"Research is not like selling hot dogs," reflects Al Globus. "You don't get much immediate gratification, and many projects don't pay off at all. But every once in a while you get a payoff like penicillin -- a payoff that compensates for all the other projects 10,000 times over." Globus hasn't yet discovered a substance to alter the future of humanity, but on his supercomputer at NASA's Ames Research Center, he has designed a program to lead other scientists closer to a technology that could change the world. And in a realm of science where results significant enough to alert the general public are rare, last week was momentous. On April 21, NASA sent out press release 97-27, quoting Globus as claiming some astonishing breakthroughs: "Thanks to the simulation of molecular-sized gears by a NASA supercomputer," he said, "hope is growing that products made of thousands of tiny machines that could self-repair or adapt to the environment can ultimately be constructed."In other words, the dream of nanotechnology and its microscopic robots might just come true. "Laboratories all over the world are learning how to control nature at the atomic level," Globus says by phone from his NASA office in Mountain View, California, citing one laboratory's recent "stretching" of a single DNA molecule and continuing progress at IBM Labs, where as early as 1988 scientists manipulated molecules to form their company's logo.Globus' own achievement is documented in a paper co-authored with three Ames colleagues and soon to be published in the journal Nanotechnology. It details their successful computer simulation of a nanomachine, using virtual benzyne molecules as "gear teeth" attached to the outside of a "nanotube" made of carbon atoms. To make the contraption work, scientists had to simulate a supercooled virtual environment made up of helium and neon gases on a supercomputer running a few billion operations per second (your average Pentium runs about 25 million). Despite the machine's power, Globus admits that the transition from virtual to real is tricky. "There's always the issue," he says, "of whether this approximation is appropriate to conditions you can create in the lab."Born in Chicago and raised in Seal Beach, Al Globus had at one time in his life settled on a career as a jazz musician, and even studied at the prestigious Berklee School of Music in Boston. But when he went to live in the Bay Area in '73 he found the boho life unsatisfying -- not so much financially as spiritually. "I played in bar bands," he says, "and one day I looked out at the crowd and realized that my role in life was to get these people to buy alcohol. That didn't seem to me a worthwhile contribution to humanity." So he signed on at the University of California at Santa Cruz in the information-science department, back in the dizzy days when the new machine and its proverbial soul were just beginning to have an impact on the public at large. When he graduated in 1979, "there was essentially one programmer for every 20 jobs available," Globus recalls. "You could go to work anywhere you felt like it."Globus, who early in his college days had become obsessed with Gerard O'Neill's space colonies after reading about them in Coevolutionary Quarterly, felt like going to work for NASA. "It was the most amazing thing I'd ever heard," he says of O'Neill's concept, and it remained so until two years ago, when Globus and NASA colleague Creon Levit heard a lecture by Xerox PARC's noted nanotechnologist Ralph Merkle. Globus recognized immediately the value of nanotechnology in the orbiting planetoid homes of his fantasies. So did NASA head Dan Goldin, who around the same time acknowledged that molecular machines might build products of profound usefulness in space exploration -- among them, materials many times lighter and stronger than any existing alloy. "We basically stopped doing what we were supposed to do," Globus says, "and went to work on nanotechnology."Globus still wants to colonize space; in fact, one of his more successful projects has been the development of image-analysis software to study bone exposed to weightlessness. But the weeny machines of the future are his passion, and a responsibility he doesn't take lightly. This November, he and Merkle are co-chairing the 1997 Biennial Nanotechnology Conference in Palo Alto, and he's considering initiating the debate on nano-ethics before things get any closer to real."Any powerful technology can be used for damage," Globus acknowledges. "I'm quite sure that nanotechnology will be no exception." Things have already begun to go awry in fiction. In Ben Bova's recent novel Moonrise, nanomachines not only construct things, but take them apart, too. Bova introduces "gobblers," nanobugs that can dismantle matter. When they fall into the wrong hands, planes crash, men die and the world nearly ends."We've seen that, since the atomic bomb, nations have had to change their behavior in order to survive," Globus says. "What is necessary, I suspect, in the nanotechnology age is that we had better learn to get along better on an individual basis."Cloning, biotechnology, nanotechnology," he concludes, "all of these things are going to require us to put a little more love in our hearts."