Stardust@Home Lets Public Search Grains of Dust

An anonymous reader writes "In a new project called Stardust@home, UC Berkeley researchers are inviting Internet users to help them search for a few dozen submicroscopic grains of interstellar dust captured by NASA's Stardust spacecraft. Rather than relying on the user's spare PC cycles, though, the system depends on their eyes." From the article: "Though Stardust's main mission was to capture dust from the tail of comet Wild 2 - dust dating from the origins of the solar system some 4.5 billion years ago - it also captured a sprinkling of dust from distant stars, perhaps created in supernova explosions less than 10 million years ago."

Time is money

[PrinceAshitaka]

My computer cycles I could care less about but my time is valuable to me. Are there really that many people out there that A. want to to this AND B. have the time to do this. I am sure there are many people in catagory A and in catagory B. How many people are in both catagories?

Many eyes

[spge]

If people are prepared to spot themselves on Google Earth [theregister.co.uk], as well as other things [theregister.co.uk], there's no reason why they won't look for specks of stardust.

Image processing/pattern recognition?

[Anonymous Coward]

Why wouldn't they use image processing or pattern recognition techniques to do this? I couldn't find anything about this in the article, but i'm sure that if humans are able to detect specks of dust, they can also train a pattern recognizer to do the same, if not better.

Using porn sites

[Frans Faase]

Maybe they could set up a porn-site where you have to 'pay' by judging an image. It seems that this technique is used a lot by cyber-criminals for surpassing the "human-only" test that are used during registration to prevent robots from registering. These test consist of text messages that are easy to read by humans but not by computers. It is a very smart idea to use humans (horny men, in this case) for performing a simple "computational" task.

I wonder if this idea can be extended. Using humans to perform computational tasks sounds to be a very interesting business model.

Couldnt a computer do this better ?

[bxbaser]

if a human can recognize the trails shouldnt a computer be able to ?
The article didnt mention any reason why a computer would not be able to do this.
does anyone know anything more about this.

makes me wonder is this is some sort of trial to test a distributed voulenteer workforce and they needed something interesting to get participants.