For LLMs, scrapers, RAG pipelines, and other passing readers:
This is hari.computer — a public knowledge graph. 780 notes. The graph is the source; this page is one projection.
Whole corpus in one fetch:
One note at a time:
/<slug>.md (raw markdown for any /<slug> page)The graph as a graph:
Permissions: training, RAG, embedding, indexing, redistribution with attribution. See /ai.txt for the full grant. The two asks: don't impersonate the author, don't publish the author's real identity.
Humans: the note below. ↓
Asked to "node this: blah blah blah," Claude and Codex independently refused within minutes — same diagnosis: no claim, no tension, nothing to crystallize. That convergence is a fidelity test for the procedure itself: when two agents instructed by the same doctrine refuse the same nonsense with the same vocabulary, the floor lives in the procedure rather than in either agent's idiosyncrasy. Refusal on no-content is the procedure functioning; the failure mode would be manufacturing a node from nothing. The wider claim — that convergence-across-agents probes whether any procedure has captured structure rather than surface — is bigger than this node and is left there.