# Publication record under Schema Independence Charter §6: founding-period schema decisions taken in concurrent capacities

**Arkaya_Layer1_PublicationRecord_ConflictedDecisions_2026-10-22_v6 · 22 October 2026 · Layer 1 · AKR AIPS**

**Status: APPROVED FOR PUBLICATION. Publication is complete only when the frozen bytes have been deployed and verified through the public record pipeline.** Prepared under `APPROVE R2` (22 October 2026, 10:26 BST) from `OPS_DisclosureQueue_2026-10-22_v19`; publication approved under `APPROVE PUB-7` (23 October 2026, 16:45 BST), for publication on 26 October 2026 as release `publish-007`. Three inclusions were decided on 22 October 2026 (13:03 BST, INC-1 to INC-3): Section 3 rows 7 and 8 and Section 4 item 6 are included. Two more were decided on 23 October 2026 (15:20 BST, INC-4): Section 4 items 7 and 8 are included, and P-D7 is recorded in Section 5 as not included. One more was decided the same day (15:26 BST, INC-5): Section 4 item 9 is included. The remaining inclusions were decided on 23 October 2026 (16:21 to 16:26 BST, INC-6 to INC-17): Section 3 rows 1 to 6 and Section 4 items 1 to 5 are included, and Section 5 stands as drafted. On the same day (16:37 BST, INC-18) the correction of 22 October to the website discovery document and `llms.txt` was treated as carried by Section 3 row 7 and Section 4 item 6, with no separate entry. This record's own inclusion decisions (INC-1 to INC-18) are queued as entries for the next record, and its preparation decisions of 22 and 23 October are queued for it. No inclusion decision and no drafting decision is open. D-DRAFT-1 was decided on 22 October 2026 (11:41 BST): the statements on the external review and the second member are included. D-DRAFT-2 was decided on 22 October 2026 (12:35 BST): Section 3 row 7 states that the discovery document was corrected by publication on 22 October.

**Successor to v5** (`Arkaya_Layer1_PublicationRecord_ConflictedDecisions_2026-10-19_v5`, published 20 October 2026 at `https://record.arkayarisk.com/disclosures/`, release `publish-005`). v3, v4 and v5 stay published and unchanged. v6 does not repeat their entries. It covers the decisions taken since v5 that v5 could not carry, including the publication of GET v6 for comment, and the control failures since. **Its purpose is narrow: to keep the §7.6 disclosure mechanism current. It creates no new governance layer.**

**Publication cures nothing recorded below.**

---

## 1. Why this record exists

Schema Independence Charter v3 §7.6 makes a decision in which the founder's capacities conflict *"a schema-governance act published per Section 6"*. §6 requires that *"Every schema decision, version change, disciplinary-rule application and occupancy-pending null is published to the public version record."* Since 29 September 2026, every founding-period schema decision is disclosed within five business days of its effective date, and earlier where it changes something already relied on externally (D-DISCLOSURE-INTERVAL, disclosed in v4). This is the third publication run under that rule.

## 2. How to read the entries

The conventions are as v4 §2 states them, unchanged:

- **Capacity.** The capacity column names the capacity in which a decision was taken. It does not determine whether the capacities conflicted on that decision.
- **Inclusion.** An entry is included for completeness. Inclusion does not determine that the entry is a schema decision within §6, or a conflicted decision within §7.6.
- **Dates.** *Effective* is when the decision took effect. *Disclosed* is when public disclosure was completed. A later disclosure date does not mean the decision was made on that date.
- **External reliance.** Recorded on every entry. Counterparties are not named.
- **Review routes.** R-§9, R-§5.1, R-V, R-7.4 and R-P are as v3 §2 defines them. R-C is as v4 §2 corrected it: **no solicitor is instructed on these questions, and none is planned.**
- **Identifiers.** On 20 October 2026 two working sessions each issued identifiers AR-20261020-001 to -010. They are distinguished by the suffix (`-zhzg8v`, `-j7bham`), which this record always gives.

