I didn't push too hard to get stuff done today. I started moving the serialization registry definitions into the modules that defined the type in question, on the logic that, since the serialization code relies so heavily on the type's layout and constructor behavior.
For after that's done, I looked over yesterday's post, and incorporated the tasks and orderings into my todo file. So, now I've got a really thorough list of things that need to happen, roughly in dependency order. It's a little intimidating, but I'll just try to make progress on it every day.
Looking over the list is also making me think about what directions I want to take this code in after I reach the end of part 13. My inclination is to figure out how to translate the code manually to Python, coming up with whatever library support I need. From there, find collaborators, and work on polishing the code and expanding it into a tutorial series. (I don't feel qualified to do a tutorial series on Coconut, and it seems like I'm using some of its features similarly to how my old boss's wife used Class Namespaces: as a stepping stone or temporary measure before or during a refactor.) I'll also want to revisit the old idea I had of adapting the code into specific games that I'm more interested in. Probably focus on trying to make non-code resources and concept stuff for them.
For now, though, the best thing to work on, is sleeping. Good night.