On 4/13/07, Detlef Schmicker <d.schmicker at physik.de> wrote:
> But other questions are not necessary too. Maybe maxima should try all
> options, and check if the result really differs.
Yes, that is an interesting idea. I tried to implement something
like that a while ago, but I ran into problems. Maybe I should
try it again.
> I would help fixing, but I would not love to be the only one working on
> the integration. I read in the archive, somebody already had the wish to
> have multidimensional integration.
Well, we (especially Raymond Toy, thanks, Ray, very much) are
working on fixing bugs in the integration code. So if you would like
to work on the integration code, I think everyone would appreciate that.
Even if it turns out that you are not interested in Lisp programming,
it would be valuable to the project for you to review the existing
integration functionality and make recommendations for future
development. I believe the opinion of well-informed outsiders
is very valuable to the project.
> (sorry, the output is not always readable in here,
> don't know a better way)
display2d : false; will make the output more readable for email.
If Maxima can deduce the sign of the constants from previous
information, it won't ask the user. So you can enter
assume (a > 0, c > 0, kp > 0);
(or whatever as appropriate) before the integral and Maxima
will ask fewer questions.
It seems likely, however, that the result does not actually depend
on the sign of these constants, so Maxima need not determine
the sign to begin with. I don't know what to do about that.
Hope this helps,
Robert Dodier