I think "irreducible to simple rules" is really the heart of that. A program is a finite set of rules which are followed consistently and deterministically[1].

I think "irreducible to simple rules" is really the heart of that. A program is a finite set of rules which are followed consistently and deterministically[1]. Programs are also almost infinitely malleable. Find a bug, fix a bug, it's gone forever[2]. People, societies, and markets have none of these properties. Everyone has their own rules, which they don't even follow consistently, and change is very difficult. People who are drawn toward one kind of reasoning are often literally incapable of the other. How many economists really understand programming[3]? The appeal of Asimov's "psychohistory" to programmers is obvious, but it's fiction. Unfortunately it's fiction that many programmers believe when they try their hand at economics or political science. [1] Mostly. I could write a whole "falsehoods programmers believe about programming" about the cases where that model doesn't apply. [2] Again, mostly. Updates take time, old versions persist, etc. [3] A few, I'm sure, but that doesn't affect the point.