From CVS to Git: thirty years of source control, lived from inside

13 hours ago

The article provides a firsthand account of the evolution of version control systems from pre-VCS methods to Git.
Early practices included using zip files with date stamps, network shares with manual versioning, and lock-based systems like SCCS and RCS.
CVS introduced concurrent edits and server-based repositories but suffered from issues like non-atomic commits and poor handling of branches and renames.
Visual SourceSafe was used in Microsoft shops, but was prone to database corruption and enforced a locking discipline that hindered collaboration.
Subversion (SVN) improved on CVS with atomic commits and better branching, but remained centralized, requiring a network connection for commits.
Git was created by Linus Torvalds in April 2005 after BitKeeper revoked its free license for the Linux kernel, focusing on distributed version control, speed, and cheap branching.
GitHub, launched in 2008, popularized Git through features like pull requests and social coding, making it the de facto standard for version control.
Despite alternatives like Mercurial, GitLab, and newer systems such as Sapling and Jujutsu, Git remains dominant due to its robust design and network effects.
The author reflects on personal experiences with each system, noting that Git's failure modes are often user errors rather than design flaws.
Git is unlikely to be replaced in the near future, as its core distributed model and content-addressed storage have proven effective for over two decades.

Hasty Briefsbeta