Wednesday, June 26, 2013

National Security and Noise in the data

I was thinking about the problems with so much security data being electronic and the apparent ease of electronic file theft, and it made me think back to some studies of the attack on Pearl Harbor in 1942. After the fact, historians can go back and try to piece together info saying that the warnings/clues were there, but before the fact it's all very noisy. Hindsight enables us to see the important info. That same principle applied on 9-11. Even when you might have some clues about an attack that may be coming - it's very difficult to put all the pieces together in advance. It makes me think maybe one of the best things we could do with national security data (or for corporations - critical corporate data) is intentionally seed the data with all sorts of bad data - making it so noisy that the info would be very difficult to use by spies. In the case of an insider like Snowden, perhaps make it so that some folks working with the data may not know whether it's real or not - despite the layers of encryption. My guess is alot of this already happens - allowing spies to capture intentionally false designs and bad info but allowing them to think it may be real. The conspiracy theorist in me thinks some leaks of this nature are probably intentional just to see what happens.

No comments: