Re: rational numbers only

[Jordi Gutierrez Hermoso]

On 1/5/06, John W. Eaton <address@hidden> wrote:
> You could also write a new user-defined data type for rational
> numbers.

I suppose this could be an interesting exercise, but who would use the
finished product? Not me, personally. It is already implemented in
other software packages and seems of little use for the kind of person
who uses Octave, the kind usually interested in numerical analysis.
The number theorists have Pari, group theorists use GAP, and for
anything else that requires symbolic manipulations we have
general-purpose CASes.

Since Octave is *nix  software, I usually prefer the philosophy of one
tool doing one job and doing it well (but then again, M-x
praise-be-unto-emacs). Swiss army knives are tools of The Devil, RMS,
and Bill Gates. :)

> it would be a lot of work to get something like svd of a
> rational matrix working (if that would even make sense).

This could make sense too, if you want to extend the previous
exercise, in order to include a radical datatype. You could have some
fun simplifying radicals, a task more difficult to program than it
might seem at first sight, because then we start dipping our toes into
(baby?) Galois theory.

- Jordi



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------