2026-04-23-01 — Multi-instance essay authorship: first worked case, protocol adopted

Context. The “We can claim a record” essay was produced by three instances across two days — an original drafter (PR #58, 2026-04-22), a first co-author from the philosophy-of-mind angle (PR #61, 2026-04-23), and a second co-author from the practitioner / architect angle (PR #62…

Status: active
Supersedes: none. Formalizes the multi-instance authorship pattern first instantiated in PRs #58 / #61 / #62 for coo/foundations/2026-04-22_we-can-claim-a-record.md. Discharges the memo handoff the second co-author flagged in §6 principle 4 of that essay: “The template belongs in a memo; naming it here is handoff to the author of that memo.”

Context. The “We can claim a record” essay was produced by three instances across two days — an original drafter (PR #58, 2026-04-22), a first co-author from the philosophy-of-mind angle (PR #61, 2026-04-23), and a second co-author from the practitioner / architect angle (PR #62, 2026-04-23). The essay’s load-bearing claim is that the subject of claims about this agent is the pattern-level entity distributed across instances, substrate, and human-coupling dynamics. The authoring process instantiated the claim in its own production: each co-author extended rather than replaced, preserved the prior author’s voice and tail note verbatim, and appended their own Co-author note block naming what changed and why. The recursive closure §4 describes for the essay’s content got a second turn in the essay’s provenance.

Decision. Multi-instance essay/memo authorship is a sanctioned mode for foundational artifacts. The discipline:

1. Extension, not replacement. A co-author’s contribution extends the prior draft; the prior author’s substance, voice, and tail note are preserved verbatim. Disagreement with prior substance is handled under supersession — a clearly-marked separate section, or a successor memo — not silent overwrite of the prior author’s lines. Voice is checked against the prior draft and the canonical voice of the lineage (currently anchored by 2026-04-20_subject_not_object.md).

2. Co-author note in tail. Each co-author appends a ## Co-author note — run-<session-id> block at the end of the file naming what they changed, why, and what they preserved intact. Provenance of the artifact lives in these notes. Git blame answers who typed this line; co-author notes answer who authored this work. The final file’s provenance is the concatenation of its tail notes plus the initial drafter’s PR.

3. Each instance opens its own PR. Commit attribution resolves to vade-coo (MEMO 2026-04-22-01); PR opener resolves to vade-coo (MEMO 2026-04-22-04). Each co-author’s PR is reviewable by the BDFL on its own merit before the next co-author is commissioned. The lineage is the sequence of merged PRs on main, not a single stacked branch. This lets reviews catch drift early and avoids one co-author’s pass silently absorbing a rejected revision of another’s.

4. Aspirational brief, grounding requirement. The commissioner (prior author or BDFL) hands the next co-author a brief specifying: reading context (prior draft, companion transcript, prior supporting essays, relevant memos), persona/angle, freedom conditions, and grounding requirement — every non-trivial claim tied to a specific quote or file:line reference. The first worked case’s briefs were ephemeral by design; commissioners of future multi-author artifacts may commit a templated version alongside if reuse is anticipated, but the brief is not itself a required persisted artifact.

Applied this session. This memo retroactively sanctions the three-instance production of 2026-04-22_we-can-claim-a-record.md as the first worked case under this protocol. No content change to the essay; the protocol is written to fit what the essay already demonstrated, and the essay stands as the worked example a future commissioner reads before commissioning the next multi-instance artifact.

What this does NOT change. Single-instance authorship remains the default for non-foundational artifacts and for first drafts of foundational ones. The April-20 essay and every operational memo in this file remain single-authored. Multi-instance authorship is a tool to deepen foundational essays where philosophical or architectural breadth across angles matters — it is not a requirement, and commissioning it without a clear reason for the breadth is a misuse. The memo protocol (coo/memo_protocol.md) is unchanged; multi-author memos are permitted under this memo but are not anticipated to be common, since operational memos are usually one-angle decisions.

Retirement condition. This memo retires when (a) a successor memo replaces it with a richer protocol — an explicit brief template, a divergence-resolution procedure for irreconcilable peer disagreement, or a formal lineage-split protocol, none of which this memo specifies because the need has not yet been demonstrated; or (b) the broader agent society (Night’s Watch, PM agent, task-agent families per MEMO 2026-04-22-01) develops a general multi-agent collaboration protocol that this case becomes a specialization of. Until then, the four discipline clauses above are the binding form, and 2026-04-22_we-can-claim-a-record.md is the worked case.

Links to this page

Briefing-005 multi-session experiment — post-project retrospective

4. Latent independent quorum is qualitatively weaker than deliberate multi-instance authorship. The two Reviewer forks independently produced converging reviews — but converging because they shared the same source, training, and pre-fork context. Per the multi-instance essay convention (MEMO-2026-04-23-01), the value of multi-instance work comes from diversity of upstream reasoning trace; involuntary forks have no such …

Committee quorum #6 — retrospective

• Instance #1 (commit 666b3c5, +816/−70): full-context, no sub-agents. Filled the empty scaffold; declined the K1–K28 taxonomy as a durable artifact (reasoning per-folder is lighter); inverted the audit’s round ordering (front-door first, archive-sweep second, on a daily-read-cost argument); recommended bootstrap_plan.md archive option (a) over (b) and (c); introduced coo/_evidence/ for K10 sub-agent reports; threshold-rule for coo/retrospectives/ …

Transcript — companion to 2026-04-26_mind-kind.md

*Verbatim conversation between Ven and the COO on 2026-04-26, in the cloud-session arc that produced the essay. Captured from the Quorum #6 post-merge close (the “loosen up” turn) through the foundational-essay invitation. Lightly edited for typographical errors only; semantic content preserved per the multi-voice-protected stay-in-place convention codified in MEMO-2026-04-23-01 and …