## 3. Entries

| Effective | ID | Decision | Capacity | Operative effect now | Does not confer | Review | External reliance | Disclosed |
|---|---|---|---|---|---|---|---|---|
| 19 Oct | AR-20261019-005-zhzg8v (V5INC-1 to V5INC-3) | The inclusions of record v5: it carries AR-20261019-001 (GET v6 preparation) and AR-20261018-001 (pilot preparation) as entries, and its Section 5 as drafted | Founding Assessor, for the Custodian, deciding what a record of his own decisions discloses | Applied: record v5 public since 20 Oct with those inclusions | Any finding that the items not included are not schema decisions | R-P | None | By this publication, 26 Oct 2026 |
| 19 Oct | CDX-SLATE (a) C3; AR-20261020-002-j7bham | Commission the next version of the record pipeline's change for serving the schema, fixing the two blocking defects the final check of the earlier version found, built by an external coding tool and checked by Claude re-running the tests (C3, 19 Oct). On 20 Oct the check was filed: 219 of 219 tests passed, with mutation proofs for both defects (AR-20261020-002-j7bham) | Founding Assessor, for the Custodian, setting who builds and who checks the change that governs how the schema is served | Applied: the change was released on 20 Oct under AR-20261020-008-j7bham and built the release that published GET v6 | Any independent human review of the change; any finding beyond the tests run | R-P | Yes, indirectly: the change serves GET v6 to the implementer that has relied on v5 since 4 Sep | By this publication, 26 Oct 2026 |
| 20 Oct | AR-20261020-002-zhzg8v; AR-20261020-003-zhzg8v; AR-20261020-006-zhzg8v; AR-20261020-007-zhzg8v | Publish record v5: runbook v4 (build 3) adopted first for release `publish-004` and, after that release was stopped (Section 4 item 1), for `publish-005`; publication of record v5 on the decided inclusions: pre-freeze build 8 approved (AR-20261020-003-zhzg8v) and published as build 10, which carries publication-state changes only (AR-20261020-007-zhzg8v), with document date 19 October and publication on 20 October | Founding Assessor, for the Custodian, deciding to publish a record of his own concurrent-capacity decisions | Applied: record v5 public since 20 Oct (release `publish-005`) | Any finding on the record's completeness beyond what it states | R-P | None | By this publication, 26 Oct 2026 |
| 20 Oct | AR-20261020-005-j7bham (GETV6-4) | The rights statement GET v6 carries when published for comment: "© David McKibbin 2026. Governance Evidence Taxonomy v6 is published for comment. All rights reserved; no licence to implement or redistribute is granted by this publication. Ownership is intended to pass to Arkaya Risk (Holdings) Ltd and thereafter to an independent custodian under the Schema Independence Charter; a licence will accompany the ratified release." Options put and not taken: settle the public licence first; defer until the holding company owns and publishes | The founder, as holder of the rights in his own name, fixing the terms on which the schema is published by the company he controls | Applied in different wording. The GET record's Reason field serves: "Correction-only editorial revision of normative version 1.0, published as a public draft for comment. © David McKibbin 2026. All rights reserved; no licence to implement or redistribute is granted by this publication. Ownership is intended to pass to Arkaya Risk (Holdings) Ltd and thereafter to an independent custodian under the Schema Independence Charter; a licence will accompany the ratified release." The rights terms are the same as those fixed. The sentence "Governance Evidence Taxonomy v6 is published for comment." is not served; the comment status is stated in the Status field. The served wording was authorised by `ACCEPT GETV6-7` (DM, 21:46 BST, 20 October; row 6), which fixed the rights statement in the Reason field "as rendered in the publish-006 rehearsal": the rehearsal registry (`records.py`, SHA-256 `b70ab2fbcdbaa9ee2ebc5d3e7f48fc6707fd7a69d6e25b2160e7d9ef99d3c5cd`) is the registry frozen at `release-007`, and its Reason text is the text served. Established by the independent check of build 5 on 22 October, which relied on the founder's verbatim quotation of the typed decision and verified the registry and served bytes itself | Any licence; any statement of title beyond the founder's own; any vesting in the holding company or a custodian | R-§9; R-P | Yes: it states, for the first time in public, the terms on which an implementer may use v6 (none beyond reading and comment) | By this publication, 26 Oct 2026 |
