My RTS will at least contain the major engine components that an RTS needs. We’ll see what it ends up as. It might become just an engine with a simple tower defense as a demo or something totally different, depending on technical hurdles and wild ideas coming up during development experiments. I am not sure whether I will support multiplayer yet, or whether I will make a series of single player tutorial steps where you write JavaScript or ClojureScript to control the AI of the game to overcome various scenarios (test cases), like CodingGame, but with fancier graphics and more complex rules similar to a commercial RTS. Currently I am leaning towards the latter, because it allows more freedom about simulation speed and no synchronization issues, but at the same time I want the game to be deterministic with a replay function, so I can’t generate too much simulation data. It will be hard to tack on multiplayer support later, so I need to make the decision before I start implementing the engine logic beyond the graphics. Currently I’ve been focusing on the graphics. I defer naming my game until I know what it will be about. I detest having to repeat myself in strategy games building the base, so I want it to be able to be automated as well as avoid any stressful micro with AI scripts, but I still want the game to be able to have human interaction with scripts as executable in-game tasks. The game should be designed such that the best human strategy should win over any AI macro/micro helper scripts.

See also for more random thoughts.