Piston Developers

The semantic distinction between probability theory and "partial diversity" might work as a toy model to explain why general problem solving is much harder than domain specific problem solving.

I believe that "partial diversity" is the "actual" measurement that should be optimized for, however, this is impossible with bounded computational capacity. So, we approximate the semantics with probability theory.

I figured out how to measure the hidden assumption and wrote up this definition as "partial diversity". A subset of partial diversity has semantics similar to probability theory.

The problem is that there is a hidden assumption of random strings that hold for infinite computing capacity, but not for bounded computing capacity. The result is that probability of success becomes ambiguous and breaks down.

This paper shows an argument that the naive interpretation of a technique, to construct general problem solvers, that works given infinite computing capacity, doesn't work in practice when translated into probability theory.

This paper defines a function that has some nice mathematical properties for discrete probability distributions.

I have a hypothesis that extreme view politics causes violence, not video games. Source: Central topic of recorded history, plus the reduction of violence with the growth of video games (playing more video games means people have less time to do extreme view politics).

In order to study dependent typed languages, it is nice to either ignore type inference rules that are specific to some language, or ignore type inference rules that are general for dependent typed languages. The MX grammar helps to distinguish between those two cases.

The idea that "growing up" means stuff gets boring... It can't be an old idea. When hunter and gatherer societies celebrated adulthood, you were allowed to participate in the hunting. This sounds more fun to me. Somewhere along the timeline, stuff went wrong...

"Growing up" is not about becoming boring. Sometimes it is quite the opposite. When teaching people stuff in a boring way is the childish problem of the system, then "growing up" means making it more fun.

I believe it is the system that needs to "grow up".

A lawyer representing me once, just before the trial, criticized me for doing math on a paper while waiting. He said that when he was young, he liked math and could spend hours on it. One day he learned to "grow up". That trial taught me how dysfunctional the system can be.