The Stage Butts algorithm is "finished" and superficially working, and it could in theory be merged at this point, depending on the appetite for risk. I am in the process of making sure it has unit test coverage, which is showing up a few bugs here and there. When I've done that, I'll call it finished without scare quotes.
I did make some (adjustable) assumptions about what the water level will be in a "full" or "empty" state and about how far open the valve needs to be for different rates of water transfer. Hopefully these are good assumptions, but if need be they are intended to be tweaked after it is up and running.
I believed that I was supposed to be working on the Stage Pi algorithm, and then, once it was done, I would move on to the Lady Hanham SAC one, assuming that nobody else had beaten me to it. My implementation of the Stage Pi algorithm is based on a statetools module I wrote, which is supposed to provide a framework for writing arbitrary control logic based on a state model/state diagram.
I tend to be a depth-first learner, so hopefully the rest of the logic will come more quickly.
