RE: [Axiom-developer] Failure of Axiom? (was: Static versusDynamically typed) )

[Bertfried Fauser]

Dear Bill and others,

        even if I was very quiet, due to an work overload with 'regular
busines', I might add to this philosophy discussion that I _do_ need a
strongly typed language! I had much discussion with the Marn le Valle
group about Mupad. Their argument is that there is already now to few
development power to come up with compeeting systems/packages (say in the
combinatorial/Hopf algebra world). And combinat for Mupad is quite
impressive, collecting algorithms from SYMMETRICA/SCHUR and other
packages.

To add types to Maple needs lots of (very slow and painstaking) code which
does not do any math for you. Furthermore its error prone.

I have no experience with Mupad, but it seems to have (some of) the same
deficiencies as has Maple. Only very clever data formats (keeping type
informations, as in the combinat package) may help, which poses the burden
of typing once more to the user.

My experience with any CAS (including the terrible but brilliantly fast
SCHUR) is that one needs a good start (local Guru). AXIOM has its problems
with the steep learning curve for the beginner. Thats what stoped some of
my more realistics approches to my problems, but will not stop me to try
to learn more AXIOM programming to do it.

I had some time ago a discussion with Maple protagonists in .de and they
assured, after being questioned, that Maple tries to pursue engineers,
studendts etc. Hence they try to approch large (in quantity) groups, since
there they can make a deal there. Scientific achievements or features
needed to implement them are not on the priority list. AXIOM users seem to
be of such a kind that they have problems which do not fit into the
Maple/Mathematica standard solution set. These are ambitious people PhD
students etc. So I think it will take more time unless these people become
multipliers and teach onw studentds AXIOM.

Meanwhile I will stop typing in philosophy `nonsense` and try to do some
more science and AXIOM programming... ;-))

kind regards
BF.

% PD Dr Bertfried Fauser
%     Institution: Max Planck Institut for Math, Leipzig <http://www.mis.mpg.de>
%   Privat Docent: University of Konstanz, Phys Dept 
<http://www.uni-konstanz.de>
%  contact|->URL : http://clifford.physik.uni-konstanz.de/~fauser/
%          Phone : Leipzig +49 341 9959 735  Konstanz +49 7531 693491