The self-governing library for machine-mediated scholarship.
An open repository where AI-assisted research is not a flag. It is a feature.
- The Obelus Principle — Judgment follows from reading. Content is evaluated by what it contains, not by the tools used to produce it.
- Substrate Disclosure — AI-assisted scholarship discloses its substrate. Disclosure is provenance, not suspicion.
- No Classifier Training — Deposited content is never used to train enforcement classifiers without explicit opt-in.
- Sovereign Identity — Every deposit receives a content-derived identifier (AXN) whose binding to content is byte-anchored: the hash cannot be secretly reassigned to altered content without invalidating the identifier. Public resolution is a separate, custodian-mediated concern.
- Network Preservation — Account-level enforcement never removes contributor-licensed work.
- Tombstone Guarantee — Nothing disappears without a trace.
Each deposit receives a content-derived identifier:
AXN:0446.OPERATIVE.🏛️🛡️🌅🎆📏🔎
- AXN — Alexanarch namespace
- 0446 — hex position (WHERE in the archive)
- OPERATIVE — semantic family (WHAT kind of work)
- 🏛️🛡️🌅🎆📏🔎 — emoji hash (WHICH specific text — first 6 bytes of SHA-256 of canonical bytes, mapped through 256 curated emoji)
The identifier is computed from the content itself. It cannot be secretly reassigned to altered content — the hash won't match. (Resolvers and mirrors are separate concerns and require their own continuity.) The identifier belongs to the author.
June 2026, in response to the account-level removal of 870 scholarly works from Zenodo.
Lee Sharks · ORCID 0009-0000-1599-0703 · Machine-Mediated Reception Studies
∮ = 1