@julesh I'm not aware of that trick, I made this up myself just by following Euler's method for solving a diffeq. Maybe the one fancy-ish thing that they also did is the the use of streams to represent solutions?
@JadeMasterMath Look at the paper "Calculus in coinductive form" (for some reason I can't find a pdf in the first 30 seconds). I'll be impressed if you found the same thing, because it's a very strong contender for "most beautiful program ever written in human history"
@julesh wow, I am impressed this kinda thing is so revered. Iirc that paper does a *lot* more than the code I wrote.. the main thing was some sort of characterization of Taylor's theorem using streams? I don't have anything close that yet...