A Rust 'better GitHub' launched as anti-AI, then dang cut that from the title
Gitdot pitched itself as a faster, open-source GitHub that won't ship AI. Commenters opened its own source, found a dozen CLAUDE.md files, and within hours a moderator had quietly removed 'anti-AI' from the Show HN title.
Gitdot is an Apache-licensed, Rust-built git forge promising instant-loading file previews, stacked diffs as a review primitive, and reproducible Nix-based CI. The minimalist UI won real praise in the thread, and snappy speed.
Then the contradiction landed: the 'we don't do AI' framing collided with the dozen CLAUDE.md files in Gitdot's own repo. The founder conceded the phrasing was a mistake, and moderator dang removed 'anti-AI' from the title.
A Show HN that argued with its own title
Gitdot launched as 'a better GitHub — open-source, written in Rust', and the founders worked the thread live on HN, revisiting design decisions in real time as 154 comments rolled in. By the end, the most notable change wasn't to the product — it was to the title.
What Gitdot actually is
Per the project's FAQ, Gitdot is a git forge pitched at developers who treat code as craft. The concrete claims:
- A high-performance git server written in Rust, with file previews that load instantly on hover.
- Apache-licensed and fully open source, with an explicit 'fork us if we misstep' invitation.
- A roadmap built on stacked diffs as the core review primitive, a Nix-based CI that's reproducible and locally testable, and a promise of no data training or selling.
The minimalist, terminal-flavored UI was the part that genuinely landed:
"The minimal look feels very refreshing, and yet it's not disorienting like many minimal web git UIs … I actually feel like I know how to navigate this thing. Site feels very snappy too …" — TazeTSchnitzel
The 'anti-AI' claim met the repo's own CLAUDE.md files
Gitdot's FAQ said it would not have AI, calling it 'a bandaid for poor design'. The thread pulled that thread until it unraveled. A commenter simply browsed Gitdot's own source — hosted on Gitdot itself — and found the contradiction:
"I'm honestly lost for words over how you can write something like this while the web app's source code is clearly heavily integrated with Claude Code." — a commenter who opened Gitdot's own repo
The repo carries a dozen CLAUDE.md agent files. A high-karma author put the logical bind plainly:
"a team like yours can either: * Use LLMs, in which case you aren't "anti-AI". … * Not use LLMs and promise never to do so. … you hurt yourself by engaging in something very old: disingenuity." — the author of *Knative in Action*
The founder conceded directly, clarifying that 'anti-AI' really meant no AI copilot in the app and no training on your data — and that the title's wording was a mistake:
"i think i've made the mistake of phrasing here with the "anti", i'll revisit it. edit: dang has removed it, apologies for the confusion i caused, that was my mistake." — the Gitdot founder in the thread
Moderator dang confirmed he'd taken 'anti-AI' out of the title 'since it was causing confusion'. Another commenter offered the cleaner label the founder was reaching for: this isn't anti-AI, it's anti-enshittification.
The other two challenges: Forgejo and a trademark
Beyond the AI tangle, two substantive critiques recurred. First, the differentiation question — why this and not Forgejo or sourcehut? The founder's answer was design plus specific bets (stacked diffs, reproducible CI), conceding Forgejo is 'a pretty great piece of software'. A theoretical-CS professor put the risk bluntly: someone could just copy the UI onto Forgejo and skip the hard parts.
Second, a commenter argued the name itself violates Git's trademark policy, which bars using 'Git' as a syllable in a new portmanteau (GitHub and GitLab are grandfathered exceptions). That sparked a genuine 'is that even enforceable?' debate, with a high-karma commenter cautioning against reading 'programmers aren't good lawyers' as proof the thing was 'vibe-launched'.
So-what: a promising forge that picked the wrong fight
The product underneath is real — fast, open, Rust, with opinionated review primitives people actually want. What the thread exposed is a positioning own-goal: leading with a values claim ('anti-AI') the codebase contradicts, in a launch where the audience can read your source in two clicks. The founders fixed it gracefully and on the spot, which is its own signal. But the lesson the thread hammered is older than Rust: on a Show HN, your repo is part of your pitch, and the crowd will diff your principles against your CLAUDE.md files before they ever clone you.
Get breakdowns like this every day
The top Hacker News stories, unpacked — by email, Telegram, or RSS.
"Why is HN so anti-AI?" — and the moderator's one-line answer that reframes it
A user asked Hacker News why its crowd is so hostile to AI. The site's moderator replied that with every A-versus-B split, the A team thinks HN is anti-A and the B team thinks it's anti-B — and pointed, as proof, at a 1,100-comment thread on this very site gushing about AI.