ABOUT COSMIC LOG

Quantum fluctuations in space, science, exploration and other cosmic fields... served up regularly by MSNBC.com science editor Alan Boyle since 2002.

Alan Boyle covers the physical sciences, anthropology, technological innovation and space science and exploration for MSNBC.com. He is a winner of the AAAS Science Journalism Award, the NASW Science-in-Society Award and other honors; a contributor to "A Field Guide for Science Writers"; and a member of the board of the Council for the Advancement of Science Writing.

Check out Boyle's biography or send a message to Cosmic Log via cosmiclog@msnbc.com.



The stardust hunt is on

Posted: Monday, July 31, 2006 8:15 PM by Alan Boyle

After months of preparation, the Stardust @ Home treasure hunt kicks off on Tuesday, with tens of thousands of Internet users primed to look for grains of dust from beyond the solar system. The research effort adds a human touch to the grand tradition of distributed-computing projects such as SETI @ Home.


NASA
An artist's conception shows tracks
of particles embedded within
Stardust's aerogel-filled collector.
In actuality, it won't be this easy to
spot the tracks.

Beginning at 2 p.m. ET Tuesday, you'll be able to click into the Stardust @ Home Web site and start scanning the actual photomicrographs of interstellar-dust collectors from NASA's Stardust probe. Using a novel "virtual microscope," you can look for the telltale trails of bits that embedded themselves in Stardust's fluffy cubes of aerogel.

During its seven-year mission, Stardust's racket-shaped collector snagged flecks from Comet Wild 2 on one side, and perhaps 50 to 100 bits of interstellar dust on the other side. Those bits could provide new clues to our cosmic origins - but they're not easy to find. The project's amateur researchers will have to look through 700,000 or more "focus movies" multiple times in order to spot the flecks.

You'll have plenty of company in the task: Andrew Westphal, Stardust @ Home's project director and associate director of the Space Sciences Laboratory at the University of California at Berkeley, told me that 115,000 users already have signed up to participate. That's not quite as many as the roughly 500,000 active users for SETI @ Home, the search for alien signals, but if all those people join in, Stardust @ Home would still rank as one of the world's biggest distributed-computing projects.

"We think that we have a science team of 115,000 people," Westphal said.

This time, the emphasis is on the people rather than raw computing power. The project depends on people actually spotting the dust tracks and reporting them, rather than just letting a computer churn away unattended.

"It's a pattern that's very obvious to people, but it's almost impossible to program a computer to recognize it," said Amir Alexander of the California-based Planetary Society, one of the beta testers. (The Planetary Society is a collaborator in the Stardust @ Home and the SETI @ Home projects.)

Alexander said doing the search didn't seem tedious to him. "It's really kind of like a computer game - once you get the hang of it, it's pretty addictive," he told me.


Stardust @ Home / Planetary Society
This screenshot provides a preview of
Stardust @ Home's "virtual microscope."

Westphal expects the project to run for many months. "The bottleneck of the project ... is how fast we can collect the interstellar images," he said. So far, 11 of Stardust's 132 collection tiles have been scanned, and another four tiles are due to be scanned every week at NASA's Johnson Space Center.

To keep the searchers interested - and to make sure they're on the ball - Westphal's team is adding simulated dust trails to the imagery. The users' skill at spotting the simulated trails (as well as the actual trails, of course) will be factored into their score. "As they go along, they will be kept up to date on how they're doing," Westphal said.

Each microscopic area will be looked at multiple times to make sure nothing is missed - and everlasting glory awaits those who are the first to spot honest-to-goodness dust grains.

Westphal said the first person to discover a particular grain has the "privilege of naming the particle," as long as the name isn't obscene or otherwise objectionable. The discoverers also "will be co-authors on any papers that we write" associated with their particular bits, Westphal said.

Everyone should start out on roughly equal footing, Westphal said. The earlier beta tests were carried out with simulated Stardust imagery. Meanwhile, the actual microscopic imagery has been seen by "nobody except us, and we've been quite careful not to look at anything too much," Westphal said.

So practice your skills with this 12-part tutorial, and prepare to begin the great stardust hunt.

