[Top][All Lists]

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

RE: Re[2]: [Axiom-developer] Issue tracker: Language selection

From: Bill Page
Subject: RE: Re[2]: [Axiom-developer] Issue tracker: Language selection
Date: Sun, 9 Jan 2005 16:24:14 -0500

On Sunday, January 09, 2005 12:37 AM Vladimir Bondarenko wrote:
> ... 
> Via my computational flair, outstanding diligence, and iron will
> I am going to transform the current mire of (implementation of)
> computer algebra. with 5,000 Maple bugs -
> a major update is coming - is just the very beginning. Fasten your
> seat belts, please.

Your language encourage one to challenge your claims!

> To start with, do you realize that actually AXIOM Computer
> Algebra System, Version of Wednesday December 15, 2004 at
> 00:50:38 has at least thousands distinct bugs?

It isn't all that relevant how you count bugs as long as you
don't count, for example an error in integer arithmetic as an
uncountable number of errors...

But 5,000+ bugs in Maple is no surprize to me at all and an
even greater number of easily identifiable bugs in Axiom would
also not surprize me. In fact, I think your definition of
mathematics must be quite restricted if your estimated number
of bugs is this small. :) Finding bugs is easy and cheap. Fixing
bugs is an entirely different matter. I hope that you are able
to reserve some of your talent and energy for implementing new
correct algorithms.

I think it is likely that using a computer to do mathematics is
inherently limited in such a way that there are always likely to
be a large number of mathematical "bugs". Therefore I am not that 
optimistic about your overall stated goals for improving computer
algebra systems to such a state that they become an universally
essential tool. But in spite of these limitations, I do already
feel that systems like Maple, Mathematica and Axiom can severe
as significant mental prosthesis, as aids in the application of
advanced mathematics in many different areas where formal
mathematics would otherwise not be used to much advantage.

Bill Page.

reply via email to

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