Mathematicians and SDI: a quandary - Strategic Defense Initiative

While more than 6,700 scientists and engineers in universities have signed petitions agreeing not to do Star Wars research, the mathematical sciences board of the National Research Council has been more receptive. It has brought together officials of the Strategic Defense Initiative Organi- zation and university mathematicians to find out what SDI is doing for math, and what math can do for SDI.

Frank Gilfeather, staff director of the math board, says that the officials were briefed on some basic math problems that have to be solved before SDI can get off the ground. According to Gilfeather, there's a need for research in engineering and computation -- and in ''speeding things up,'' i.e., finding short cuts that would enable computers to respond to continuously updated in- formation about incoming missile tracks, and about which warheads are real and which are decoys. In math lingo, this is an ''optimization'' problem, and what makes the SDI version of it more difficult than any that mathematicians have faced is the sheer volume of constantly changing data that must be processed as computers simultaneously determine how best to deal with attacking missiles. Mathematicians will have to seek solutions similar to the Karmarkar algorithm, which is used to route computers to the fastest solutions of complex problems of minimizing costs and finding the best distribution of resources. Other branches of mathematics, like statistics, are faced with equally difficult SDI problems.

The academy board's action has divided mathematicians. Some are enthusiastic, because they think money from SDI may pay for computers on which to do research and that math breakthroughs made for SDI could be applied to such areas as artificial intelligence, telecommunications, and aero- dynamics.

Other mathematicians oppose the Star Wars overtures -- they don't think they and their colleagues should work on SDI, because the research is ''immoral'' or because strategic defense is impractical and too expensive. But since neither the math board nor any other body can tell researchers what to do, each will have to decide on hi own whether to accept SDI money, and whether math will gain or suffer by increased funding from defense sources.

COPYRIGHT 1987 Discover
COPYRIGHT 2004 Gale Group