Maxima for numerical methods (v. Scilab)?

I only use commercial software if there's no alternative.  For that reason I
have been using Maple and Matlab for some years.  Even though I have written
a Matlab book (
http://www.course.com/catalog/product.cfm?category=Computer%20Science&sub=
category=Digital%20Image%20Processing&isbn=0-534-40011-6),
I don't expect students to buy it: they just use printed notes on which the
book was based.  Many of my students come from disadvantaged backgrounds,
and I never expect them to spend yet more money on books or software.

As for a book; sure, why not?  A very good book, available online, is
"Numerical Computing with MATLAB" by Cleve Moler (founder of The Mathworks)
available at http://www.mathworks.com/moler/index.html.  I'd be happy if I
could produce something similar-ish to that, but based on free software.
Anyone interested in being a co-author?

Thanks for your replies so far.

cheers,
Alasdair