dy(x)/dy



On Sun, Nov 30, 2008 at 1:39 PM, <N.A.Beishuizen at ctw.utwente.nl> wrote:

> What I meant was why d: diff(y(x),y) is not evaluated as 1
>

Because in Maxima, the function y and the variable y are considered as two
distinct things.

>> Why this behaviour and how can I make it work the way I want
> >> without explicitly declaring dependencies?
>

Why do you not want to declare dependencies?


> >  depends(y,x) ?
>
> that contradicts the statement:
> >> without explicitly declaring dependencies?
>
> Actually, y does not depend on x, but dy/dx does exist in my problem
> (determining nth prolongation in symmetry analysis).


To tell Maxima that dy/dx is in general non-zero, you declare the
dependency.


> The alternative is using 'diff(y,x), but then 'diff(y,y) will also stay
> unevaluated, unless I build in checks to determine what the second argument
> is. Maybe I should use a defrule and treat dy/dx as something special.
>

I think you are making things too complicated for yourself by fighting
against Maxima's usual conventions.  I doubt you need a defrule.  Maxima in
general works much better if you work with variables and dependencies rather
than with unknown functions with arguments.

               -s