Re: Comments on Spearman and Williams solvable quintics


Richard Fateman wrote:
> 3. Every expression or subexpression of the form x^(1/n) is
> ambiguous potentially. How do you propose to simplify them
> unambiguously.
>
> 4. Solving quintics better is not an important issue in
> the design or implementation of computer algebra systems;
> generally the solutions that come out of the formulas
> using radicals cannot be used for anything else because
> (except in very simplified circumstances) they are too
> complicated in form for the simplifiers to handle, and
> so might as well be opaque "root #n out of m of this
> formula".
>
> Regards
> RJF
I agree strongly with these last to points. Allot of solutions to the
roots of polynomials I have seen on this news group have been pretty
unintelligible. However, I still feel the research is important as I
hold "faith" that CAS in the future will be better able to deal
with long complicated, intelligible and somewhat ambiguous expressions.

.

Relevant Pages

  • Re: Help with equation
    ... how do you simplify that by taking the log of both sides? ... to me like you have a sum there on the right side, ... minus 30th root of ... take the positive root, because then you get a number that makes the ...
    (comp.sys.hp48)
  • Re: Help with equation
    ... The missing was na error on my part Typo give it a try if you ... how do you simplify that by taking the log of both sides? ... > since the 30th root is even, it gives both positive and negatives, so ... > take the positive root, because then you get a number that makes the ...
    (comp.sys.hp48)
  • Newtons Method---Iteration
    ... We wish to find the root of the equation: ... It said to write down the Newtons method and simplify (I got this ... Would that be the Secant method or the accelerated Newton's ...
    (sci.math)
  • Re: OOo problems with rpm
    ... > 1) Download the rpm ... > 2) open a command line ... > 3) su to root ... Let me simplify that: ...
    (alt.os.linux.suse)