From: //noreply@axelvogt.de   
      
   On 10.04.2016 23:32, clicliclic@freenet.de wrote:   
   >   
   > Axel Vogt schrieb:   
   >>   
   >> I confess that I did not try to implement the linked algorithm:   
   >> what does it give for the unit disk and the unit roots (I would   
   >> not mind if some small rotation may give difficulties) ?   
   >>   
   >   
   > The papers are dealing with the case of the *open* unit disc, so roots   
   > of unity will be counted as being outside. But there is no loss of   
   > generality since replacing x by 1/x in the polynomial maps the closed   
   > outside of the disc onto the closed disc.   
   >   
   >> BTW: one always can reduce the task to the unit disk.   
   >   
   > Yes. And one can do it symbolically, the disc center and radius then   
   > becoming parameters in the polynomial.   
   >   
   > Martin.   
   >   
      
   Hm ...  x -> 1/x maps the unit circle to itself ...   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)