the same, but the algorithms use different steps.   
      
   You seem to not understand algorithm analysis.   
      
   The result of two sorting programs may be the same,   
   but one might take time O(n*log(n))  and another   
   might take O(n^3).  Are the two sorting programs "=" ?   
      
   You seem to think that GB analysis should rely on   
   "your gut".  Certainly your chosen implementation   
   affects the timing, but I have not seen any reason   
   to believe that what you have written is better   
   than the best alternative implementations.   
      
      
   The history of polynomial GCD algorithms from about   
   1965 to the present represents substantial improvements   
   in the asymptotic complexity of the problem, as well as   
   actual practical programs.  (An additional valuable   
   program is a heuristic GCD, used in several systems).   
   There is certainly no need to rely on "your gut" here.   
      
   In any case, to say that these are "="  to GB is   
   a theoretical result which is basically irrelevant to   
   anyone with an interest in computing polynomial GCDs   
   in a CAS.   
      
   You refer to handwaving.  The observations you report   
   on your timing  + the observations I report  (which can   
   be reproduced by you or anyone who downloads the software   
   and runs the tests)  are not handwaving. They are   
   observations.  Your timings show that your programs,   
   as tested, are much slower.   
      
     No conjecture.   
      
   To be clear   
      
   In mathematics, a conjecture is a proposition for which no proof has   
   been found (yet).   
      
      
   Here is a conjecture.   
      
     You will  post to   
   sci.math.symbolic again in the next 10 days.   
      
   You can prove the conjecture by sending another message within 10 days.   
   You can disprove it, by waiting.   
      
   You see, the proof depends on incomplete information,   
   information or knowledge not available at this time.   
      
   So it is a conjecture.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)