It probably is easier to do something like what Rayex is doing, but the main reason I’ve not spent effort improving that instead is that I think shoehorning a C-based state machine-like API into Elixir isn’t what will make a great tool for making a game in Elixir. Of course, you can abstract a more idiomatic layer in Elixir on top of Raylib, but I think it would end up being confusing to reason about. Needing to understand Elixir, the abstraction layer, Raylib, and how the abstraction layer communicates to Raylib, just seems like too much. That doesn’t even get into the complexities of actual game development. It’s no wonder nobody is hungry for that system to exist.
The ease I’ve had in jumping into various projects in Elixir and understand what’s going on and even make changes has been inspiring. I personally feel like that ease could apply to game development as well, which would allow for rapid development. Which isn’t a new idea - that’s exactly why scripting languages are used in many games. I just have a crazy idea of using Elixir for the core engine and the “scripting” language.
There’s also a ton of interesting things thanks to easy DSLs, like creating shaders, objects, scenes, etc., but that’s a ways away.
