The Stardust mission will return its comet dust samples to earth on the 15th of January if all goes well. Observers in the north west USA will get to see the Sample Return Capsule as it passes through the atmosphere if the weather cooperates, more details at the above link.

Seems Berkeley will run the Stardust@Home project (non BOINC) that recruits volunteers to find the dust samples. If you wish to participate you can signup at the website but please note the following;

"First, you will go through a web-based training session. This is not for everyone: you must pass a test to qualify to register to participate."

---

is the aerogel the dust collector something full of buble filled foamy glass?! and are the participants supposed to find a dust particle among the bubles?

---

Mandtugai!

is the aerogel the dust collector something full of buble filled foamy glass?! and are the participants supposed to find a dust particle among the bubles?

The samples will be visible to the naked eye believe it or not.

---

"By faith we understand that the universe was formed at God's command, so that what is seen was not made out of what was visible". Hebrews 11.3

I love that aerogel--its so cool! I would love to see it for myself one day up close and personal.

---

OOPPPs Never mind

Got on the mailing list.

---

TEAM
LL

is the aerogel the dust collector something full of buble filled foamy glass?! and are the participants supposed to find a dust particle among the bubles?

The samples will be visible to the naked eye believe it or not.

Uhuh :)

Finding the incredibly tiny interstellar dust impacts in the Stardust Interstellar Dust Collector (SIDC) will be extremely difficult. Because dust detectors on the Ulysses and Galileo spacecraft have detected interstellar dust streaming into the solar system, we know there should be about 45 interstellar dust impacts in the SIDC. These impacts can only be found using a high-magnification microscope with a field of view smaller than a grain of salt. But the aerogel collector that we have to search (is) enormous by comparison, about a tenth of a square meter (about a square foot) in size. The job is roughly equivalent to searching for 45 ants in an entire football field, one 5cm by 5cm (2 inch by 2 inch) square at a time! More than 1.6 million individual fields of view will have to (be) searched to find the interstellar dust grains. We estimate that it would take more than twenty years of continuous scanning for us to search the entire collector by ourselves.

Dunno who wrote it, but I made it more legible. (word) ;)

---

My sister and I have our names on a microchip on the craft :)

---

"By faith we understand that the universe was formed at God's command, so that what is seen was not made out of what was visible". Hebrews 11.3

yeah me too, it is realy fun sending our names for 7 years and 2.8 billion miles. and are'nt they going to display the chip in smithsonian ?!

well the mission page says the dusts will be visible with cone shaped shooting trace inside the aerogel?!

---

Mandtugai!

trace

---

Mandtugai!

yeah me too, it is realy fun sending our names for 7 years and 2.8 billion miles. and are'nt they going to display the chip in smithsonian ?!

well the mission page says the dusts will be visible with cone shaped shooting trace inside the aerogel?!

look Stardust@home

---

"By faith we understand that the universe was formed at God's command, so that what is seen was not made out of what was visible". Hebrews 11.3

once stardust@home begins im putting all my cpu time to that...

Wonder how serious it is. Send my own name in as wel as Wilhelmina Kudtje from the same ip adress. LOL

once stardust@home begins im putting all my cpu time to that...

This isn't your computer doing the work, it's you. You have to pass a test and prove that you can find these things by visually scanning movies. I have a tough time believing this can't be done by computers... Humans may be good at pattern recognition, but computers are far superior in objectively completing tedious tasks.

---

Here's an Installation Guide.
Try the Wiki for other questions.

I have a tough time believing this can't be done by computers... Humans may be good at pattern recognition, but computers are far superior in objectively completing tedious tasks.

I couldn't disagree more.

For want of a better name, security images (the anti-spammer kind that you have to type in if you want to start up an account) pose a difficult task to most computers. The image can be defeated, but the amount of brute-force pattern recognition needed is not feasible. That's why they work well (unless you're dyslexic or using a non-visual browser). Think of the amount of processing needed to find the "Nina"s embedded in Hirschfeld's art. Humans can do it very well; computers, simply, suck at it.

That said, I'm sure it'll be real fun to pass the time looking at images one by one for hours on end. It'd be great. Pure joy. Sheer fun. Wouldn't miss it for the world... That damn capsule had better crash... Can't believe I signed up...

---

touched down whooo!

---

Mandtugai!

For want of a better name, security images (the anti-spammer kind that you have to type in if you want to start up an account) pose a difficult task to most computers. The image can be defeated, but the amount of brute-force pattern recognition needed is not feasible. That's why they work well (unless you're dyslexic or using a non-visual browser). Think of the amount of processing needed to find the "Nina"s embedded in Hirschfeld's art. Humans can do it very well; computers, simply, suck at it.

But now ask humans to solve 3 million of those anti-spammer things. It would take far fewer man hours to write a program to do it. Sure, it would take some sophistication, but for a 200 million project, I think they'll selling themselves short. There's no telling what kinds of biases will be introduced by doing it this way... and I think they'll have to search everything more than 4 times to be sure no one missed it.

Do you really think the amount of processing time/code writing time would be less than 30,000 hours? They could have been writing the code during the 7 years it was in space... ample time to develop better pattern recognition software technologies. Then they'd spend a month taking all the necessary images and another month running and tweaking the program. Now they're counting on volunteers to do it in 6 months... your guess is as good as mine as to how reasonable that is.

Well have fun... wanna find the redshifts of my 650 galaxies while you're at it? I could have a program do it, but I'm just too lazy to write the software...

[edit]
Oh, and what happens next time when this project works so well that they want to make it 10 times bigger... do you think this kind of effort is easily scalable? Computers and software get faster and cheaper, humans don't.
[/edit]

---

Here's an Installation Guide.
Try the Wiki for other questions.

Stardust lands in Utah successfully.

---

"By faith we understand that the universe was formed at God's command, so that what is seen was not made out of what was visible". Hebrews 11.3

Do you really think the amount of processing time/code writing time would be less than 30,000 hours? They could have been writing the code during the 7 years it was in space... ample time to develop better pattern recognition software technologies. Then they'd spend a month taking all the necessary images and another month running and tweaking the program. Now they're counting on volunteers to do it in 6 months... your guess is as good as mine as to how reasonable that is.

Software could do the job, but this article explains why it would be impractical to use pattern recognition tools.

To save the reader reviewing the full text (though I do recommend it) the bottom line is that pattern recognition software must be trained. Since they have no verified particles on which to train the software, they are a bit stumped.

---

To save the reader reviewing the full text (though I do recommend it) the bottom line is that pattern recognition software must be trained. Since they have no verified particles on which to train the software, they are a bit stumped.

I already read it and don't buy it. Users will be trained with particle hits in astrogel from particle accelators. If humans can be trained, software can be too.

---

Here's an Installation Guide.
Try the Wiki for other questions.

"If humans can be trained, software can be too"......

I'd think twice about that statement!

Yeah, software could probably be created for Stardust@Home but when we're talking about looking for patterns, nothing beats human eyes. HUMANS are programmed to recognize patterns in just about any old mess (i.e. pictures in the clouds, constellations, satan's face in the WTC explosion, etc).

This isn't to say that computers aren't a lot FASTER than humans and could probably flag more potential impact sites in a shorter amount of time, but I bet they'd miss a lot too. Besides, what a great way to get Joe Average directly involved in real research.

I don't know about you but potentially having my name printed in a scientific journal article is pretty durn cool!

---