Ghostty leaving GitHub landed on HN as a warning about trust, not just hosting
Original: Ghostty is leaving GitHub View original →
Hacker News did not read Mitchell Hashimoto's Ghostty post as a mere hosting migration. The emotional hook mattered, but the thread stayed alive because people read the move as a trust signal. In the blog post announcing the decision, Hashimoto says he is GitHub user 1299, joined in Feb 2008, and even says he cried while writing the piece. That history turned the move into more than a project logistics update. If someone with that level of loyalty decides to pull a flagship open-source project away, users naturally treat it as a warning.
The interesting part is what HN argued about next. Some commenters treated the story as proof that GitHub reliability and product focus have degraded past the point where affection can compensate. Others pushed back on the easy villains. A GitHub employee in the thread argued that the deeper cause is scale and the shock of agentic coding, not just Microsoft ownership or a sudden AI rot. That response did not settle anything, but it did sharpen the disagreement. Is GitHub slipping because the product is distracted, or because the whole platform is trying to absorb a new development paradigm while usage explodes?
- Ghostty is leaving GitHub, but the destination has not been finalized publicly yet
- The post frames the decision as emotionally difficult, not as a casual forge experiment
- HN quickly moved from one project move to a larger argument about GitHub reliability
- Alternatives such as Forgejo, Codeberg, and self-hosted flows kept coming up in discussion
That is why the thread had more weight than a normal maintainer diary entry. Community discussion noted that the social layer and the code-hosting layer may need to come apart if developers want more flexibility without losing discovery and collaboration. Other commenters were more blunt: staying on a service that keeps disappointing users is not a path to improvement, it is just inertia with better branding. Either way, the conversation was bigger than Ghostty.
The real story here is not that one terminal emulator project wants a different forge. It is that a long-running GitHub loyalist made the emotional cost public, and HN instantly mapped that feeling onto its own backlog of reliability frustrations. Once that happens, every outage and every slow page stops looking isolated. They start to look like proof. Source links: Hacker News thread, Mitchell Hashimoto’s post.
Related Articles
HN focused less on telemetry as an idea and more on whether opt-out controls work when gh runs inside CI, servers, and automation.
HN reacted because fake stars are no longer just platform spam; they distort how AI and LLM repos look credible. The thread converged on a practical answer: read commits, issues, code, and real usage instead of treating stars as proof.
Why it matters: model launches become more consequential when they land in tools developers already use every day. GitHub says early testing found GPT-5.5 strongest on complex multi-step coding tasks, and the rollout ships with a 7.5x premium request multiplier.
Comments (0)
No comments yet. Be the first to comment!