Press Release

Alien-hunters who don’t do their homework

By SpaceRef Editor
April 25, 2001
Filed under , ,

The game may be up for people who cheat in “distributed computing” projects, in which computer owners donate or sell their surplus computer time. Projects like this include the Search for Extraterrestrial Intelligence (SETI), where participants use their PCs to hunt for intelligent radio transmissions from outer space.


SETI was the first large-scale distributed-computing project, and it relies solely on volunteers. But in the past few years, companies have been racing to capitalise on the idea, paying participants in proportion to the computer downtime they donate. Instead of searching for ET, these companies tackle giant number-crunching problems such as identifying genes or cancer-blocking molecules, or cracking codes.


And whenever money is on offer it’s only a matter of time before some clever hacker figures out how to make it look as if their computer did calculations that it did not-allowing them to earn cash while they play computer games or surf the Net.


But now Philippe Golle and Ilya Mironov, both graduate students in computer science at Stanford University, have developed a way to detect cheats. They insert a few fake “hits” into each data set sent out to be crunched. If a participant returns the data without finding the planted hits, the company knows the data was not searched.


Suppose a company was hired to uncover a secret key used to encrypt a plain-text message. Each participant’s computer must analyse thousands of possible keys. Cheats could claim to have tried all their keys, but since they don’t know how many fake correct keys are hidden in their data, they must try all of them to earn their fee.


Even when no payment is involved, as with SETI, some people cheat to be listed as a top performer on the project’s website. With 2.7 million members, SETI sends each data set to multiple participants, thus ensuring all the data is thoroughly searched. But for companies without that luxury, the Stanford scheme could be useful, says David Anderson, director of SETIhome and chief technical officer of United Devices, a commercial distributed-computing company.


###

Author: Catherine Zandonella

New Scientist issue: 28th April 2001

Please mention New Scientist as the source of this story and, if publishing online, please carry a hyperlink to: http://www.newscientist.com


Contact: Claire Bowles

claire.bowles@rbi.co.uk

44-207-331-2751

New Scientist

SpaceRef staff editor.