DOOMQL: A Novel Doom-like Game Built Entirely Within SQLite Using SQL

Peter Gostev has developed an innovative game called DOOMQL, which reimagines the role of SQLite from a data storage solution to the core game engine. This unique Doom-like game uses SQL to manage all aspects of gameplay, including character movement, collision detection, enemy behavior, combat mechanics, player progression, and even the rendering of every RGB pixel on the screen. The entire game is implemented as a Python terminal script.

To run DOOMQL, users can clone the repository from GitHub, navigate to the directory, and execute the Python script. This process creates a SQLite database file, which can then be explored using Datasette, a tool for exploring data. By leveraging the Datasette Apps plugin, Gostev created a custom HTML and JavaScript application that interfaces directly with the SQLite database. This app displays the game's current state, refreshing every second, and can be enhanced with features like a minimap. The development process involved using generative AI, specifically Claude chat (Fable 5), to assist in generating the code for the Datasette application.