Re: [Axiom-developer] Re: literate programming pamphlet files for MathAction

[Martin Rubey]

Ralf HEMMECKE writes:

 > As far as I know there are 3 languages, all slightly different.
 > A Aldor
 > S SPAD
 > I Axiom interpreter language
 > 
 > I could live with A \ne I, because the interpreter is there to make life 
 > easier for the lazy user. For serious programming, however, Axiom should 
 > use Aldor, not SPAD. I believe that SPAD and Aldor still differ in some 
 > places, unfortunately.

That is exactly what I meant. Tim Daly said that using Aldor for compiling
Axiom files (i.e., .as files that contain Aldor code but rely on the Axiom
library) should work, but either I'm missing the obvious or it is not obvious
how to do it...

I think this should be first priority for those who can resurrect this.

For lisp and especially compiler gurus -- Camm, I believe that's you... :-) --
a great (non-priority) project might be to make a (simple! non-optimizing) lisp
implementation of the Aldor compiler that could be freely distributed and run
within axiom.

One good reason for that is the following: a bug can have different
origins. Sometimes it is obvious, that it is an algebra bug, or a compiler bug,
or a lisp bug, but sometimes this is not clear at all.

Having different, but semantically -- modulo bugs -- equivalent implementations
would ease debugging quite a bit, I imagine.

On the other hand, given our limited resources, this seems quite out of
reach...

Martin