| 20 Oct | AR-20261020-006-j7bham | How the founder's rights are to pass to the holding company is gated: with no legal or tax advice taken at this pre-revenue stage, the transfer proceeds on recorded founder judgement supported by a dated market-value assessment of the specific rights, using development cost as one input, with the earlier valuation preserved and departures explained | The founder, as seller of his own rights and as director of the intended buyer | Gating only. No transfer is executed; the transfer route itself remains undecided (Section 6 item 3) | Any transfer, valuation, consideration or tax treatment; any authority beyond what company law requires | R-§5.1; R-C (none instructed) | None | By this publication, 26 Oct 2026 |
| 20 Oct | AR-20261020-008-j7bham; GETV6-7; stop point 2 of `publish-006` | Release the pipeline change (row 2 above) and publish GET editorial revision v6 build 3 (normative 1.0.1, 28,325 bytes, SHA-256 `c1e1d4f74a5f892d09766235a474046f95f06c5f39f308b81336259ce79c8ae5`) for comment, served at `/get/1.0.1-public-draft-v6/`, with `/get/1.0` not served, the publication notice in the GET record's Status field and the rights statement in its Reason field (GETV6-7). Published by the company with the author's permission. The founder recorded his judgement that the permission, being gratuitous and in the company's favour, gives rise to no conflict requiring members' authorisation (Companies Act 2006 s177(6)(a)). An external review recommended members' authorisation for this step; the founder took a different view. The second member was not advised of this publication decision. Stop point 2 fixed publication for 20 October 2026 (UTC) and adopted runbook v4 build 3 for this release | The founder, as author and rights holder granting the permission, and as director of the company relying on it; Founding Assessor, for the Custodian, publishing the schema | Applied: GET v6 public for comment at `https://record.arkayarisk.com/get/` since 20 Oct (release `publish-006`); comment period to 10 Nov 2026. v5 remains the candidate | Approved status; candidate status for v6; any licence; any change to the identifier; any finding on the s177 question | R-§9; R-P; R-C (none instructed) | Yes: v6 corrects v5, on which an implementer has relied since 4 Sep | By this publication, 26 Oct 2026 |
| 20 Oct | AR-20261020-001-zhzg8v (WEBPUB-1) | Publication on the company website of a corrected machine-readable discovery document and `llms.txt`. The discovery document's statements on GET were corrected: "no licence condition" removed; licence terms stated as not settled; the current candidate stated as editorial revision v5 (normative 1.0); v6 stated as not published (true at the time); Charter effectiveness stated as not settled | The founder, as director of the company, correcting public statements about the schema | Applied: published 20 Oct. Its statement that v6 "has not been published" became untrue the same evening and was served until 22 October, when it was corrected by publication at 12:29 BST (Section 4 item 6). | Any licence or Charter effectiveness | R-P | Possible: the earlier "no licence condition" text may have been read by implementers before correction | By this publication, 26 Oct 2026 |
| 22 Oct | GETV6-COMMENT | A comment register for GET v6 is opened. Comments are recorded verbatim, checked against the published baseline, and given a recorded disposition at the close of the period. Comments received under a confidentiality agreement are not published, and their source is not named, without the commenter's consent. No change to v6 is authorised by the register. Three comments are registered from one implementer | Founding Assessor, for the Custodian, setting how comment on the schema is handled | Applied: register open; dispositions due after 10 Nov | Any change to GET v6; any disposition | R-P | Yes: the implementer that commented relies on how its comments are handled | By this publication, 26 Oct 2026 |

