Richard Fateman wrote:
> Raymond Toy wrote:...
>
> ...
>
>> Maybe the expectation is that for compiled functions, you get to debug
>> the result using the Lisp's debugger.
>
> ..
> The expectation of the compiler [to the extent one can say that
> computer programs have expectations :) ]
> is that you've debugged the program before you compiled it! This
> expectation
I don't debug my Lisp programs before compiling them. Why should I do
that with maxima? :-)
>
> is unreasonable when the uncompiled function is so slow that the buggy
> part is not reached for
> a long time. However, wrong-number-of-argument errors can sometimes
> be detected at compile
> time (type-checking etc.) So maybe that should be done by the Maxima
> translator etc.
I think a warning would be ok, but given the dynamic nature of Maxima
(and Lisp) where I can redefine functions at any time, an error would be
wrong.
Ray