[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Axiom-developer] Aldor?

From: Bill Page
Subject: RE: [Axiom-developer] Aldor?
Date: Fri, 13 Jan 2006 13:22:46 -0500

On January 13, 2006 12:06 PM Jay Belanger wrote:
> ... 
> So that any potential signers could get a quick grasp of the
> situation, a little history might be useful.  What about
> something like:

Thanks for the draft. I think it is very good. I have inserted
a few explanations below. I will leave it up to you exactly
how to modify the text.

>    Axiom is a sophisticated computer algebra system originally
>    developed by the Numerical Algorithms Group (NAG).

Axiom was actually developed as an IBM research project. It
was sold to NAG with the intention that NAG would market it
commercially, which they did for a few years. NAG also made
some improvements and developed and marketed a PC version. (The
PC versions depended on software called texplorer which
was separately licensed and not available to at the time
Axiom became open source.)

>    After NAG discontinued work on Axiom, they generously
>    released

After NAG discontinued Axiom as a commercial product ...

>    the source code under a free license.

... part of the Axiom source code (excluding the Aldor compiler) ...

>    Axiom is now very usable and is currently being actively
>    developed by an enthusiastic community. It is expected to be
>    useful in research and education for quite some time.
>    The language used for writing Axiom programs is called SPAD
>    (short for Scratchpad, the original name of Axiom).  While Axiom
>    was still being developed by NAG, an improved extension language
>    called Aldor was created.

Aldor was written while Axiom was still an IBM research project.
SPAD and Aldor are only used to write Axiom library code - not
Axiom per se. Axiom and the SPAD compiler are written in Lisp.
Aldor is written in C.

>    While the Aldor compiler was originally part of the Axiom system,
>    it soon became a separate project and was not part of Axiom when
>    the Axiom sources were relicensed.

As I understand it, Aldor did not exist as a separate project until
after NAG discontinued the sale of Axiom as a commercial product.

>    Binaries for Aldor have been made freely available and the sources
>    are available under a generous license.  However, the license does
>    not qualify as a free or open source license.  While Axiom can
>    still be built using Aldor, the lack of a free license for Aldor
>    prohibits it from being a standard part of Axiom, and so nothing
>    in the base Axiom system can assume that Aldor is included.
>    A free license for Aldor would allow it to become a standard part
>    of Axiom, which would improve Axiom and make Aldor a more
>    widespread language.  A free license available as soon as possible
>    would allow current development on Axiom to use Aldor.
>    We, the undersigned, respectfully request that the Aldor compiler
>    be released under a free license at the earliest opportunity.
>    Sincerely,

I think the rest of the proposed letter is fine.

> ... 
> > The Axiom website is a wiki (user editable), so it would be very
> > easy to add a page consisting of the petition to Free Aldor. We
> > could freeze the text and then readers could "sign" it by simply
> > adding a comment.
> That might not be a bad idea.  We would have to get Steven Watt to
> look at it, of course.

Ok, unless someone has serious objections, I think we can post a
revision of a letter such as you wrote above on the Axiom wiki.

> > From the letter from Steven Watt that you quoted, it sounds like
> > he's willing to free Aldor when he gets the chance.  If he's like
> > me, he just needs a reminder now and then.  Whatever we can do to
> > provide a nudge would be useful. 

Yes, I think that is exactly the case.

Bill Page.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]