In 2022 [0], Jan David Nose (jdno) released a game / programming challenge called _Auto Traffic Control_ [1]. Rather than controlling airplanes with UI interactions, the simulation is a GRPC server that expects the player to construct a game client that issues flight paths.
I tried to make a game client for jdno's ATC [2]. My client was deficient, as I tried to treat the movement as a fixed interval, by only submitting flight plans that moved in cartesian directions. Only by cheating, modifying the simulator, could my greedy algorithm land any planes [3]. I suspect I was supposed to continuously model and poll the future positions of airplanes when developing a flight plan.
I suspect that I won't have much competition on the leaderboard, despite the asterisk, as jdno eventually archived the project [4]. Originally, the project was self-publication while looking for work. The Rust Foundation hired him a little while afterward.
[0] https://jdno.dev/auto-traffic-control/
[1] https://auto-traffic-control.com/
[2] https://github.com/Nzen/hircine-anise-okkoz
[3] https://www.youtube.com/watch?v=czsSjy1Uqyk 1.5 minutes
[4] https://github.com/jdno/auto-traffic-control/commit/b5972dfb...
I think about this game quite often and would love to eventually pick it up as a project again. I was only able to scratch the surface of what I had in mind, and I’m sure there’s so much to learn from players about what makes a good programming game.
Imagine how cool it would be if something like this was based on the actual FF-ICE / FIXM flight plan format.
I have to admit that I have exactly zero knowledge about actual flight plans or anything else related to air traffic control. Just loved playing Flight Control and used it as inspiration for this game.
Having said that, I think there are tons of interesting ways to build on the idea. :)