Personal
InformationWilliam M. MacDonald is a physics professor at the University of Maryland whose primary interest is the development of educational material to teach physics using computers.
Professional InformationProfessional
biographyWilliam M. MacDonald received his undergraduate degree Summa Cum Laude from the University of Pittsburgh and completed his doctoral dissertation under the direction of Eugene P. Wigner on the "Validity of the Isospin Quantum Number for Light Nuclei". As a graduate student he spent two summers at the United States Bureau of Mines in Pittsburgh and published papers with John M. Richardson on use of Thiele semi-invariants for solving combustion equations and on a generalization of the Hartree-Fock equations for systems of fermions, bosons, and Boltzmann particles at non-zero temperatures. Other summers he spent at Los Alamos National Laboratory and at Project Matterhorn at Princeton University on theoretical computations for the hydrogen bomb. After earning his doctorate, he spent a year at the Lawrence Berkeley Laboratory where he, Marshall Rosenbluth, and David Judd developed the so-called Fokker-Planck equation for collisions in plasmas and applied it to predict the failure of the magnetic mirror machine for thermonuclear fusion. Besides studies in plasma physics, the RMJ equation has been the basis of many studies in astrophysics. At the University of Maryland he investigated the effect of Coulomb distortion on super-allowed beta decays and its implications for the conserved vector current theory and the Cabibbo angle, collaborated with Martin Walt of the Lockheed Missiles and Research Laboratory on a series of papers analyzing the lifetimes and spectra of geomagnetically trapped particles, with Leonard Rodberg developed the shell model approach to reaction theory, investigated fine structure in low-energy neutron cross sections, and did the first, and only, calculation of the spreading width of an isobaric analog doorway state produced through coupling to complex configurations by Coulomb and charge-dependent nuclear forces.
In 1979, MacDonald organized a conference at the University of Maryland to discuss the establishment of a national supercomputer center for theoretical nuclear physics. This resulted in a panel chaired by him that produced a report recommending the establishment of such a center. Although this was budgeted into the Long-Range Plan for Nuclear Physics submitted to the Department of Energy that year, no action was taken by that agency. However, the Physics Division of the National Science Foundation followed up on this concept in 1981 by forming a Subcommittee on "Computational Facilities"[1] (on which he served) that considered this issue and recommended the establishment of National Supercomputer Centers. Subsequent studies and reports then led to the establishment by the NSF of centers at Cornell University, Princeton University, the University of Illinois, the University of Pittsburgh, and San Diego,
Since 1985 MacDonald has been most interested in the role of the computer in physics education, particularly at the undergraduate level. For more than a century both the content and the organization of undergraduate physics courses have been largely dictated by the restriction to problems that can be solved exactly in terms of algebraic expressions and the more elementary functions of mathematical physics. The advent of the computer has made it possible for a student to investigate problems of real interest, such as the effect of air resistance and spin on the trajectory of a golf ball[3], or of the deviations from the inverse square law of gravitational attraction upon planetary motion. Professor MacDonald was a member of Maryland Project in Physics Education and Technology and developed utilities, programs, and material for students to use computers in introductory physics courses. The guiding principles of the M.U.P.P.E.T.[2] project have inspired similar efforts at universities throughout the world.
The use of the computer for teaching physics requires publication of high-quality software for lecture demonstrations and computer laboratories. This software must be extremely flexible so as to allow the student to investigate many aspects of a physical system, including special cases that are exactly solvable and that can be used to bridge the transition to cases solvable only numerically. The material should be easily integrable into existing courses, yet provide a gateway to the development of courses more reflective of the approach and current interests of working physicists. To meet this need the Consortium for Upper-level Physics Software (CUPS) [4,5] was organized in 1990 and directed by Professors Maria Dworzecka and Robert Ehlich of George Mason University and by Professor MacDonald. The CUPS project was funded by the Physics Division of the National Science Founation, with additional support provided by IBM, Apple Computer, and George Mason University. Twenty-seven physicists from the U.S.A., Canada, Australia, and the United Kingdom were responsible for developing packages consisting of a text-book and at least six multi-part simulations for each of the nine physics courses that are the core of the undergraduate physics curriculum in most parts of the world. The nine packages were tested throughout the United States and in several other countries. Publication of the CUPS series by John Wiley and Sons was completed in 1995. The CUPS simulations have already won many of the annual awards given by "Computers in Physics."
For many years, beginning with APL, MacDonald has been interested in the use of "interpreted" computing languages in the sciences and engineering. More recently he became convinced that symbolic manipulation packages that combine numeric and symbolic computating with graphical displays for visualization will produce fundamental changes in both research and education in the sciences and engineering. In 1990 he began teaching and developing a course called " Mathematica for Scientists and Engineers" that used Mathematica, a language based on rules applied through pattern recognition, that integrates symbolic and numerical computing with extraordinary graphics capabilities. In 1995 he proposed the addition of a new course to the physics curriculum that would introduce upper level physics majors to the theoretical methods used in physics and integrate the use of a symbolic computing program. The course, " Intermediate Theoretical Methods," was adopted as part of the University of Maryland physics curriculum in 1996.
Selected
Publications