Enforced publication controls

Provenance, rights & correction

Every object in the 1,399-position historical inventory is accounted for here, and most of them are not on the public site. 1,609 unique source objects carry 1,664 ledger records; 460 of those records are published and 1,204 are not. This page explains what decides which, using the ledger itself. Every number below is computed from the same JSON files linked in the table, so you can check any of it without taking a word of it on trust.

One field decides it

A record is published when it names the source it came from, and withheld when it does not. That is the whole rule, and it is worth seeing it operate on two records from the same corpus, both reviewed on 15 August 2026.

Flashcard fc-001 asks what the primary mechanism of action of classical psychedelics is. Its sourceEvidence field reads Nichols, D.E. Pharmacological Reviews 68(2):264-355 (2016), a real review, free to read in full at PubMed Central. It is published. Another flashcard in the same corpus has an empty sourceEvidence field. It is withheld, and the public ledger counts it without naming it. Its claim is not suspect and may well be sound; it is simply not traceable from this ledger to anything, and an untraceable claim is what this project has decided not to ship.

1,199 of the 1,664 ledger records carry no source evidence at all. That number, not any judgement about quality, is why the public library is a quarter of the archive.

Why 1,204 records are withheld

Each withheld record states its own reason. There are five, and they divide as follows.

The second row is the one that shapes the reading experience most. Every quote card in the archive, all 337 of them, is withheld, because a quote card is by construction somebody else’s words, and reproducing those needs a basis attached to that specific quotation rather than a general policy. The timeline is withheld entire for a related reason: a dated assertion about what happened in a given year is a factual claim, and 196 of them arrived without anything saying where the date came from.

What the 460 published records rest on

The published records carry one rights basis: original editorial or bibliographic and educational metadata, with cited third-party sources named rather than reproduced wholesale. In practice that means a published record can tell you that a passage discusses Huxley and Osmond, and can cite Moksha by author, title and year, without setting out Huxley’s sentences.

That distinction is the reason the library can exist at all. Facts and bibliographic particulars are not themselves owned by anyone: the Supreme Court settled the point in Feist Publications v. Rural Telephone Service Co., 499 U.S. 340, decided 27 March 1991, holding that facts are not copyrightable and rejecting the “sweat of the brow” theory that would have protected them. The expression of those facts is a different matter, which is why the quotation corpora sit in the withheld column and the bibliographic ones do not.

Three counts, reconciled

The archive reports three different totals, and the gap between them is a real feature of the material rather than a rounding error. 1,399 is the historical presentation inventory: the number of positions the library occupied when it was catalogued, which counts 55 mini-essays twice because each appears once inside Writing and again as a Mini-Essays item. 1,609 is the count of distinct source objects once those 55 are recognised as the same work in two places. 1,664 is the number of ledger records, which is 1,609 objects plus the 55 duplicate presentation records, each of which is kept and marked with duplicateOf and canonicalWorkId rather than deleted, so the historical count remains reproducible.

Relationship and responsibility

Psychonaut Bookworm is an independent RN Collins editorial prototype. RN Collins is the editorial owner and the responsible publisher for the original material here. Fat Nugs Magazine has no ownership, hosting, publishing, sponsorship, endorsement, revenue or licensing relationship with this edition, and none is claimed.

What was checked on 13 October 2026

Education, not medical advice. Nothing in this library is medical advice or a guide to using any drug.

On 13 October 2026 every citation on the 460 released pages was checked against PubMed, Crossref or the known public record. Journal articles were matched by author, title, journal and year; books, laws and court cases were confirmed against the known record. Wrong journals, years, titles and authors were corrected. Sources that could not be found are now marked “Source not found; under review” on the page, next to the citation and next to any sentence that rests on it. Claims that went beyond their evidence were regraded, including the Rat Park experiments (now described the same way everywhere: Alexander and colleagues, 1978 to 1981, and later attempts with mixed results, including Bozarth and colleagues, 1989, and Petrie, 1996) and the 2020 DMT entity survey (a survey of self-selected people who had already had an encounter, which cannot confirm what the entities are or how common encounters are). Dose amounts and drug product, company and magazine names were removed from the text, and pages with no sources of their own were given sources or labelled “Sources pending review”.

What remains: an independent check of every factual statement in the body text; the organizational, trade and press sources, which were listed but not checked; and claims about 2025 and 2026 events, which change quickly. The full citation ledger, with every original citation, its status and the corrected citation, is CITATIONS-Oct13.md in the open edition of this library. Changes to sentences in the body text were made in the source repository and are recorded in its history, which is not public. Earlier wording of the citations is kept in the ledger.

Publication controls, as shipped

Withholding is enforced in the data rather than announced in a policy: a withheld record is absent from the public data object, not merely flagged inside it. The public ledger files list every published record in full and count withheld records by reason only, without their ids, titles, excerpts or sources, so nothing that is held back can be read through the ledger. Since October 2026 one file, data/releases.json in the source repository, decides what is published: a record appears when its release date has arrived in Honolulu and it is not on hold, and the site rebuilds daily. Each ledger record states its evidence, its rights basis, its publication status and, where it applies, its suppression reason. Every record was reviewed on 15 August 2026, and the manifest carries that date, so the age of this review is visible rather than implied.

Open the machine-readable manifest

Correction, rights, or withdrawal request