[Top][All Lists]

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

Re: a meta name for axiom, open.axiom, fricas. was: Re: [Axiom-developer

From: Bill Page
Subject: Re: a meta name for axiom, open.axiom, fricas. was: Re: [Axiom-developer] Re: [#55] Axiom doesn't run on Windows
Date: Fri, 21 Dec 2007 01:04:22 -0500

On 12/20/07, M. Edward (Ed) Borasky wrote:
> Bill Page wrote:
> > On 12/20/07, M. Edward (Ed) Borasky wrote:
> >> ... There is no confusion on my machines simply because I don't
> >> need either FriCAS or OpenAxiom -- and I haven't installed them --
> >> and I don't intend to.
> >>
> >
> > So are you arguing against Tim's point of view? If there is no
> > confusion then why all this concern over a name?
> >
> I'm arguing in favor of a single Axiom, called Axiom. If FriCAS
> and OpenAxiom want to rejoin the main branch, that's one way.

I believe that is what nearly everyone wants. However I am not sure if
this is what Tim wants since he specifically invited both Waldek and
Gaby to create forks when they could not come to some agreement about
basic development methodology and the future of the project. And the
current argument over names seems to further distance Axiom from these

> Another way is to ignore the other two and just use Axiom.

I think evolution does not favor this strategy. ;-)

> There are lots of things I don't put on my machines because they're
> not ready for prime time, and FriCAS and OpenAxiom fall into that
> class for a number of reasons including the potential directory name
> clashes. And I run "testing-level" (borderline unstable) boxes.

There are no directory name clashes. I run all three side-by-side at

Installation was simple. The only thing you need to be concerned
about is the proper setting of the AXIOM variable.

> I simply see no benefit to a recent fork of Axiom -- not enough has
> changed since the fork(s) to the parts of the package I really care
> about for me to want to spend time testing them.

Your time it your time, but I think you are quite wrong about how much
has changed in the forks. For example with FriCAS you can use hyperdoc
without having it crash unexpectedly. You can make deep fundamental
changes to the Algebra without having SPAD abort because of an
inconsistency in the database files. And you can build a complete
Windows version using clisp in cygwin. There are also a large number
of differences that are not so visible to users but that make life
easier for developers, e.g. the autoconf-based build system.

> Now if I find something broken in Axiom and it gets fixed, being
> open source, FriCAS and OpenAxiom are perfectly capable of
> picking it up.

Of course that works both ways and there are several examples of that
over the last few months.

> Or if someone puts something *useful* in one of the two, I'd take
>  a look. I just don't think that's going to happen because of why
> the forks happened -- they came about for reasons other than
> the usefulness of the core language for applied mathematics.

I disagree. But first my apology in advance to Tim for dwelling on
features of projects other than Axiom on this list. My intention is
only to counter the claim Ed makes above.

Unlike the original Axiom project which appears to focus on
documentation and the "30 year horizon", both FriCAS and OpenAxiom
specifically state "usefulness of the core language for applied
mathematics" as one of their immediate goals. E.g.

"OpenAxiom strives to support ubiquitous, advanced, high quality open
source computer algebra on major operating systems, in particular
major Unix variants, GNU/Linux variants, Windows, and handheld
devices. It aims at being the open source computer algebra system of
choice for research, teaching, engineering, etc."

And FriCAS:

"Longer term plans.

The compiler for Spad language included in Axiom is buggy and has
serious performance problems -- new compiler is needed. New
mathematical algorithms. Assertion support."

> In other words, "come back when you've got better math than
> Axiom". :)

Fair enough. One example of that already is Martin Rubey's "GUESS"
package although that may not be of much specific interest to you.

But in general my sincere wish is for a more collaborative,
co-operative, less competitive and more unified environment. As you
said: "a single Axiom".

Bill Page.

reply via email to

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