Internet users are being recruited to sift through pictures of the Stardust spacecraft's collection trays, looking for bits of cosmic dust that could contain clues to the origins of the solar system.
The NASA-backed effort, known as Stardust @ Home, is being run by researchers at the University of California at Berkeley — but the nonprofit Planetary Society is helping to get the word out about the project and recruit volunteers.
Internet users will have to sort through 1.6 million "movies," each showing a minute slice of Stardust's aerogel collectors, in hopes of spotting the faint tracks of interstellar dust particles captured while the collectors were exposed to space. The Berkeley team estimates that about 45 grains of dust are embedded in the aerogel.
“Think of the dust particles as bugs smashing into a really soft windshield,” Bruce Betts, the Planetary Society’s director of projects, explained in a news release. “On long trips, you expect to collide with a few bugs as you cruise down the highway, but in this case, you want to collect them intact, without smooshing, or vaporizing, them.”
The idea that hordes of Internet users can help with a scientific experiment was pioneered by the SETI @ Home project, which also has received support from the Planetary Society. In an e-mail Q&A with MSNBC.com, Betts discusses the new project and its connection to SETI @ Home.
Will Stardust @ Home be based on the same
It is really a different beast, though the computer end of it came from the same people. You are still putting data out and taking results back, but now the computer at the other end is a person's eyes and brain, not their computer processor. Also, the data transfer isn't automated. Users connect to the Web site and follow links to start looking at movies. Every time they complete a movie and return their result, they have the option to look at another.
Is this really the most efficient way to scan for grains? The Stardust @ Home folks have consulted with some top experts in automated image analysis, and the prospects weren't very good for doing it automated. This is compounded by the fact that they don't know exactly what the tracks will look like, though they have guesses. The human brain still wins for image recognition, especially if there might be some variation that you can't predict. This program also has the very nice side benefit of actively involving the public.
Will users know immediately whether a grain has been detected, or will the results of the work unit have to be passed back to the server for further analysis?Since they are doing the detecting, not their computer, they will know they have submitted a detection. As the press release says, there will be many fake tracks sent out to people to make sure they are paying attention and to assess their success at detections. People will get feedback on what come from their detections, though details of this aspect are still being worked out.
How will users who identify grains be identified, and will they be able to follow scientists as "their" grain is analyzed?We are still working this out, but yes, they should be able to follow their grain, and those who are among the first to identify each grain will be acknowledged.
To pre-register for Stardust @ Home and Planetary Society updates, check out the Web pages at Berkeley's and the .