Wouldn't it be something if you could combine the power of your computer with those of hundreds, thousands, even millions of others? If you're connected to a local area network at work or the Internet at home or the office, you're already experiencing the benefits of computer connectivity. But typically you're just sharing information, programs, or storage space with other computers. You're not pooling the processing power of each PC. "Distributed computing" takes the network one step further. In many ways, it's an out-of-this-world concept, and it's thus fitting that the first popular distributed computing project involves the search for intelligent life elsewhere in the universe. [email protected], setiathome.ssl.berkeley.edu, has been attracting volunteers since May 1999. Already a whopping 3 million have opened their computers through the Internet to the Space Sciences Laboratory of the University of California at Berkeley, helping the lab analyze radio signals from outer space for signs of intelligence. It's all very, well, far out. But it's very logical as well. By breaking down a huge computational project into smaller tasks and distributing them to many different computers, the work gets done faster. The process itself is painless. You download a small program that typically acts as a screensaver -- Windows, Mac, and Linux users can all play. The screensaver kicks in when you're not using your computer, so it doesn't interfere with your work. Every few days it gathers together its data and sends it back to the mother ship, that is, the Berkeley lab. You don't need a high-speed Internet connection -- a 28.8 modem works fine -- and for disk space all you need is 10 spare megabytes. To promote security, the [email protected] software is designed to download and upload data only from the Berkeley lab. For similar reasons, the Berkeley lab hasn't publicly released the source code of the software. [email protected] has received a lot of attention, but it's not the only option available in digital altruism. If you're a more down-to-earth type, you can participate in cancer research through the United Devices Inc.'s project, which was developed in conjunction with Britain's Oxford University and is supported with funding from computer giant Intel Corp. Nearly half a million people so far have joined this effort. The process is similar to that with [email protected], but instead of analyzing radio signals from space, your computer analyzes molecules for their suitability in the formulation of anti-cancer drugs. Helping others is great, but if you're more attuned to doing well than doing good, a number of companies have begun to offer distributed computing services that can harness your organization's existing computers to solve complex business or research problems. Entropia, at www.entropia.com, has the bottom line firmly in mind with its offerings, which it suggests for use in areas such as financial analysis, Web testing, bioinformatics, or computational chemistry. The company doesn't ignore social responsibility, however, allowing individuals to download its software for what it describes as "cause computing." You can currently sign up for two projects. The first, in conjunction with the Scripps Research Institute, La Jolla, Calif., involves the study of drug resistance and drug design in the fight against AIDS. The second, in conjunction with the University of Rochester's William E. Simon Graduate School of Business Administration, involves the study of world market volatility. All this is very impressive sounding, and some of the rhetoric inevitably gets more than a bit grandiose. United Devices enthuses: "We all get to feel terrific because we are changing the world." Not to be outdone, Entropia claims to have "quite literally built a computer the size of planet Earth." Still, the numbers, if not yet the results, really are impressive. [email protected]'s 3 million volunteers collectively have created a relatively inexpensive distributed supercomputer with greater raw processing power than the fastest discrete supercomputer. [email protected] contends its network of computers is rated at 15 teraflops and has cost just $500,000, while IBM's ASCI White supercomputer is rated at 12 teraflops and costs $110 million. Distributed computing does have its limitations, however. Unlike discrete supercomputers such as IBM's, it's not effective with problems such as weather forecasting in which individual calculations affect each other. With distributed computing, as currently developed, the calculations must be independent of one another. And though it has great potential, distributing computing is still too new to have scored any significant achievements. [email protected], for instance, has not detected any radio signals that indicate the presence of extraterrestrial intelligence. Yet the search, and the development of ever more sophisticated computer technology, continues. Reid Goldsborough is a syndicated columnist and author of the book Straight Talk About the Information Superhighway. He can be reached at [email protected] or http://members.home.net/reidgold.