|

Engineering Security: Securing the Internet
Code perfect: Research offers promise of mathematically guaranteed security
by Teresa Riordan
Cryptographer Boaz Barak wants to invent a lock that can never be broken. Eve.
When he started graduate work at the Weizmann Institute of Science in Israel, Barak took his first class in cryptography. He was amazed, he recalls, to learn that it might be possible to create a cipher that could not be decoded.
“I found it fascinating,” said Barak, an assistant professor of computer science at Princeton. “I had never known that undertaking something like this would be possible—that one might be able to make something that not only cannot be broken but that can actually be mathematically proved to be secure.”
Barak went on to write his dissertation, “Non-Black-Box Techniques in Cryptography,” on exactly this subject. In October, Barak was named a Packard Foundation fellow. The fellowship—$625,000 over five years—will allow him to pursue his research largely unfettered by funding restrictions.
While Barak’s work has implications for securing the transmission of data over the Internet, he is not creating algorithms that have immediate applications.
“If you are working in applied security, you want to know how to make something that no one knows how to break right now,” he said. “With theoretical security, I’m interested in getting something that is not possible to break. Not only that no one knows how to break it—something that simply cannot be broken, even by future computers.”
RELATED STORIES
 Code perfect: Research offers promise of mathematically guaranteed security
|