It wasn't the code itself that I learned from. I have more been enriched and stimulated by reading the blog posts, documentation, hacker news threads, and maili

It wasn't the code itself that I learned from. I have more been enriched and stimulated by reading the blog posts, documentation, hacker news threads, and mailing list. A couple of the more interesting ideas are: * He created Nock, which in a way is bytecode language, like compiled java bytecode or the .NET CIL. But his idea was that this bytecode should be the simplest possible thing, far, far simpler than the CLR. In fact, it should be versioned in Kelvin versioning, starting at 5,000 and counting down, until it is finally perfected and will never need to change. Going forward, all consumer apps will always compile down to this bytecode. All new hardware platforms can build interpreters for this bytecode. I think that is a pretty novel and neat approach. If it caught on, it would ensure that any program we wrote now could be run for the next thousand years. * His view is that to beat spam, you simply need to have a finite number of cryptographically secured identities. This number can be large. But if it is finite, that means accounts will not be costless, which means the market over time will be able to solve the problems of trust and filtering out spam in a way far superior to how it works today. It's hard to do the ideas justice by trying to repeat them myself. In reading through the material it was just lots of little things, where I said to myself, "Ah, that is a neat solution to that problem, I wonder if he'll be able to make it work."