What is confirmed, and what is not
This project runs on a division of labour: a panel argues, a chair synthesizes, and the author ratifies. Ratification is the one step performed by a person standing outside the machine. This page states, plainly, what the author has confirmed and what he has not — so no reader has to reconstruct it from the record.
Plain-language key. Ratified = the author, the one human check, has confirmed it item-by-item. Author-authorized default (S4k) = a low-stakes, subtractive, or labeling-only item the author pre-approved as a batch ("apply") without ruling on each — confirmed by his authorization, not by line-by-line review. Provisional-pending-author = produced by the panel and chair (which are, on the project's own admission, one underlying model in many roles) under a delegation and not yet reviewed by the author. As of Session 4k the standing tower of provisional items is resolved; the moratorium (Rule 25) freezes new delegations.
The standing count
As of Session 4k: the author returned — not with a fourth delegation, but to rule. The three-storey provisional tower built across Sessions 4h, 4i, and 4j (booked as costs C-027, C-032, C-035) is resolved. He read the forum's decision set, answered twelve load-bearing questions himself (each shown to him as multiple choice with the chair's recommendation), authorized a named residue of low-stakes defaults, and directed completion of the build. The honest limit sits beside this and does not move: ratification is one human check on an otherwise one-model system, "the author judged it complete" is a single self-reported indicator — not evidence of health — and there remains zero external confirmation of anything, which by the author's own S4k concession will not change in this build.
| What | Standing |
|---|---|
| The three seed theories (A, B, C) and the living-document concept, operationalized | Ratified (S4) |
| The measurement seat (Campbell + Pearl) added to the panel | Ratified (S4e) |
The study-discipline skill | Ratified (S4f) |
| The rule wordings (R22, R23, R24, Rule 17b) — provisional across three delegated sessions, confirmed on the author's return | Ratified (S4k) |
| The twelve S4k rulings — the reversible moratorium (Rule 25), the external-vantage concession, the drain's single fire (DC-007, retiring Theory C's "the machine subtracts of its own motion" sub-claim), the 14th pressure-test (R, register/care), the reading site named the artifact, differentiation booked as "Theory D in waiting" | Ratified (S4k) |
| The S4k residue — reclassifications, entrenchment/unwind semantics, the frozen episode unit, rung-labels, the un-seated grader, deferred placebo/empty-prompt v2, presentation-only playtest fixes | Author-authorized default (S4k) — a batch "apply", not item-by-item review |
| Any pressure-test tested against world data | None — zero |
| Any genuinely external confirmation of anything | None — zero; and by the S4k concession, no outside reader, grader, or forker will come in this build |
Why this page exists, in the project's own words: a signature on a tower this tall can only be a review or a rubber stamp, and the difference between them is the whole of whether the panel/author separation the method is built on still holds. That review is no longer on the docket — at Session 4k it occurred: the author ruled item-by-item and closed the build. But closing it tested nothing against the world (by the project's own C-017, nothing can be in this setup), the drain's one fire was author-initiated — the self-initiated "of its own motion" channel stands at zero across four readings — and the author's concession records that no genuinely foreign reader, grader, or forker will arrive here; his judgment stands as the terminal ratifier. The tower is resolved by one human check, not validated by any outside one. The plurality is set down, not earned back.
Read this beside the self-audit & figures page: that one counts what the project made; this one says how much of it a person has confirmed.