4.5-4.8

1. This was a nice history lesson. The hardest part was the different modes.

2. There's a lot of ways to measure whether some encryption is "secure enough". If there is no demand for the data being encrypted, any encryption (or none) is good enough. DES and some other algorithms seem to get their security from the fact that breaking it is computationally infeasible, but general purpose programming on data-parallel processors (graphics cards) is becoming more and more accessible. I'm not exaggerating when I say that a top of the line GPU today (with 240 compute cores) is over two orders of magnitude faster than the top Intel chips (Core i7 as they're called) today. So what, do we continue to increase key length? We've done that a little with RSA and it's a testament to Moore's law and capitalist ingenuity that we've had to. But when quantum computing becomes available, the time to break RSA becomes polynomial (maybe even linear) with the key length, and not exponential. It's almost stupid to get cracking on encrypted data now since you could wait for (or work toward) quantum computers and crack the encryption long before your classical computing device yielded any fruit. I guess what I'm saying is computationally infeasible is super relative and proofs of security should not rely on such a property.