MAIN PAGE

Email this EMAIL THIS

Comments

I'm wondering how difficult this would be to automate this search a bit.  For instance, many web sites are now protected with an alpha-numeric registration system that requires a user to input a string of characters that are displayed by off-center graphics (yahoo mail is a good example of this).  This is supposedly hack-proof against autoregistration, but there have been many improvements in OCR technology that allows autoregistration to take place anyway.

I wonder if the same hacking technology could be adapted to this with any success.
I just went through the tutorial, and it seems to me like it will be lots of fun.  I can't wait to get started.  To use humans instead of computers to search stuff like this is a great thing because it gives us more understanding and the feeling that we, regular everyday people, can be part of not only this world, but the rest of the universe.  Some of this dust could have been created billions of years ago!  And I can be one to see traces of it for the first time!  That is truly awesome.
I've just passed the 10 test "movies". I then registered, only to see that the login to begin the actual searching was disabled. Now, the entire home page has to links to anything. Looks like they've disabled the entire entry site, most likely due to high traffic volume problems. I guess it's to be expected.

The searching actually seems very relaxing to me. Looking forward to actually beginning sometime in the near future!
I went through the tutorial process (100% :-D) and logged in, then viewed about 20 of the "real" virtual slides.  Four or five of them were "calibration images" that were thrown in to see if I was paying attention.  I was.  Unfortunately their website was shut down right about then.  I hope it was because of too much enthusiastic help, and not some jerk spamming their site!
Searching the slides is actually pretty addictive and relaxing.  It's nice to be able to do something important for the scientific community, and it would be pretty cool to be the first to spot a specific impact trace in the Aerogel.
Does anybody know whats going on???
I have registered and searched for an hour. Now the website seems to be turned off since yesterday evening. Its frustrating when you dont know whats going on??
Hope somebody has some info.
Thanks
I would hope that anyone interested in helping would not want to miss one slide! I asm very excited to be able to participate. Imagine, you could see something no one has EVER seen before. I am encouraging everyone to help. Think what a fun family project it can be for the kids. Lord knows, most public schools would never be able offer an opprtunity like this to a child.  
Hi, frustrated seekers: I'm trying to get in touch with the Stardust @ Home folks, but based on the forum comments, it sounds as if spurious images are showing up in the "Virtual Microscope" ... e.g., images of a bride and groom, lake and trees, etc. So the S @ H team is trying to figure out where this weird imagery is coming from. It sounds as if the stock images were interspersed on the Amazon server that is hosting the aerogel images. It also sounds as if they were bogged down with traffice when they opened the doors. I'm sorry to hear about the glitches, I guess you'll have to periodically check the Stardust page for updates relating to this problem.
OK, I heard back from Andrew Westphal at Berkeley. He says the site is back up. They had to track down the "weird image" problem, and found it was due to a "communication issue" or "crossed wires" with the server where the images were being stored. "We hadn't seen it" before the launch because the glitch happened relatively rarely, but users did notice it and mentioned it on the S @ H forums. There was also a hardware problem that created hangups just after the official launch. Aaanyway, Westphal says the issue was resolved and the system is back in service.
I was on for about 6 hours last night. I did have trouble getting into the site initially but once i logged on, i didnt lose it again. I managed to view about 75 slides. it was incredible. beware, they slip a lot of "calibration" slides in too. I had 21 slides out of 75 were calibration.
It was very exciting. i saw about 20 possible tracks out of all the slides. What a great age we live in!! I'm ready to go to Mars now!!


SEND A COMMENT

PLEASE READ: All comments must be approved before appearing in the thread; time and space constraints prevent all comments from appearing. We will only approve comments that are directly related to the blog, use appropriate language and are not attacking the comments of others.

Message (please, no HTML tags. Web addresses will be hyperlinked):

TRACKBACKS

Trackbacks are links to weblogs that reference this post. Like comments, trackbacks do not appear until approved by us. The trackback URL for this post is: http://cosmiclog.msnbc.msn.com/trackback.aspx?PostID=1554

Latest Tech & Science News

Syndicate This Site

Add Cosmic Log to your news reader:
live.com xml
myyahoo msn
bloglines newsgator
google