An engineer documented their journey of creating a version control system from scratch, inspired by the desire to understand the inner workings of Git. The project, dubbed "tvc" (Tony's Version Control), focused on core functionalities like hashing, compression, and commit creation. This allowed them to grasp how files, diffs, and commits are structured under the hood.

The system utilizes SHA-256 for hashing and zstd for compression, diverging from Git's SHA-1 and zlib for compatibility. The implementation, written in Rust, included commands for listing files, hashing content, compressing/decompressing objects, and generating tree and commit objects. This approach provided a hands-on learning experience about content-addressable storage.

Commit creation involved storing object types, filesystem snapshots, parent commits, author information, and commit messages. They opted to simplify by not differentiating between author and committer. The project's challenging aspect was parsing object formats. The author suggests using a well-defined format like YAML or JSON for future iterations, showcasing a practical learning experience.

This project underscores the complexity of building a version control system. Despite the rise of cloud-based version control tools, understanding the underlying principles remains valuable for developers. The author's code is available on GitHub, offering a practical example of the system's architecture and design choices. It serves as a great learning resource for those interested in source control.