Re: CASs vs. octave numerics (was: Re: Symbolic expand function doesn't always do anything)

[Joe Koski]

on 9/14/04 1:37 PM, John W. Eaton at address@hidden wrote:

> On 14-Sep-2004, Jonathan Stickel <address@hidden> wrote:
> 
> | edA-qa mort-ora-y wrote:
> |
> | > I must admit though, I truly don't understand why there is a separation
> | > between tools such as Octave and Maxima.  Why would I choose to use one
> | > over the other?
> | 
> | I've been troubled by this as well.  For the most part, though, I use a
> | CAS to arrive at formulas which I then use for numerical calculations.
> | I'd love to see an example where one interfaces symbolic manipulation
> | with numerical methods, but from my experience they are often completely
> | separate.
> 
> In case it is not already obvious, writing something like Octave or
> Maxima is a big job.  Probably the people who work on Octave are
> primarily interested in numerics, and those who work on Maxima are
> primarily interested in symbolic manipulation.
> 
> My experience was that when I started working on Octave in 1992, I saw
> the job of doing a numerically oriented language is being hard
> enough.  I did not know enough about symbolic manipulation to even see
> how to make it possible for Octave to do that and numerics.  If I knew
> then what I know now, maybe I would have done things differently.
> 
> OTOH, the existence of the "sym" data type in octave-forge shows that
> it might not be too difficult for Octave to support symbolic
> manipulation in a relatively clean way.
> 
> Since Maxima is free software, I don't know that it makes sense to try
> to duplicate the capabilities of Maxima in Octave.  It would probably
> be better to make it possible to use Maxima within Octave, either by
> sending and receiving data to a separate Maxima process, or by
> modifying Maxima so that it can be embedded in Octave.
> 

As a maxima user, I monitored the maxima mail list for a while about 9
months ago when I was first getting maxima operational on my Mac, and indeed
they helped me to modify maxima for the common lisp compiler version that I
had. 

At that time there was a message thread where someone suggested accessing
octave from maxima, and the group decided that it would be difficult, and
maybe the R-language was a better candidate language. At least the same
common usage thoughts have occurred on the other side of the octave/maxima
fence.

Maxima will write results as Fortran coded expressions, which could probably
be converted directly into octave expressions. Maybe the easy first step fix
is for maxima to write octave/Matlab compatible expressions as well as
Fortran. A suggestion probably better made to the maxima mail list.

Joe

> jwe
> 
> 
> 
> -------------------------------------------------------------
> 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
> -------------------------------------------------------------
> 



-------------------------------------------------------------
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
-------------------------------------------------------------