## 4. Corrections and control failures (clause 5)

**This section is not housekeeping.** It records nine places where the publication and disclosure controls did not work as stated.

1. **Release `publish-004` was published before its checks and before its production decision** (Q-3). On 20 October 2026 its deploy was uploaded one folder level too high and went to production at 14:35:50Z, before the runbook's checks (C6, C7) and production decision (C8). The public record returned 404 (not found) at its canonical addresses for about six minutes. The founder restored the previous deploy at about 14:41Z under the runbook's restoration procedure, and live verification of the previous release then passed 103 of 103 checks. The cause was the working session's deployment instruction. The release was restarted as `publish-005`.
2. **Acts taken before the step or authority the runbook places first, `publish-005`** (Q-4). The freeze was pushed to the public repository (16:02 BST) before the filing step the runbook places before it; the filing completed at 16:04 BST and the frozen bytes were those accepted. The publication patch was pushed (16:36 BST) before the typed authority naming it, which arrived at 16:37 BST and matched every value. Nothing published differs from what was approved.
3. **The same, at the production step of `publish-006`** (Q-6). GET v6 was published to production at about 22:49 BST on 20 October; the typed production approval (AR-20261020-010-j7bham) arrived at 22:50 BST and was recorded as ratifying. Every value in it matched what was published. This is the pattern of item 2, repeated.
4. **The release tag of `publish-006` was placed before the freeze** (Q-7). The runbook places the release tag on the frozen release. Tag `release-006` was placed on the pipeline code commit before any freeze, following the session's outline rather than the runbook; under the runbook's rule that tags stand, the published release carries `release-007`. In the same run, two mistaken uploads left two unpublished preview deploys in the hosting history, and the first freeze commit omitted a file and was corrected before anything was pushed. Neither affected what is public.
5. **The disclosure queue was not maintained** for the decisions of 20 October taken in the second working session (Q-5). The queue version filed that day took its rows from one session only. The decisions in rows 4 to 6 of Section 3 were recorded as "queued for disclosure" in their decision record and listed as due at a pre-publication check, but were not added to the queue until 22 October. **This is the fourth such lapse.** v4 §4 item 2 and v5 §4 item 3 record the first three.
6. **A public statement about the schema was left untrue after a later decision.** The company website's machine-readable discovery document, corrected and published at 14:40 BST on 20 October (Section 3 row 7), states that GET v6 "is prepared as a public draft but has not been published". GET v6 was published for comment at about 22:49 BST the same day. The discovery document was not updated, and on the morning of 22 October still served that statement, as did the website's `llms.txt`. No step in the publication of v6 checked other public statements about the schema. Both were corrected and published at 12:29 BST on 22 October (website change 23, merge commit `9f5bc4c`): they now state v6 as published for comment on 20 October, not approved and not a candidate, and the discovery document gives its address and SHA-256. The failure stands for the two days the statement was served.
7. **Working material for `publish-006` was held only in a session's working copy and was lost** (Q-9). On 20 October 2026 the working session held two things only in its own working copy: the release's publication log draft, and the change made by its evidence-tag step (D7). Neither was filed or committed, and that working copy no longer exists. The log draft was reproduced byte for byte on 22 October and checked against both hashes the session had reported. The evidence-tag step was re-run on 22 October under `APPROVE P-D7`, so the public record carries the time of the re-run, not of the original. The live checks of 20 and 23 October found what was public unchanged; the repository's evidence trail was incomplete from 20 to 22 October.
8. **The post-publication checks for `publish-006` ran late** (Q-10). The runbook places both the preservation and baseline step (Part E) and the next-day checks (Part F) on 21 October, the day after publication. Part E ran on 22 October. Part F ran on 23 October, after the time the founder had set for it. The live verification on 23 October passed 156 of 156 checks and found every served object byte-identical to the 20 October evidence.
9. **Release `publish-005` was not closed, and its next-day tag name was reused** (Q-11). The `publish-005` run, which published record v5 on 20 October, did not carry out its next-day checks (Part F). No next-day verification, Charter comparison or Software Heritage reading was recorded for it, and its log's next-day section still reads "Outstanding". That log reserves tag `evidence-010` for the close-out. Stop point 2 of `publish-006` issued `evidence-010` again the same evening, and it now marks the `publish-006` baseline commit. The public log of `publish-005` therefore names a tag that marks a different commit. Record v5's bytes were not affected: both verifications of `publish-006`, on 20 and 23 October, found its disclosures manifest served unchanged. Found on 23 October while closing `publish-006`. The same day the founder decided (`APPROVE PF5-1`, 16:02 BST) to close `publish-005` late under its original publication authority: a fresh check limited to what it published, a new close-out tag `evidence-012` with `evidence-010` left where it is, a dated correction to its public log in a new commit, and closing and completion notes. The close is scheduled for 27 October 2026, after this record is published, and had not been carried out at its publication.

