Re: Computer Algebra Algorithms lisp vs. C.

Modern FPLs are vastly better suited to CAS development than C++. The
resulting programs will be smaller, simpler, faster, more robust and
efficient than a C++ program written in the same amount of time.

Can you show us some code that support your claim? 


I sold the code to WRI and developed it under NDA, of course. However, I can
give you some information about it.
...


I think we do not agree perhaps because we do not speak of
the same thing when we speak of CAS. For me the parser is not
the essential part of CAS development, of course it has to
be done, but once it is done, we can forget about it.
The essential part is for me the coding of mathematical algorithms,
e.g. gcd algorithms, or the Risch algorithm for symbolic integration
or Gaussian quadratures for numeric integrations, etc.
I strongly believe that the language of implementation is not
important for this, but the data structure and the mathematical
knowledge that is put inside the algorithms will be essential.
If you want to convince me from the contrary, I suggest that
for example you show me a polynomial gcd code that is
significantly faster than e.g. fermat or giac.
.

Relevant Pages

  • Re: When in doubt, drown em in lawyers
    ... Gregory Weston wrote: ... asserted that algorithms are inherently patentable. ... Mathematical algorithms are inherently unpatentable. ... talking about whether an abstract process is automatically disqualified ...
    (comp.sys.mac.system)
  • Re: When in doubt, drown em in lawyers
    ... Gregory Weston wrote: ... asserted that algorithms are inherently patentable. ... Mathematical algorithms are inherently unpatentable. ... There's no such thing as a free lunch, but certain accounting practices can ...
    (comp.sys.mac.system)
Loading