There are a number of observational phenomena around how scitech happens, but it is all fuzzy and disconnected: finance people talk about Wright’s Law^1, business school people talk about modularity^2, complexity scientists talk about evolutionary dynamics^3, historians talk about incremental vs radical innovation^4, and video games model progress as tech trees^5.

Jerry Neumann ‘unified’ the ideas of incremental/radical innovation in neumannOneProcess2020: showing that if you model technology as a modular tree that advances through a stochastic process it produces a power law in innovation size but that interacts with our intuitions that things are all normally distributed to create the sense that there are two kinds of innovation when in fact there is just a ~continuous distribution. The question is, could you pull this trick for all of the phenomena we observe about science and technology?

The first step is obviously to just keep a list of the disparate phenomena that people have observed.

- S-curves
- Incremental vs Radical Innovation
- Learning-by-doing/cost curves
- Wright’s Law
- Progress Functions
- Modularity
- Evolutionary dynamics
- The Valley of Death

^1: See Wright’s Law Predicted 109 Years of Auto Production Costs, and Now Tesla’s