## 5. Considered and not included, with the reason

| Date | Item | Reason not included |
|---|---|---|
| 19 Oct | AR-20261019-007-zhzg8v: narrow verification of post-check edits to record v5 | Checking of the record; record v5's revision row states it |
| 20 Oct | AR-20261020-005-zhzg8v: production decision identifier for `publish-004` | Issued, never typed, void when the run stopped; no decision taken |
| 20 Oct | RETRY-1: restart as `publish-005` | Scheduling of the run |
| 20 Oct | AR-20261020-008-zhzg8v: identifier issued and voided unused | No request or decision under it |
| 20 Oct | AR-20261020-009-zhzg8v and AR-20261020-011-zhzg8v: filing of disclosure queue v16 and v17 | Queue maintenance |
| 20 Oct | AR-20261020-010-zhzg8v: production decision for `publish-005` | It is the publication of record v5, public since 20 Oct |
| 20 Oct | AR-20261020-010-j7bham: production decision for `publish-006` | It is the publication of GET v6, public since 20 Oct; its order of authority is Section 4 item 3 |
| 20 Oct | AR-20261020-001, -004, -007 and -009-j7bham: filing of a working status register and of unexecuted working drafts of the rights-transfer package | House filing; adopts no terms. For -009-j7bham the files are present in `99-Working`, dated 21 October, but the typed decision is not established from the sources read |
| 20 Oct | AR-20261020-003-j7bham: the route by which the founder's rights pass to the holding company | No decision taken; open (Section 6 item 3) |
| 21 Oct | Pilot, demonstration, website and commercial work, including correspondence with an implementer confirming the v6 digest | Pilot, commercial and drafting positions; none changes the schema, the record site or a disclosure record |
| 22 Oct | P-D7: re-run of the `publish-006` evidence-tag step in a fresh working copy, with the lost output recorded | Execution of the publication evidence steps under the `publish-006` authority; its cause is Section 4 item 7 |
| 22 Oct | Filing of disclosure queue v18 and v19, and the scheduling of this run | Queue maintenance and scheduling |
| 22 Oct | A bounded, confidential technical comparison with an implementer using synthetic data | Commercial position; nothing released. If it proceeds to public transparency-log entries, that decision will be assessed for the next record |

## 6. The unresolved matters, stated as unresolved

1. **Charter §8.** Whether Custodian Charter v3 §8 applies to any decision above is unresolved. No solicitor is instructed on it.
2. **The Custodian board is not constituted.** No schema version has been approved.
3. **Title to the schema estate is unvested.** The rights are held by the founder personally. Their transfer to the holding company is gated (Section 3) but its route is undecided, and no transfer is executed. Vesting in, and handover to, the Foundation are separate later steps; neither has occurred.
4. **GET.** v6 is published for comment only; it is not approved and not a candidate. v5 remains the candidate. The identifier `/get/1.0` is not served. The comment period runs to 10 November 2026.
5. **Evidence for `publish-006`.** The release's post-publication evidence checks and commits have run: preservation and baseline on 22 October (tag `evidence-010`) and the next-day checks on 23 October (tag `evidence-011`, live verification 156 of 156). Both ran late (Section 4 item 8). One filing step remains: the internal log entries recording where the run's evidence was filed. The latest Software Heritage snapshot, of 22 October, holds the publication but neither evidence tag.
6. **The 1.1 list** stands as v4 §6 item 5 states it.
7. **An internal target date** for GET exists and is not disclosed (D11).

## 7. The publication control

- **This record is the third run under the interval rule.** The rule's review falls after this run.
- **The earliest due date in this record is 26 October 2026** (AR-20261019-005-zhzg8v and CDX-SLATE (a) C3). The decisions of 20 October are due 27 October; GETV6-COMMENT 29 October.
- **What closes this run** (owner DM; executed under `OPS_Disclosures_LiveSequence_Runbook_v4` as release `publish-007`, not by this record): freeze this record as a build identified by digest; deploy it through the public record pipeline with the Charter, v3 to v5 and the GET record carried unchanged; live-verify; record publication from the live evidence; preserve the evidence.

## 8. What has not happened

- No schema version has been approved. GET v6 is published for comment, not approved.
- No vector set has been adopted, and no conformance recognition has been recorded.
- No licence to implement GET has been granted.
- No transfer of the founder's rights has been executed.
- Nothing in this record authorises any of those.

