A new way of keeping private correspondence private

When Barack Obama became American president, one of his first tussles with White House lawyers was over whether he could keep his beloved BlackBerry. (Yes, he did.) The reason why the lawyers were wary was that e-mail cannot be destroyed. People do not know where the information they are sending is being stored and when, if ever, it is deleted. Such unknowns make it possible for seemingly long-gone data to turn up in a court under the order of a subpoena, or worse, in the hands of a hacker. On August 13th, though, a team of computer scientists led by Roxana Geambasu of the University of Washington, Seattle will unveil to the 18th USENIX Security Symposium in Montreal an e-communications system that destroys messages soon after they have been sent.

The technique devised by Ms. Geambasu and her colleagues uses one of the least secure areas of the web to store encryption keys that self-destruct after a certain period. Peer-to-peer networks, or P2Ps, originated in the late 1990s with the rise of music-sharing networks such as Napster and KaZaA. Individual users would log on and allow other people to download music from their computers while simultaneously downloading music for themselves. In recent years P2Ps have become vast file-sharing networks for information in all its forms. Dr. Geambasu and her colleagues realized that because computers logged on and off P2Ps at a fairly steady rate, they could use these networks as places to store encryption keys temporarily.

The researchers developed a piece of software called “Vanish”, which encrypts information before it is sent, breaks the encryption key into pieces and then sends the bits out to randomly selected “nodes” created by computers that are logged on to the P2P network. Once sitting on a node, the pieces of the key wait for another copy of the Vanish software to access them in order to read the encrypted message. However, the pieces of key do not remain on the P2P in perpetuity. When a computer is disconnected from the network, the node it formed ceases to exist and any encryption-key data stored there are lost.

This means that, in the first few hours following the dissemination of a key, the number of pieces required to make a sent message readable can easily be gathered. However, as time goes by and more computers on the P2P are disconnected, the encryption can no longer be cracked and the message in effect self-destructs. At the moment, the message lasts for about eight hours before vanishing but Dr. Geambasu suggests that this duration could be extended or diminished by adjusting the number of pieces of key sent out.

Past systems designed to encrypt e-mails all relied on ways of storing their keys safely. Government agencies, however, have found ways of making computer companies hand over those keys. The new software leaves nothing to hand over, so it looks as if it will be able to make old messages completely unreadable. Perhaps the ultimate measure of how secure the system really is will be whether Mr. Obama himself ends up using it.

Source: Economist.com