[Aldor-l] Using Aldor or SPAD

[Martin Rubey]

Dear Gabriel, Ralf, Christian, Bill, ...

Gabriel Dos Reis <gdr at integrable-solutions.net> writes:

> | Gaby, have you any idea how long it could take to produce another
> | reasonably efficient Aldor compiler?
> 
> It all depends on what you want to have in the "reasonably efficient"
> compiler. It should be a mater of days or weeks to have a good front-end. It
> would take momths to have decent optimizations in.  but, that is definitely
> in our reach.

What implementation language would you propose? Is the legal situation affected
by the choice of the implementation language? Gaby, did your students see the
source code of the aldor compiler?

I think we should not wait any longer. I'm currently interested in two
projects:

* species, together with Ralf and Nicolas

* symmetric functions

Both of these projects depend *heavily* on types being first class objects, and
on allowing dependent types. Both of these features are missing in SPAD.

In fact, there are two things missing in Axiom:

1) a replacement for SPAD that allows one to use the Aldor language

2) a replacement for the interpreter, that allows one to use the Aldor language

I wouldn't be surprised if 2) would be easier to achieve, but I don't
know. Gaby, how much money do you need for 1), for 2) and for both?

Martin