A Rust dev bets agents will retire Git, ships Oak, offers an e-ink display

Oak replaces Git's commit model with content-addressing and a branch-per-session unit, so a cloud agent can edit any repo in seconds with no full clone. The launch's headline incentive is a personalized e-ink display for the first 100 paying users.

In brief

Its trick is content-addressed lazy mounts: instead of describing each commit, you describe a whole session, and agents skip the clone entirely. The author has been building non-Git VCS since before the agent hype made it fashionable. But the launch shipped a reward gadget and zero benchmarks against Git, and at least one Git lover quietly agrees the incumbent is vulnerable anyway.

Oak is an open-source, agent-first version control system in public beta (v0.99.0), written in Rust and built around content-addressing rather than Git's commit graph.

Describe the session, not the commit

Oak version control system
Oak version control system

The atomic unit is branch-per-session with a branch description in place of per-commit messages. Content-addressed lazy mounts let agents, local or cloud, edit any repo in seconds without a full clone. Two crates carry it: oakvcs-core (BLAKE3 hashing, content-defined chunking, a Blob/Manifest/Commit/Tree model, optional SQLite and git-compatible backends) and oakvcs-cli (machine-readable JSON, documented exit codes, non-interactive safety). There's no separate LFS; monorepo support is planned. The pitch is faster agent workloads and no worktree wrangling for parallel tasks.

The problem it solves, and the launch that undersold it

The author, zdgeier, gave the clearest statement of the actual problem:

"'download everything' means that you don't have to do a full or partial clone of the repo to make a change. You can imagine agents running in the cloud need to spin up and access the repo for every task, this can be slow for large repos… worktrees can be a pain to manage…" — zdgeier, the project's author

That clarity didn't reach the launch copy. A reader tallied what was on offer:

"Lots of self-promotion, but no concrete comparisons where this tool does a better job than git… > For the first 100 users that subscribe to a paid plan I will send you a personalized e-ink display I don't understand anyone who feels incentivized by this. Brogrammer 2.0 is weird." — a commenter who read the launch offer

The "why not just build on Git?" objection got a reply rooted in the author's own history, which predates the agent wave entirely:

"Haha I wish, but I've been working on VCS's separate from git for a while now. … I've wondered for years before agents if something could be made using something different, rather than building something on top." — zdgeier, the project's author

Even a Git partisan conceded the ground is softer than it looks:

"'Git is forever' Many things were forever until they suddenly died, but I think this is especially true for git… I also think git is an ill-fit for the majority of modern commercial software projects and there will be a breaking point…" — Ludwig Weinzierl, a Rust/ML/security engineer

The debate in the thread circles one bet: that the next breaking point for Git isn't a better diff or a friendlier CLI, but a workload Git never anticipated — fleets of agents that want to touch a repo without ever owning a copy of it.

Get breakdowns like this every day

The top Hacker News stories, unpacked — by email, Telegram, or RSS.

Next story · 29 of 30 Read more
🚀Tech

The best sync engine you don't manage runs in one Chromium-only browser

Steve Harrison hands a web page an ongoing read/write handle to a folder on your disk and builds a Lightroom-style photo manager that never leaves the browser. Mozilla and Apple have both formally vowed never to ship the API that makes it work.

★ 106 · 💬 97 · steveharrison.dev · 3 min