I am accusing RJF of been dammned stupid. Everybody knows   
   that polynomials are not an Euclidean Domain, so Algorithm E   
   cannot work for polynomials.   
      
   In the present incarnation, it shows for the given pair,   
   that it loops. Same for the other pair f,g, I already showed.   
   And yes Algorithm E is not an algorithm for multivariate   
      
   polynomials. What remains to be shown is that Algorithm C   
   isn't either an algorithm for multivariate polynomials.   
   But this is way over your head RJF.   
      
   Am Samstag, 8. Juli 2017 19:12:02 UTC+2 schrieb Richard Fateman:   
   > On 7/8/2017 10:05 AM, bursejan@gmail.com wrote:   
   > > No, I didn't say Knuth is wrong, where do you read   
   > > this into my post RJF? Can you point me to such a claim   
   > > by me that Knuth would be wrong?   
   > >   
   >   
   > You said   
   >   
   > "   
   >   
   > Algorithm E from Knuth loops   
   >   
   > ?- X is 2+9*A^2-5*B, Y is 5*A^2*B+5*A*B^2, euclid_gcd(X,Y,R).   
   > [2,3]   
   > [3,2]   
   > [2,5]   
   > [5,2]   
   > [2,5]   
   > [5,2]   
   > etc...   
   > "   
   >   
   > If a procedure loops indefinitely, it is not an Algorithm.   
   >   
   > So you accused Knuth of defining, as Algorithm E, something   
   > that is not an algorithm.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)