From: fateman@cs.berkeley.edu   
      
   On 3/2/2014 4:28 AM, G. A. Edgar wrote:   
   >> I have always thought that sqrt(a/b) and sqrt(a)/sqrt(b) are   
   >> exactly the same   
   >>   
   >   
   > Certainly sqrt(a/b) and sqrt(a)/sqrt(b) are both square roots of a/b.   
   > But a complex number has two square roots.   
      
   So does a complex number with a zero imaginary part.  That is, a real   
   number.   
      
     It could happen (as Axel   
   > explains), that the "principal branch" choice for sqrt results in   
   > opposite choices for these two.  If a,b are both positive, this does   
   > not happen, and you get the same square root.   
   This assumes, contrary to generally accepted college mathematics   
   "complex analysis" courses, that the sign of the argument of the square   
   root (etc) determines the branch cut. Which it does not.   
   >   
   > Similar things can happen with other powers, with logarithgms, inverse   
   > trig functions, and so on.  Unless you choose the arguments nicely   
   > enough.   
      
   No, choosing the argument does not choose the branch cut.  Unless maybe   
   you are in high school and believe that log(|x|) is a legitimate   
   solution to ignoring the complex plane.   
      
   It is possible to choose branch cuts. Students learn to do it. Conformal   
   mapping programs must do it.   
      
   RJF   
   >   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)