Seti@home gets an upgrade

It could make the greatest discovery of all time

By BBC News Online's Damian Carrington

When the Seti@home project began in May 1999, using a computer screensaver to search for alien signals seemed almost as crazy as searching at all.

On Friday, the second version of Seti@home is launched, so BBC News Online spoke to Dr David Anderson, the Project Director at the Space Sciences Lab, University of California, Berkeley.

He told us about beating hackers, the fears of running out of data and what will happen when they have finished the sky visible from their telescope.

What improvements or changes are there in v2.0?

Enhanced security - this prevents people from sending back altered or duplicate results. We've also put in better support for proxies and firewalls. This well let more people in businesses, which usually have firewalls, use SETI@home. Finally we've put in Improved graphics to show the curve-fitting process which is central to our data analysis.

What is latest information on the amount of data analyzed?

We've now got about six months of data analyzed.

Is it true you actually have too much computer time now and are increasing the complexity of the analysis to use it all?

Yes. We'll be adding code that checks for pulsating signals. Currently we check only for continuous signals. We're also currently analyzing each piece of data twice, and comparing the results, to guard against results that are wrong either due to tampering or computer malfunction - of which there is a measurable amount. Bu it looks like this new science code won't make it into version 2.0, but will appear in a 3.0 in about a month's time.

Are you now keeping up with the data produced and how big an archive of data is there to delve back into?

We're analyzing slightly faster than data is being produced. We still have about a two-month backlog. Hopefully, the new analysis will prevent us from running out of data.

How do you answer critics who say that your software is inefficient and wastes computer time?

It's absolutely true that our software is not 100% efficient. For example, drawing graphics uses computer time. But it's not a waste of time - it engages users, and helps bring in more participants. Also, our software is not optimized for every type of computer processor out there. For example, our Windows version uses only standard Intel 386 instructions, not MMX, or other proprietary variations. We don't have the resources to manage hundreds of specialized versions.

How do you gauge success when the chances of achieving your goal - a signal - are very small indeed?

Like other Seti projects, we're covering a certain part of the sky in a particular frequency band. If we do that as well as we can, with the available technology, then we've succeeded. Some people might find this a bit discouraging, but we feel it's very much worth doing.

Seti@home has other measures of success as well - bringing a huge number of people into the scientific process, raising public awareness of Seti, and establishing internet distributed computing as a way of doing science research. I think we've been quite successful by these measures.

Given the vastness of space and the short time we have been on Earth, don't you think the chances of success are so small that your efforts could be better employed in other areas of astronomy?

No! I personally feel that intelligence is the most interesting phenomenon there is, and that contacting a distinct intelligence - presumably with their own versions of math, philosophy, music, art, etc. would be the most significant event in human history. It's possible that there are lots of intelligent civilizations out there, and they may have technology that lets them send radio signals with much greater power than what we have. So our chances of success may be larger than we think.

What plans, beyond V2.0, do you have for the future to improve Seti@home?

One big goal is to provide more feedback to users about the state of the project - e.g. what kinds of signals we are getting and how interesting they are. We hope to have some kind of weekly newsletter on the website fairly soon.

In 15 months or so, we will have finished scanning the part of the sky that Arecibo can see. After that we may have a follow-on project using southern-hemisphere telescopes, and/or different frequency bands.

What other projects do you believe are contributing well to the overall Seti effort?

Some people, including our chief scientist Dan Werthimer, are getting interested in optical Seti, that's looking for laser flashes. See http://seti.berkeley.edu  There's also a project to build a "compound eye" radio telescope, called 1HT, to be used full-time for Seti.

 TOP