SAN FRANCISCO – The annual Cryptographers' Panel at the RSA Conference 2010 is part state of the union on cryptography and security, and part homage to the pioneers of encryption. It can be a dizzying discussion on hash functions and broken encryption algorithms; a nirvana for nerds. But this year, however, the Shamirs, Rivests and Diffies and Hellmans of the cryptoworld were joined on stage by the National Security Agency, making for a bit of good natured contention as well.
Brian Snow, former NSA technical director of information assurance, took his share of jabs from the heavyweight panel and connected with a couple of roundhouses of his own about the agency's capabilities compared to those of the commercial world. The result was an entertaining 50 minutes that left an overflowing hall of attendees wanting more.
Moderator Ari Juels, chief scientist and director of RSA Labs, prodded Snow with a question about the NSA's interests and advantages over the private sector, which sparked a lively back-and-forth with notable cryptographers Whitfield Diffie and Adi Shamir. Snow said the NSA has a "more nuanced posture" and works on a range of areas that likely has little overlap with the corporate America, such as nuclear command and control (C&C) systems.
"Where we do overlap, we cheat," Snow said. "We read what you publish, but we do not publish what we study. We have good budget and an aggressive, talented staff. We have PhDs doing nothing but cryptography; that's a nice department. We have a better knowledge base and more stuff than what you have. The NSA is still ahead, a small handful of years, on average."
The comment rankled Diffie, best known for his groundbreaking work with public-key cryptography, who countered, for example, that nuclear C&C is not out of bounds for cryptographers.
Shamir, one of the creators of the RSA public-key encryption system, also challenged Snow by pointing out that in a few recently declassified NSA technical journal titles, there was no mention of public key cryptography. "Doesn't that demonstrate that the NSA would have been way behind?" Shamir asked.
Snow answered: "People invent things in parallel, and sometimes don't always use the same terminology."
Spirited disagreements aside, the panel touched on its usual wide array of security topics, such as the recent deaths of PKI innovator Shaun Wiley and Ned Neuberg, a former NSA agent who tried to recruit Diffie during the 1980s; renewed interest in Suite B cryptography; a tribute to the work of Ralph Merkle, another public key cryptography pioneer; and David Chaum, inventor of many cryptoprotocols for his work on voting system security.
The panel, which also included Ron Rivest and Diffie-Hellman protocol co-inventor Martin Hellman, closed out its annual session with a discussion on whether any of them had ever done anything foolish that turned out to be a wise decision.
Rivest, co-inventor of the RSA algorithm with Shamir and Len Adelman, said it was foolish to assume what we know now is the best that can be done. "Foolishness is having the merit to step out there and draw the line and say that's the best I can do," Rivest said.
Shamir, meanwhile, was a little less philosophical.
"I'm about 99% fool," he said. "Every morning, I go to my office as a scientist, and work on problems that I've been looking at for a years with no success. It's a long shot and about once every three months, I have a good idea. In the other 99 days, I work on something and make no headway whatsover. That is normal in our profession. My employer could have hired someone who would be 100% successful because they have set out simple tasks to achieve; for some reason, they picked me over the other guy."
And Diffie was a little more direct: "I've rarely done anything else [but be foolish]."