Phase 1: The Pixurl Project & The Library of Babel

Every complex architecture begins with a strict constraint. For our engine, the foundational concepts began with a wild, borderline impossible ambition: to build a decentralized "Library of Babel" for pixel art. The core constraint was strict and elegant—generate images deterministically purely from URL parameters without relying on a backend database.

The Iterations

The project initially explored ambitious 32x32 grids before refining down to a minimalist 4x4 grid. This forced constraint drove creative solutions for managing visual state. When you cannot store thousands of pixels in a database, every single byte of data matters.

The Breakthrough Algorithm

The breakthrough wasn't just in drawing pixels; it was in how we encoded them. We engineered a hybrid encoding algorithm utilizing bit-packing and flag characters. Instead of just storing raw color values, the system allowed for recursive image references—essentially creating pointers to other image IDs directly inside the URL string.

This extreme compression technique turned URLs into unique, claimable "pixel artifacts." It meant an entire sprite sheet could be represented by a string of text that anyone could copy, paste, and render instantly without any server calls.

This direct mapping of state-to-URL was the seed that eventually paved the way for encapsulating the logic into the <pix-url> web component, and later, for the base64 URL-loading system currently used in our Web IDE.