SETI@home

"I think it's just the open-face fish sandwiches that everybody wants."

Siduri: So what's the relationship between the SETI Institute and SETI@home?

Seth: Cordial. Well, there aren't many SETI experiments in the world. To begin they figure SETI's all one giant organization, which it's not. SETI's just an acronym. There are a half a dozen SETI experiments going on. But in terms of radio SETI, the big ones are right here in the Bay Area. What the Berkeley guys do, they're just two guys.

Siduri: So there's no formal relationship.

	This is Seth's telescope, but sometimes he lets other people drive it

Seth: Well, we support them, we give them a bit of money, because we know these guys. But it's a different experiment. What they do—I'll tell you what they do, because it's pretty interesting. If you look up you can see there, [pointing to a photograph] you see the way this thing works, which is that there's a big reflector: the radio waves come in, they bounce off, and they go into here, this buckyball [there's a bit I can't make out here; Seth is facing away from the tape recorder] boink boink, and then into the receiver. So that's what we're using. By moving this thing around on these tracks, we can point to anywhere within 20 degrees of the receiver, and then you just wait for the Earth to rotate. You get a lot of the sky with this thing. And then this telescope is pointed straight up and you can actually see... [voice trails off as Seth turns away again]

But there's a second receiver over here [mumble] so this is an old kind of receiver, picking up signals in random spots on the sky. We're looking at stars, but this is just pointed at random spots in the sky. Maybe somebody's using this thing [the main antenna] to look at pulsars, while this [the second receiver] is pointed at some random spot in the sky. The Berkeley guys collect data from this thing. And it's random spots in the sky, and it's moving around, and they never look at any patch for more than about one and a half seconds, so they can't use this integration trick. But they get all the data they want out of it, right? They figure, look, we don't know where ET is, and random spots are as good as any other spots. And you know, you could make that argument, nobody knows. We think it's better to look at stars, but nobody knows.

So they collect the data with that thing. And then they just have a setup over in the control room. And they collect this data and they just send it on back to hard disks in Berkeley. And then they take about 2 percent of those data and distribute them on the web for SETI@home. That's how that's done. It's much less sensitive, that data, and the real problem, actually, for them, is the followup. Because people find signals of course, just as we do. They'll find signals. And you're processing this chunk of data, these data were taken three weeks ago or three months ago, and you find something and send it back to Berkeley. And so they're running software that notes that, and just waits for that antenna perchance to be aimed back at the same patch of sky again, which it will, ultimately, once every two years or so. And it will say okay, and have that data processed, and compare those data with the ones that were processed maybe two years earlier. You know, this processed by Melvin and this processed by Molly, and see if they found the same signals at the same frequency, in which case they both get the trip to Stockholm.

Actually, even that wouldn't work, because you can work out the statistics, how often that would happen by chance, just random noise. You have to find it three times. And then it's probably...

Siduri: But people are falsifying signals, at SETI@home. There have been people actually...

Seth: Really?

Siduri: Just to get the...

[There's a break here, because we had to switch tapes in the recorder. Meanwhile, here's an old article about people tampering with SETI@home data: http://www.spaceviews.com/2000/01/23a.html]

Seth: Does this mean you're doing this?

Siduri: No, no. No, I'm not, but I was reading about security in peer-to-peer systems, and...

Cellphone: Ring! Ring!

Siduri: Oh, I'm sorry, that's me.

Seth: This could be the big one.

Cellphone: Ring!

Siduri: Be quiet. Are you going to be quiet now?

[The phone is quiet.]

Siduri: Um, and they were talking about how the fact that a person is getting no money from falsifying their end of the system does not necessarily mean that they won't do it. Like, there's all sorts of other reasons.

Seth: Money is not the only incentive?

Siduri: Exactly. And they pointed to problems that the people at SETI@home were having as an example of that.

Seth: Interesting. Yeah, I guess that's true.

Siduri: Nobody's getting paid to do this work. But people want to increase their standings, you know, so they'll just make up all kinds of stuff.

Seth: Lots of reasons other than money. I didn't know that they were actually faking the data. Because what the guys over at Berkeley normally do, is they send the same data set to at least three random downloads. So that they compare returned signals. And if somebody fakes a signal in their data, their chunk of data, it won't agree with the other two people. There's usually more than two others, but they would get the other two holders, who will agree but they won't agree with him. That kind of voting system. So I think that's the way they try and guard against this.

Siduri: So what's in Stockholm? Why would they get a trip to Stockholm?

Seth: Oh, yes, Shannon, you obviously haven't investigated that. Isn't that where they give you the Nobel Prize?

Siduri: Ohhhhkay, I see.

Seth: I think it's just the open-face fish sandwiches that everybody wants. "What's in Stockholm." Not much, I'll tell you.