Given the horsepower the NSA is throwing at this stuff, testing for book ciphers doesn't seem like an intractable problem. I'd suggest we simply start attachin

Given the horsepower the NSA is throwing at this stuff, testing for book ciphers doesn't seem like an intractable problem. I'd suggest we simply start attaching modestly large files of random numbers encrypted (or not) using ridiculously secure keys to our emails. Incidentally, if the ratio of content to encryption key length is high, the likelihood of false positive decryptions will be high.