Virtually Real Science
From simple number-crunching to fancy graphic display, each new application for computers has been quickly adopted by scientists. Many researchers are especially keen on computer simulation, which takes advantage of the machines' capacity to answer What if?
For example, Juergen Kienle of the Geophysical Institute and Zygmund Kowalik of the Institute of Marine Science worked together on a computer simulation of tsunamis that might be generated in Cook Inlet if a portion of Augustine Volcano fell into the water. Kienle and Kowalik fed in the proper equations and data, and out came a series mountains-in-nets graphics showing how high the wave would be at Homer Spit and other coastal locations, depending on tides, where the volcanic rock fell in, and how much of it fell.
Their simulation's graphics were easily understood, but primitive compared to what can be done with computer simulations now. They can be used to simulate a virtual reality which the researcher can experience directly.
Many people have heard of virtual reality in connection with entertainment. With appropriately wired goggles and gloves, audience members would be in the movie, not merely watching it. Stoop down, and something on the floor would apparently come closer; back up, and the door at the end of the apparent hall would recede. (And, if you're that sort of computer program, flex your trigger finger and your ray gun would shoot.)
Architects have also been working with virtual reality. The programs let them apparently walk through a building before any of it is actually built---useful in both uncovering errors and reassuring clients.
But there's no rule that says virtual reality programs have to be written to real-world scale. If architects wanted to "walk" through the heating ducts of a building instead of along its hallways, the program could be adjusted so that the virtual walkers were the size mice instead of men.
And if---for example---a molecular biologist wants to check the way a protein fits onto a cell membrane, virtual reality programs could let that protein seem to be biologist-sized. That's just about the state of the art right now, according to the April 3 issue of Science. Author Robert Pool notes especially the work of the Department of Computer Science at the University of North Carolina, Chapel Hill.
Pool reports that scientists use graphics workstations routinely in designing medications; they draw proteins in three dimensions to envision how drugs will fit into active sites. The UNC researchers believe they can go one better by letting the drug designer walk all around the molecules they study, and---thanks to a remote-control arm set to mirror the electromagnetic force between the drug and the protein---get a genuine feel for how the medicine's molecules will slide into the protein's receptor sites.
It's easy to see that playing with individual molecules would be compelling, illuminating, and hard to leave alone. I foresee electronically addicted molecular biologists.
They won't be the only ones. At a meeting of the American Physical Society, physicist John Joannopoulos handed out 3-D glasses to people attending his presentation. Joannopoulos was not content to tell them about results of his work on impurities in silicon; he showed them. It was no cartoon, but an accurately simulated representation of an invading oxygen atom and the electrons it disturbed, all obeying the laws of quantum mechanics. The simulation is useful and accurate enough so that Joannopoulos has made discoveries that could lead to improved processing of silicon.
Which, in turn, could lead to better computer chips. That seems only fair. For virtual reality to become a fully useful research tool, the simulations will need better images and faster responses---that is, better computers. Who knows---maybe next time, Kienle and Kowalik will be able to show that tsunami actually washing over Homer Spit, even though nobody gets wet.
