Secrets and Lies: Digital Security in a Networked World, 15th Anniversary Edition

7

Cryptography in Context

If cryptography is so powerful, why do security breaches occur? Why are there electronic theft, fraud, privacy violations, and all of the other security problems discussed in the previous chapters? Why isn't cryptography the perfect answer to all our security needs? Why am I bothering with the rest of this book?

Surprisingly enough, it's not because of bad cryptography. (Enough of that is out there, but the problems are even more serious.) The answer lies in the difference between theory and practice.

Cryptography is a branch of mathematics. Mathematics is theoretical; mathematics is logical. Good mathematics starts with sound premises, follows a single road—proof after proof—over complex terrain, and ends with unassailable conclusions. By its nature, it looks good on paper.

Security is rooted in the physical world. The physical world is not logical. It is not orderly. There is no single road. There are theories and conclusions, but in order to accept the conclusions you have to accept the premises, the models, and the relationship between the theories and the world. And that's not easy. People don't play along. They do the unexpected; they break the rules. Hardware is the same way: It breaks down, it misbehaves once in a while. Software, too. Software should be logical and orderly—it's only ones and zeros, after all—but it is often so complex that it behaves more like an organism than a piece of mathematics. No matter how good the cryptographic theory ...