The minisymposium will take place Sunday, June 1. The times and rooms are yet to be arranged.

Professor Jonathan M. Borwein,

Director, Centre for Experimental and Constructive Mathematics

Simon Fraser University

Burnaby, B. C.

* Title:*

Doing Mathematics on the Web

* Abstract*

``Doing" intentionally captures both ``communicating and publishing" and ``thinking, calculating and computing". I will talk about some of the issues involved in publication over the internet and show some of the tools that CECM has developed and is developing for

- interactive mathematics documents
- intelligent linkage of text and computational tools
- direct collaboration between researchers; or
- between researchers, students and practitioners

Sam Dooley

IBM T. J. Watson Research Center

dooley@watson.ibm.com

* Title:*

Derivational Problem-Solving System for Elementary Linear Algebra

* Abstract:*

A primary obstacle in the application of computer algebra to mathematics instruction is the mismatch between the goals of the symbolic systems and the requirements of instructional settings. A derivational system for a domain of computation is a regime for organizing problem-solving facilities which allows a student to build competence in a collection of computational skills while providing a correct and consistent approach to informal symbolic reasoning. This talk presents a description of the components of a derivational system and discusses how these components realize an advantage over more conventional computer algebra systems in achieving the goals of mathematics instruction. A derivational system for elementary linear algebra is described to illustrate the approach to symbolic problem-solving which is presented by such a system.

Keith Geddes

Symbolic Computation Group

Department of Computer Science

University of Waterloo

* Title:*

Maple as a Tool for Scientific Computation

* Abstract:*

The Maple scientific computation environment supports both symbolic and numeric mathematical computation. This presentation explores the enhanced mathematical problem-solving capabilities which can be achieved via a hybrid symbolic-numeric approach in various problem areas. The objective is to achieve an appropriate combination of symbolic mathematical analysis and numerical computation.

Applications include the development of algorithms for efficient numerical evaluation of functions, the evaluation of definite integrals in the presence of singularities, and the generation of problem-specific numerical methods for solving differential equations. The hybrid approach will be seen to be particularly suited to the teaching of scientific computation.

David J. Jeffrey

Department of Applied Mathematics

University of Western Ontario

London, Canada

djj@apmaths.uwo.ca

* Title:*
Mathematics for the Million:

Automating Mathematics Using Computers

* Abstract:*

Computer Algebra Systems (Derive, Maple and Mathematica) are now being
used quite widely in educational contexts. For educators and students
alike, the systems hold out the promise of increasing their
effectiveness. By means of examples, and---I hope---demonstrations, I
shall discuss some of the difficulties that the developers of Maple
and Derive face in making their systems reliable tools for student
use. The problems are not problems in programming, but rather are
problems in mathematics and in the way people think of
mathematics. The responsibilities for change and improvement lie not
solely with the developers, but also with the mathematical and
educational community. The mathematical level of the talk will mostly
stay at the level of first-year mathematics.

Fri May 2 12:23:10 EDT 1997