RE: [Axiom-developer] SAGE, Axiom, and usage

[Bill Page]

On August 23, 2006 2:21 PM Gaby wrote:
> | > 
> | > Yes, and as a matter of fact, I'm deeply sceptical of your
> | > previous assertion.
> | 
> | Which assertion?
> 
>   #  Gaby would like to introduce his students to "symbolic
>   #  computation", *but really Axiom (and Aldor) are not very
>   #  good at this -- by design.*
> 
> (emphasis is mine).
>

Sorry. I did not mean to imply that Axiom and Aldor are not good
"by design". Certainly a lot of effort went into making Axiom
with Aldor the best possible tool for doing general mathematics by
computer. The point I was trying to make was that the basic design
of Axiom was such as to emphasise mathematical *structure* over
linguistically oriented formal symbolic computation. It is clear
from the early literature on Axiom that this was a deliberate design
choice. Other early systems being developed at the same time, such
as Reduce, took the opposite view.
 
> | 
> | On the contrary, I do not think I am "painting it into a corner".
> 
> what you said only reinforces the perception I have had since
> some time now, from discussions on this list.
>

Could you explain what you mean "the perception you have had"?
Do you mean that idea that I am (we are?) are "painting Axiom
into a corner" by emphasising how it differs from some other
systems? I don't understand why you would think that.
 
>...
> between 1995 (when I first heard of it, and later got presentation
> by Stephen about A# at FRISCO workshops, and repeated "conversion
> attempts" from colleagues -- mostly French you suspect) and 2002,
> nearly nothing widely appreciated happened to Axiom -- contrast
> that to other CAS on the market.

On the contrary this was the period of time when NAG was investing
a lot of time and money into developing Axiom as a commercial
product. An entirely new and I think potentially quite revolutionary
user interface was developed. Axiom was ported to a new lisp environment
that permitted Axiom to be delivered on Windows. And the numerical
abilities of Axiom were greatly extended. Perhaps it is true that this
is not now "widely appreciated" but I think that is only because it
turned out that NAG decided to abandon it's attempt to market this
new version of Axiom. :(

Unfortunately when first Aldor and then later Axiom were made open
source very little of the innovative work done by NAG during this
period could be made available in the open source release due to
licensing restrictions. :( :( :(

> I call that deep hibernation; you may call it differently.

Yes. I call that a "damn shame". :)

> 2002 was when Tim came to Lyon putting Axiom hard forward at the
> international workshop on open source CAS :-)

Yes. My understanding is that Mike Dewar from NAG had been looking
for someone willing to take on Axiom as open source for some time
when Tim stepped forward. Thank you Tim.

> | 
> | Could you explain what you mean by "retroactively used to redesign
> | its past foundation"?
> 
> My understanding of your comments is that "people tried to show Axiom
> as competing symbolic computation systems, it fails.  Let's try to
> present it as not having anything to do with that, by design."

I think that is an unfair assessment of my statements.

> I don't believe Axiom's foundational and design principles can be
> meaningfully understood that way.  I don't believe earlier failure
> had to do with the fact that Axiom was presented a symbolic
> computation system.

I did not say that. The reasons for Axiom's failure as a commercial
product were no doubt very different and largely non-technical.

> But I suspect all have our own religions and beliefs :-)

Yes I suppose, but what does that have to do with Axiom?

Regards,
Bill Page.