[Axiom-developer] Re: [Aldor-l] Using Aldor or SPAD

[Ralf Hemmecke]

On 08/19/2006 12:23 AM, Gabriel Dos Reis wrote:
> Ralf Hemmecke <address@hidden> writes:
>
> [...]
>
> | > It should be a mater of days or weeks to have a good front-end.
> | 
> | Oh, really? 
>
> when you get a handful of motivated people, yes.

Maybe we should open up a Wiki and call for people on the FrontPage 
similar to the FreeAldor link.

> | Why then wait any longer? Maybe, because there is no 100%
> | clear definition of the language Aldor?
>
> That is an important factor I guess. I.e. what do we want?  What kind
> of problems do we want to see solved by the language?

I believe that some research is involved here. I would definitely like 
to see better support in the language to state the input-output 
conditions in formal terms that could be used by a theorem prover.
If that just appears in the comments it is lost for checking program 
correctness.

> | > It would take momths to have decent optimizations in.
> | > but, that is definitely in our reach.
> | 
> | Does somebody volunteer to start such a project, preferably under GPL?
>
> well, I'm already burried with Axiom and other projects; but I do
> definitely have an interest here.

Axiom is really big, may I ask on which part of Axiom do you want to 
concentrate?

In my eyes Aldor is/should become one of the most important pieces in 
Axiom. The mathematical knowledge that is hidden in all the SPAD 
domains, must be rewritten in Aldor and in a pamphlet style.

> | I don't want the Aldor **language** to split, because of different
> | compiler implementation, but staying in an uncertain future is just
> | bad for the Aldor language to reach the masses.

> it would not be good for anybody to have the community split; however,
> more than one implementations are beneficial for both the language and
> the community.

The availability of manpower is a big question mark for a second 
implementation of Aldor.

Ralf