drector@orion.oac.uci.edu (David Rector) (06/05/91)
For those who asked, this is a summary of the responses to my request for
information about numerical analysis and graphics work being done in
Smalltalk. My thanks to all of the people who contributed information
to me.
Book:
Object Oriented Programming with Smalltalk
by: Dusko Savic.
Ellis Horwood series in computers and their applications.
ISBN 0-13-040692-9
It discusses how to use Smalltalk/V to do such things as:
- Matrix operations, including determinants, Gauss-Jordan
elimination, linear transformations, and eigenvalues.
- 2-D and 3-D graphical transformations.
- Runge-Kutta method for ordinary differential equations.
- Solving parabolic partial differential equations.
Comments of respondents:
1. All this is quite good. Although he does suggests that the collection
iterators such as select: collect: etc shouldn't be used. Ignore this
advice.
2. BEWARE, there are many code errors in the book.
-------------
Project:
Computer Aids for Industrial Productivity. (CAIP)
Sandra Walther, Dick Peskin
Rutgers University
CAIP Center P. O. Box 1390 Rutgers University
Piscataway, N. J. 08855-1390
peskin@caip.rutgers.edu
walther@caip.rutgers.edu
A project which is using Smalltalk for research into scientific
visualization and computational fluid dynamics. They have Smalltalk
systems for solution of systems of non-linear differential equations
and some early work on the PDE solvers. This work has accompanying
graphics systems for display of results. In addition they have done
work on matrix manipulations.
-------------
Other sources:
1. You should talk with Kurt Hebel (hebel@cerl.cs.uiuc.edu), who
has developed a lot of numerical software for Smalltalk-80.
I think that most of it is adaptable for Smalltalk-V.
2. Knowledge Systems, Inc. has a good math toolbox for Smalltalk/V.
-------------
My own work:
For the next two years, I will be developing tools in Smalltalk for
teaching numerical analysis. I intend to make them publically
available when finished.
Since my research area is formal semantics in computer algebra, I intend
to make a virtue of necessity by studying the data types needed for
numerical analysis programming.
I welcome contributions and suggestions.
--
David L. Rector drector@orion.oac.uci.edu
Dept. of Math. U. C. Irvine, Irvine CA 92717