While poking about defining the data required to properly model the game, I've also been thinking about the way the actual game logic will operate on these disparate objects. I'm currently considering a Non-Deterministic Finite Automata approach, where game entities are just collections of game state and potential/executing behavior, but I'm worried that it will lead me down a rabbit hole of attempting to implement an overly-generic framework for building the game, rather than actually working on the game. Oh, rabbit-holes. Thanks to you, I should have my level 10 spelunker's merit badge by now...
It's an exceptionally dangerous proposition for me due to two things. First, the abstract concept is like a neon sign in my head. I can see the core implementation like it was drawn up on my favorite 12'x8' whiteboard at work. Secondly, the challenges in implementing a solid, generic framework for developing future games are, honestly, more interesting to me than the challenges of designing a rule set for a specific game.
I might have to do a little more thinking on this and see if I can break it down into a set of quickly-attainable stories to put in Pivotal. If so, maybe I can keep myself from haring off course too far and kill two birds with one stone. Namely, get the game firmly started while implementing a system that will make feature addition and future game design a lot simpler.