Linus Torvalds published the first public release of Git on April 7, 2005, describing it as a "fast, scalable, distributed revision control system" with an unusually rich command set. The tool emerged from Linux kernel development needs after the BitKeeper license dispute forced Torvalds to build a replacement in roughly two weeks. Licensed under GPLv2, Git provided both high-level porcelain commands and full access to plumbing internals — a design choice that enabled scriptable workflows and third-party tooling from day one.

The initial README reveals Torvalds' characteristic humor in naming: "git" was a "random three-letter combination" meaning everything from "stupid content tracker" to "goddamn idiotic truckload of sh*t" depending on mood. Documentation pointed users to gittutorial and giteveryday for onboarding, while development centered on the [email protected] mailing list with patch submissions governed by explicit coding guidelines.

Security issues were routed to a separate [email protected] list, and the maintainer issued regular "What's cooking" status reports. This distributed model — no central server required, every clone a full repository — represented a fundamental architectural shift from CVS and Subversion's centralized approach.

Git's 2005 debut solved the kernel's scaling problems through content-addressable storage and cheap branching. That same architecture now underpins virtually all modern software collaboration, from GitHub to CI/CD pipelines, proving the "stupid content tracker" became the industry standard by refusing to be clever about the wrong things.