Kevin Simler

A little context on the medium, to set the stage:

This is my first attempt at what @ncasenmare has been doing (brilliantly) for years: using simulations to explain complex ideas. It's possible to convey the same ideas using text and static images. But simulations are a lot more compelling, IMO. And also more *real.*

What do I mean by "real"? @DavidDeutschOxf has this great line in FoR — that reality "kicks back." Meaning it exists outside the mind and is capable of independent behavior.

1/ The model is more real, but the connection to the reality it is trying to model is not. You do this thing in this article that I see around here that I think is a general error mode, and it is exemplified by "I know this is an oversimplified model but...".

2/ This does a few things. The first obvious one is that oversimplification is more likely wrong. The next is not so obvious. By handwaving away this oversimplification, by not including a great deal of examination of the connection between model and reality, reality is lost...

3/ The model dominates attention and the phenomena are viewed through it, without any idea of how to see around it. This connection needs to be the *central* focus of the modelling process.

this is a very fair point. but i don't know how to balance it against my need to develop intuition around these kinds of complex systems. i can't focus on precise modeling until i have a basic sense for something