axiom-developer
[Top][All Lists]
Advanced

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

Re: [Axiom-developer] Axiom, FriCAS, forks and teeth


From: C Y
Subject: Re: [Axiom-developer] Axiom, FriCAS, forks and teeth
Date: Wed, 11 Jul 2007 11:03:37 -0700 (PDT)

--- Gabriel Dos Reis <address@hidden> wrote:

> The notion that Axiom would need just to be written once and never
> rewritten again is -- I'm afraid -- completely deconnected from
> reality, and reflects an amateurish approach to software that
> saddens me.

This may indeed be amateurish, but after all I AM an amateur.  I am
aware that Axiom will need to be extended and built upon indefinitely,
if for no other reason than because mathematical research is
continuing.  What I want is for solved problems to remain solved, just
as TeX's solution to typesetting has kept it a solved problem for
decades.  What is the limit?  How robust and future proof can the
system be made?  I want to find out.
 
> (1)  As a law, useful software evolve; otherwise they die.

Certainly.  If actual new ideas appear, they should be incorporated.  I
would hope that these would be in areas like new mathematics, new
output formats, new graphics backends, etc - i.e. changes not involving
the core of the system.

Maybe this is a practical impossibility, but I am not yet convinced of
that and would like to try.

> (2)  Axiom is to support computational mathematics.  While 1+1=2
>      will ternally be true, computation technologies *evolve*,
>      *change* and many parts of Axiom will have to be changed,
>      evolved, rewritten.

My hope is that changes and new abilities can be added strictly to
address new ideas and new operating environments, and that the internal
design and logic will scale independently of any external technology
changes.  Maybe it can't work this way, but I am hopeful that new
effort could someday be focused almost exclusively on mathematical
features and not superstructure.

> When you think you have 30 years to write Axiom, think about what
> the environment in which it will be deployed will look like.  If you
> don't need to make it useful, you can disregard my comments.  If you
> don't need to sustain its development till 30 years from now, please
> disregard my comments.

Speaking ONLY for myself, short term usefulness is not my primary goal.
 Please note I am NOT saying that it is an unimportant goal, just that
it is not what drives my personal interest in this project.

I am tremendously impressed by what Knuth has achieved with TeX, which
is still used and has gone unsurpassed for decades.  New features have
been added as new technologies are created (pdf export, for example)
but the core of TeX has stood the test of time.  I want a CAS which
will do the same thing for computational mathematics - stand the test
of time.  It took something like 10 years to create TeX, IIRC.  The
results have paid off, at the expense of a long lead time.  The system
is still extended today, of course, but the foundation has remained
steady.

THAT is the kind of project I want to work on.  The TeX of computer
algebra systems.  Am I disconnected from reality?  Possibly.  But we
have a wide variety of commercial systems and even free alternatives
like Maxima that are addressing real problems today.  Axiom's codebase
itself can be forked to a project with different objectives.  MY
PERSONAL INTEREST is in the long term, TeX-like system that can last -
can it be done?  What would it take?

I don't ask anyone to accept my interests as the "correct" ones - they
are simply my interests and the reason I spend my time on this project.
 Other people have other interests, and that is great.  I intend to
pursue mine - the experiment of Axiom as a long term project.

Cheers,
CY


       
____________________________________________________________________________________
Get the Yahoo! toolbar and be alerted to new email wherever you're surfing.
http://new.toolbar.yahoo.com/toolbar/features/mail/index.php




reply via email to

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