Home > Positioning > Subjects > Git

Git

Git is a distributed version-control system: a tool for recording the history of a set of files and coordinating work on them across many people and machines. Its defining choices — content-addressed snapshots, a complete copy of history in every clone, and integrity guaranteed by hashing — set it apart from the centralised systems that preceded it, and it has since become the near-universal standard for source control.

Three ideas organise the design:

• Snapshots, not differences. Where earlier systems (RCS, SCCS, CVS, Subversion) stored history as a series of per-file deltas, Git records the state of the whole tree at each commit as a snapshot, sharing unchanged content rather than recomputing diffs.
• Distributed, not central. Every clone is a full repository with the complete history; there is no privileged central copy. Committing, branching, and inspecting history happen locally; exchange with others is a separate, explicit step.
• Integrity by content-addressing. Every object is named by the cryptographic hash of its content, so the identity of any version depends on the entire history leading to it; corruption or tampering changes the name.

Origin

Git was written by Linus Torvalds in April 2005. The Linux kernel project had used the proprietary BitKeeper system under a free-of-charge licence; when that licence was withdrawn in 2005 — after Andrew Tridgell reverse-engineered BitKeeper’s protocol, prompting its owner Larry McVoy to revoke it — the kernel needed a replacement, and no free alternative met Torvalds’ requirements for speed and a genuinely distributed model. He began Git on 3 April 2005; it was self-hosting within days, managed the kernel’s 2.6.12 release that June, and reached version 1.0 that December. Torvalds handed maintainership to Junio Hamano in July 2005, who has maintained it ever since.

On the name, Torvalds: “I’m an egotistical bastard, and I name all my projects after myself. First ‘Linux’, now ‘git’.” The repository’s own description reads “the stupid content tracker.”

Pages

• The object model — the content-addressable store and its four object types (blob, tree, commit, tag), the index, and the move from SHA-1 to SHA-256.
• Refs and history — branches, HEAD, and tags as pointers; the commit graph; branching, merging, and conflict; rewriting history and the reflog.
• The distributed model — every clone a complete repository, remotes and exchange, and why a central server is a convention rather than a requirement.
• Ecosystem and adoption — Git’s dominance, the hosting platforms built on it, independent implementations, and its reach beyond code.

Persons

• Linus Torvalds — creator of Git and of the Linux kernel.

Sources

• Git and the Pro Git book — the canonical documentation.
• Git — Wikipedia, for overview and history.

---

See also: Linus Torvalds