News and other tidbits that Chad Cloman finds interesting enough to share
I have no idea how this is possible, and the math is beyond me, but an IBM employee named Craig Gentry has found a way to add and multiply encrypted data without first decrypting it. It’s called “fully homomorphic encryption.”
Link: http://www.technologyreview.com/…
(via Slashdot)
DNA can be used to make a one-way process suitable for encryption. I don’t completely understand the details, but the concept is interesting. I’m also thinking that DNA cryptography might be easier to implement than quantum cryptography.
Link: http://it.slashdot.org/…
So you think your encrypted data is safe? Here’s what would really happen.
Link: http://xkcd.com/…
It has been known for some time that the advent of quantum computers will completely destroy our existing public key encryption system, which depends on the difficulty of factoring a very large number. The appropriate quantum factoring algorithm already exists—we simply need to develop a functioning quantum computer on which to run it. Two research groups have moved the technology forward by creating very small proof-of-concept quantum computers that perform a modified version of the factoring algorithm. Their quantum computers are not scalable but do demonstrate that some of the core technology is working.
For those interested, the technical writeups are available here and here.
The main lesson from this is that you cannot encrypt data with today’s technology and expect it to be safe for more than a few decades at most (who knows—it could be years instead of decades). Also, I wonder if cryptologists are looking for something other than factoring to replace the one-way algorithm essential to public key encryption.
Link #1: http://it.slashdot.org/…
Link #2: http://arstechnica.com/…
Researchers were recently able to factor a specially formed (but hard to factor) 1039-bit number in a mere 11 months. It shouldn’t be too long before those 1024-bit encryption keys can be broken in a realistic amount of time. My key is 4096 bits, which was specifically discouraged by the key generation software because it was considered to be massive overkill. Maybe it pays to be paranoid. Of course quantum computers, if they ever become a reality, will make existing encryption methods obsolete.
Link: http://actualites.epfl.ch/…
(via Slashdot)
