MENU

HELP FIND ALIEN LIFE FORMS!

DOWNLOAD SETI@HOME

       SETI@HOME, DEVELOPED at the University of California at Berkeley, was officially launched on May 17, 1999 — although beta testers and Unix users got an early crack at it. Since then, the number of personal computers running the program has rapidly increased.
       Users from 226 countries and territories around the world have devoted about 280,000 years of computer time to the effort, with as much as a millennium’s worth added every day, organizers say. Versions of the program are available for Windows, Mac, various flavors of Unix, the Be operating system, OS/2 and more.
       SETI@home’s rapid rise to become the world’s biggest experiment in distributed computing has come as a surprise to Berkeley astronomer Dan Werthimer, the project’s chief scientist.
       A year ago, “I thought we’d be lucky to get 100,000 people using it,” he recalled.
       Werthimer said the 2 millionth volunteer downloaded the program Monday, providing an e-mail address and Web address based in Russia. He said there were no plans for ceremonies recognizing the first anniversary or the milestone user. But the Planetary Society, one of the project’s sponsors, announced that SETI@home users could download certificates of appreciation from the nonprofit group’s Web site.
       
HOW THE PROGRAM WORKS
       Here’s the basic scheme behind SETI@home: Radio signals received by Berkeley’s Project Serendip at the world’s biggest telescope dish — the 1,000-foot Arecibo Observatory in Puerto Rico — are carved into 330-kilobyte “work units” and distributed by a central computer to users over the Internet. The free client software works like a screen-saver, analyzing the data in the background or when your computer is otherwise idle.

       After a few hours or days of data-crunching, depending on how high-powered your computer is, the results from the work unit are uploaded, then a new work unit is downloaded to start the cycle again.
       When the results are returned to Berkeley, particularly interesting signals — for example, strong and steady tones in a narrow frequency band — are flagged for further analysis. For 40 years, SETI scientists have been engaged in an increasingly sophisticated search for such signals, believing that they could represent an intentional transmission from an advanced civilization beyond our solar system.

       Werthimer said analysts have followed up on some strong signals flagged by the software but hadn’t come up with any unexplained spikes. He said the spikes turned out to be either intentional test signals, radio interference from earthly transmitters or orbiting satellites — or spurious computer data that didn’t match up with the actual Arecibo readings.
       Project director David Anderson said SETI@home was only just starting the process of reviewing its huge database of candidate signals.
       “Until now, we’ve been stockpiling signals that may or may not be interesting,” he said. “There may be something extremely interesting lurking in our database right now, but until we go through this process of exhaustively scanning everything ... we won’t really know. And it’s conceivable that this back-end processing phase may go on long beyond the end of the project.”
       The next version of the SETI@home client software, now in beta testing, will analyze the data for pulsed signals (“bip-bip-bip ...”) as well as continuous tones (“beeeeeeee ...”), Anderson said. He said a search for pulsed radio signals “has never been done before in SETI research,” although it’s been done by astronomers searching for pulsars.
       The upgraded program will take about 20 percent longer to analyze each work unit, but that shouldn’t pose a problem since SETI@home is currently having a hard time keeping ahead of its “avalanche of users,” he said.
       Version 3.0 will have new graphics to represent the new search features, although Anderson said “I’m not sure if anyone is going to be able to understand it — it tends to look just like a lot of random noise flying by rather fast.”
       
TAKING THE LONG VIEW
       This first run at SETI@home was due to last two years, and Werthimer said the organizers still have to raise the money for a SETI sequel.
       SETI@home’s annual expenses add up to about $500,000, compared with about $100,000 a year for Serendip and $30,000 for Berkeley’s optical SETI projects, he said. But Werthimer cautioned that “it’s not fair” to compare the costs in that way, since SETI@home had so many startup costs.

       Assuming that money is raised for the second round of SETI@home, the project’s organizers plan to expand the program, widening the spectrum being searched and bringing in more radio data from Southern Serendip, a sister project conducted at the 213-foot (65-meter) Parkes Radio Telescope in Australia.

       Project director David Anderson concedes that the odds of finding an alien signal this time around are, well, astronomically low.
       “My most likely scenario is that 50 years from now, we’ll have a giant radio telescope out in the outer solar system, and it’s quite possible by then that the desktop computer will be as fast as all the (mainframe) computers we’re using in SETI@home,” he said. “I think when we get those two things the odds will go to 50 percent.”
       In the meantime, Anderson and others are looking into adapting the power of distributed computing for other applications. Even before SETI@home, the strategy was used by groups such as Distributed.Net to crack encryption codes that were once thought unassailable. Another project, called Casino-21, plans to apply distributed computing to climate modeling.

       Anderson said he is studying a proposal involving biodiversity research.
       “These biodiversity people have huge databases of sightings of particular animal species over the entire world for the past couple of centuries,” he explained. “What they’re interested in doing is combining that information with geographical information, like altitude and climate data, and basically compute mathematical models of what the habitats of different animals are.”
       If such models could be developed, researchers might be able to forecast how changes in particular habitats could affect various species — “to predict extinctions, basically,” and take appropriate action well in advance, Anderson said.
       “That’s an area that the world can get excited about,” he said, “in some ways much more than SETI.”

MSNBC. 18 March 1999. SETI Software Hits Milestone.