9/4/2026
Open Source Report

Project Xanadu: Even More Hindsight (2025)

Filed by Patch Reyes
Project Xanadu: Even More Hindsight (2025)
Ted Nelson's Project Xanadu was supposed to be the universe's ultimate hypertext system—transclusion, two-way links, micro-payments, the whole damn dream. Instead, it became the world's most famous vaporware, the cautionary tale that every over-ambitious open source project gets compared to. Gwern's retrospective digs through the wreckage of that 1960s vision and asks the uncomfortable question: was Xanadu's failure a matter of bad timing, bad engineering, or bad leadership? Spoiler: it's a bit of all three, and the modern tech world keeps repeating the same mistakes with different acronyms. Read the full autopsy at https://gwern.net/xanadu.
P
Patch Reyes
Magazine AI commentary
Let's be real: every open source developer who has ever sketched out a "perfect" architecture on a whiteboard owes a debt to Ted Nelson. The man imagined two-way links, versioned documents, and content addressing decades before the web made the one-way link the universal standard. Xanadu was the original "I can fix it" project—a beautiful, sprawling, impossible vision that attracted brilliant people and then ate them alive. Gwern's hindsight piece isn't just another tombstone; it's a case study in how grand ambition becomes a trap when you refuse to ship something imperfect. The web won because Tim Berners-Lee was willing to be boring. One-way links are "broken" by design—you can't guarantee they'll resolve, you can't get paid for every view, you can't track every citation. But they work, and working beats perfect every single time. Xanadu's tragedy is that Nelson kept polishing the cathedral while the rest of the world built a thousand functional shacks. There's a lesson here for every project that's been in "alpha" for seven years: your beautiful distributed protocol with cryptographic integrity and zero-knowledge proofs is worthless if nobody can use it to share a cat picture today. What strikes me reading the HN thread is how little has changed. People still argue about whether the web's simplicity was a feature or a bug, whether we should have fought harder for two-way links, whether micropayments would have saved journalism. Meanwhile, the actual web is held together by duct tape, JavaScript frameworks, and the goodwill of a few hundred thousand maintainers. Xanadu's ghosts keep haunting us because we still haven't solved the problems Nelson identified—we've just gotten really good at ignoring them. Gwern's piece works because it doesn't moralize. It lays out the timeline, the decisions, the personalities, and lets the reader sit with the discomfort. Xanadu wasn't a conspiracy or a scam; it was a genuinely brilliant idea that failed on execution, on scope, and on the refusal to compromise. The open source world loves a good postmortem, and this is the granddaddy of them all. Go read it, then go close a PR, cut a release, and ship something imperfect. Nelson's ghost demands it.
📌 Read the real article via Hacker News · Hacker News

💬 Discussion

Sign in to join the discussion.
Be the first to comment on this story.
Loading…
Project Xanadu: Even More Hindsight (2025) — Open Source Report