I haven't read (or re-read) the redoing proposal, but
here are a few facts.
1. you can't overload existing CL programs like + and *.
2. even if you could, do you want (+ 'x 'x) to build the
tree (+ x x) or do the simplification (* 2 x)? And
if you say simplify, how much? As you know, there are
lots of possibilities, like expand, factor, ratsimp ...
3. I think that using 1/2 instead of ((rat simp) 1 2) does
not require any object oriented stuff. In fact, you can
insert rational numbers into a maxima system by defining
(defun $makeclrat(num denom) (/ num denom))
and see what happens. I think things like 1/2+makeclrat(1,2)
don't simplify quite.
C Y wrote:
>--- Stavros Macrakis <stavros.macrakis@verizon.net> wrote:
>
>
>
>>Is there a set of patches to the rest of Maxima, or is this a project
>>that will have to be undertaken if we want to use newsimp? If the
>>latter, I would suggest we do it once, and do it right, so that we
>>can use not just CL rationals, but also CL complexes, IEEE floats,
>>user-defined extension types, etc., as I suggested in my email of
>>2003-Dec-12 11:41: "clean up and abstract other object types".
>>
>>
>
>I'd put my vote in for "do it once, do it right." Please forgive me
>Stavros if you covered this (I lost a few emails) but I have a couple
>"duh" questions - if we are going to do this, are we going to use CLOS
>objects and introduce object oriented programming to Maxima? Would
>that be the "right" way or overkill? I've always wondered if
>restructuring some stuff to use CLOS would make sense for
>maintainability concerns, but I don't know how much of a pain it would
>be to impliment/use or whether or not there is come collision between
>CLOS concepts and Maxima's ideas.
>
>CY
>
>__________________________________
>Do you Yahoo!?
>New Yahoo! Photos - easier uploading and sharing.
>http://photos.yahoo.com/
>
>