As fun as simply stacking cows sounds, I've got plans of turning this into a simple puzzle game. Unfortunately the code is currently as buggy as hell and I've screwed up the cow data structures by working things out on the fly. Furthermore I'm not sure from this initial state whether my original concept is going to work. I'm not yet ready to scrap getting something done before the end of the contest however; I'll spend some time tomorrow morning sketching out some new ideas and see if I can rescue this concept somehow, either with some better data structures or maybe with some savage redesign elements.
I've also started planning ahead for the engine refactoring of the Diagonal Game Library. By a strange twist of fortune I'm in the position in my grad research where I need a simple graphical test bed to try out some algorithm ideas. Given I'm rusty on nearly every windowing system out there I'm strongly considering adapting my game library, since I'm presently pretty clued in to how it works (thankfully as a grad student I own my IP by default, so it's possible to cross over code like this.)
The main drawback is that Diagonal currently has no GUI functionality and the input system is extremely basic. But given building a GUI system is one of the things I mean to add I'll move that the front of the queue and see how it goes. Since the guts of the system work on an event system I don't think it'll be too difficult to put in a simple GUI system framework with the next refactoring cycle.
I guess this means my next game will need to be window based. Or I could just use the opportunity to add menus, windows and level editing capability to Brixtar. Or (most likely) both.
I'll decide by tomorrow evening whether I can rescue this tower based game, or whether I should just cut my losses and work ahead on the Diagonal refactoring.