[CPSR Home Page] | [CPSR Newsletter Index] | [Winter 1997 Issue--Table of Contents ]

Letter to the Editor

by Tom Shea

CPSR News Volume 15, Number 1: Winter 1997

A Modest Proposal

Dear Editor:
I read with interest the article "Star Wars: Down but Not Out" by Carl Page in the Fall 1996 issue of the CPSR Newsletter.

Mr. Page noted that the so-called Star Wars program has had incredible persistence, even after it had been opposed by many prominent scientists and had suffered numerous technical failures in key areas. He gave some reasons for this, but one other reason occurred to me: Administrators, scientists, and staffers at the National Labs and within the weapons community are fighting for their livelihoods. They are smart, tough, and politically canny people who want to continue doing what they've become good at doing. Of course the Star Wars program has persisted despite obstacles!

Now that the Cold War has ended, what is to be done with such technical talent? There's been no national initiative to help the sword makers get into worthy ploughshare manufacturing enterprises. They're good at making weapons and they have secure government jobs; it's tough to translate bomb-making expertise into useful, equally secure peacetime pursuits. Moreover, the weapons makers are not necessarily the ones with the vision to see what they can do next that would be of equally high importance to their country.

I have an idea of what we might ask them to do next: use their skills in computer modeling and scientific visualization to create animated, data-driven graphic models of large economic flows. For example, an accurate economic model of the U.S. Federal budget could be constructed with data for the current year, as well as data accrued over the years. It should track both the Federal budget (anticipated revenues and costs set forth in the President's annual budget, which is mainly hopeful fiction) and what actually happened. Current-year-over-previous-years models could be run, as well as models showing trends over many years. The output could be accurately adjusted for the results of inflation; or, on the other hand, the results of inflation could be seen and understood, perhaps for the first time.

As was done with sensor data collected on unseen atomic reactions, these data streams could then be used to animate graphic metaphors that would make the underlying realities visible. One useful graphic metaphor would be pipes and vessels, such as those used in models of refineries and power plants. The pipes representing incoming revenues and outflowing expenses would increase or decrease in size to represent changes in money flow on a monthly or weekly basis. Storage tanks (which also would vary in size depending on how much money they hold) would represent the general fund, and other vessels would represent surpluses or invested funds. Underground wells (with varying depths) would indicate debts. The visual output of these models could be made available to experts, policy makers, and the general public by the simple expedient of transferring the output to video tape.

Attention could then be turned to modeling and visualizing even larger economic flows: international trade deficits and surpluses, the ebb and flow of international monetary funds across various world currencies. Also, once such models have been perfected at one level, they could be adapted to state governments (almost as complex and as little understood as the federal government) with equally beneficial results. Yes, there are already computer economic models aplenty, but most are special-purpose, proprietary, and very limited in scope, compared to what I'm proposing. Also, they're not readily understandable to the layman because the step of making them "scientifically visualized" has not been taken.

Granted, the problems involved in such an enterprise would be huge and challenging. But they are solvable-precisely by people with this kind of rare technical expertise and dogged determination to overcome obstacles. I'm told that the folks at Lawrence Livermore and elsewhere have modelled on computer the changing weather patterns of the entire globe (in spite of unbelievable complexity of fluid dynamics on this scale and the problems posed by chaos theory). Supposedly this global weather model is usefully predictive for up to 10 days ahead. If this can be done, then economic modelling can certainly be accomplished.

Now that the Cold War is over, the well-being of our country depends on economic strength, broad-based understanding of financial realities, and informed economic choices. Yet the level of real understanding of economic issues (underneath all the stormy idealogical debates) is pitiful, even among policy makers. A major presidential contender proposed (without much data) that the cure for the nation's ills was a flat 17% income tax. He may be right or wrong, but the point is, who really knows whether the idea makes sense or not? We are trying to make major decisions on the basis of soundbites and piecemeal statistical comparisons, almost invariably chosen for a partisan purpose. Except for perhaps a dozen senior economists, business people, and policymakers in the world, everyone is guessing about macro economic realities. And even these few experts disagree.

The techniques of computer modeling and scientific visualization could change all that if properly applied. Computer models can allow people to back off from "non-negotiable" positions without "losing face." Such models actually amplify human understanding, which, without assistance, is simply inadequate for comprehending problems of very high complexity. If computers have a beneficial use in the world (and they do) this is it: making the complex understandable and making the invisible visible.

Sincerely,
Tom Shea
sheatom@mail.apple.com

[Previous Article] | [Table of Contents] | [Next Article]

© Computer Professionals for Social Responsibility
P.O. Box 717 Palo Alto, CA 94302-0717
Tel. (415) 322-3778 Fax (415) 322-3798 webmaster@cpsr.org