RE: [Axiom-mail] A(ld,xi)o(r,m)

[Page, Bill]

On Monday, September 27, 2004 4:17 PM Frederic Lehobey
address@hidden wrote:

>> ...
>> The problem to me seems to be that parts of aldor are distributed
>> ONLY as binary files.  There is no source only distribution I am
>> aware of, and if I understand correctly the intent is that there
>> not be one. That's why I've personally never viewed Aldor as a
>> logical program to use with Axiom.
>
>  Seconded.

You do realize of course, that not so long ago Aldor (then called
AxiomXL or the "library compiler") *was* a part of the commercial
version of Axiom? Aldor was "released" by NAG for non-commerical
development a few years before Axiom itself to a different group of
developers and under different conditions at a time when open source
was not so well developed as a concept.

Axiom can and does operate without Aldor but in principle Aldor
would provide better optimized library code and would compile
significantly faster than the compiler that is built-in to Axiom.
Since building Axiom and compiling all of the library code using
Axiom's compiler, on even a fast computer can take several hours,
I think using Aldor with Axiom could provide significant advantages -
especially during development and testing of new code.

There are a few people (and I am not one of them) who would have
to meet and agree on a common strategy, licensing, etc. before
Axiom and Aldor could become one again or at the very least
cooperate for mutual benefit. I think the best thing we could do
here is to demonstrate that there is an active interest in making
this happen.

BTW, I plan to have Aldor running in interactive move available
via MathAction in the next few days - though not linked to Axiom
(yet).

Regards,
Bill Page.