Stave Off

This is my entry for the Spring Lisp Game Jam 2024.

Playable here.

Stave Off is a Breakout game with a twist: you have to survive, there are no levels and the longer you've been playing, the harder it gets (well that was the plan, anyway).

How to run or build

First you need ClojureScript (quick start).

Then you can use the targets in the Makefile:

• make repl lauches a browser running the game and starts a repl connected to it.
• make release produces a release build in ./out/main.js.
• make serve launches a local web server that serves the release build.
• make dist produces a release build and packages it in dist.zip, ready to be uploaded to itch.

If you are on Windows, you'll have to adapt the commands on your own to get things to work.

Post-mortem

This is the first time I use a Lisp and the first time I write anything significant in a functional style. I chose ClojureScript for its ability to compile down to JavaScript, which allows having a web version for the game, which is always nice for game jams.

Ultimately I didn't really enjoy using ClojureScript to develop this game. The language is diametrically opposed to how I like programming languages: static and procedural, as opposed to dynamic and functional-style. The alien syntax is annoying to work with, though one thing I regret a bit is that I didn't write any macros, which I read is probably the only redeeming factor about it. I had a bit of fun at the beginning of the jam getting accustomed to all the weirdness but by the halfway point, I found myself not wanting to implement any new feature and try to write as little code as possible because it felt like a chore to program anything. Near the end of the jam, the only fun I had was implementing the particle system. It's very possible that my half-baked architecture was ill suited for functional programming, so I'd be interested to try making another game in another functional language. The repl driven approach sounded nice at first, but I feel like my tooling (VS Code & the cljs repl) just wasn't good enough for it because it felt unwieldy. I tried to lean hard on what I understand is the Clojure way of doing things, by using maps for most things. Multimethods piqued my interest and I feel like the data-driven approach Clojure goes for works better than the traditional object soup for dynamic languages. However, I don't think I'll be writing any more ClojureScript or Clojure going forward.