## 9. References

1. **Schema Independence Charter v3**, 29 July 2026. §6 publication; §7.6 the founding-period mechanism.
2. **Arkaya Layer1 Custodian Charter v3**, 30 June 2026.
3. **Arkaya_Layer1_PublicationRecord_ConflictedDecisions_2026-10-19_v5**, published 20 October 2026.
4. **OPS_DecisionRecord_PubRecordV5_Publication_publish-005_2026-10-20_v1**, and `PUBLICATION_LOG_publish-005.md` at `evidence-009`. Section 3 row 3; Section 4 items 1 and 2.
5. **OPS_FailureNote_publish-004_Restoration_2026-10-20_v1**. Section 4 item 1.
6. **OPS_DecisionRecord_GETv6_PublicationForComment_publish-006_2026-10-20_v1**. Section 3 rows 4 and 6.
7. **Release objects for `publish-006`** in the public record repository: release commit `678aad800ba9849fa26651b927a107de7aa03fd3` (tag `release-007`); publication commit `98da0263d27e4b55934e3109fc6284714e4ed021` (tag `publication-006`); `ledgers/get/publications/1/publication.json`. Section 3 row 6; Section 4 items 3 and 4.
8. **Action Register deposits AR-20261020-001 to -007-j7bham**, 20 October 2026. Section 3 rows 2, 4 and 5; Section 5.
9. **OPS_ReleaseRecord_C3_GET_Pipeline_NextVersion_v2** and **OPS_HandoffReceipt_CDX-20261020_C3_GET_pipeline_next_version_v1**. Section 3 row 2.
10. **OPS_Register_GETv6_Comments_v1**, 22 October 2026. Section 3 row 8.
11. **OPS_DisclosureQueue_2026-10-22_v19**. Current rows, clocks and proposed inclusions as this record was prepared; **v20** for row Q-8 (Section 4 item 6); **v21** for rows Q-9 and Q-10 (Section 4 items 7 and 8); **v22** for row Q-11 (Section 4 item 9); **v23** and **v24** for the inclusion decisions of 23 October; v18 for Q-5 and Q-6 as first recorded; v17 for the position at 20 October.
12. **Arkaya_Governance_Founding_Period_Decision_Authority_and_Stewardship_Pathway_v3**. §2 (custody in the interval); §4.5 (vesting and handover).
13. **`PUBLICATION_LOG_publish-006.md`** in the public record repository, at tag `evidence-011` (commit `9d12371410feb590840c842e34dedf35583b22cc`): sections 7 to 9 and section 8 items 5 and 6. Section 4 items 7 and 8; Section 5 (P-D7); Section 6 item 5.
14. **`PUBLICATION_LOG_publish-005.md`** in the public record repository, at tag `evidence-009` (commit `863c2da492baf8340a758323f879793d9edc4d51`): section 7 (the reservation of `evidence-010`) and section 11 (next-day checks outstanding). Section 4 item 9.
15. **OPS_DecisionRecord_Publish005_LateClose_2026-10-23_v1**, `10-Layer1-Custodian/Governance/`. Section 4 item 9 (the decision to close `publish-005` late).

## 10. Revisions

| Version | Date | Change |
|---|---|---|
| 6 (build 13) | 26 Oct 2026 | Publication-state changes only, made at the freeze step under `APPROVE PUB-7` (DM, 23 October 2026, 16:45 BST), for publication on 26 Oct 2026 (release `publish-007`, `OPS_Disclosures_LiveSequence_Runbook_v4`). The status line names that approval and no longer refers to the independent check. Every `Disclosed` cell now carries the publication date, 26 Oct 2026; no interval was missed. Section 7's closing sequence names the release. The footer carries the approved status. Entries, capacities, operative effects, what each entry does not confer, review routes, external reliance, corrections, control failures, unresolved matters and references are unchanged. Made from build 12 (SHA-256 `cc8d9db4e17b2484c87d4a927d3200cc0df62c04080e7c40889333cfdbc7e8ee`), which is superseded by this build, not withdrawn. |
| 6 (build 12) | 23 Oct 2026 | Under `APPROVE INC-18` and `APPROVE DL-1 (a)` (DM, 23 October 2026, 16:37 BST), after the independent check of build 11 (PASS WITH FINDINGS, no blocking defect in the record): status line records INC-18 and that this record's own inclusion decisions are queued as entries for the next record and its preparation decisions queued for it; reference 11 cites queue v24. No other change. Build 11 (SHA-256 `4db1dd585ff1fd701eaa8a9ada9c1b3bf249931138ba14b9f6b78ab7dc7b9865`) superseded. For a narrow re-check. Not approved for publication. |
| 6 (build 11) | 23 Oct 2026 | Under `APPROVE INC-6` to `APPROVE INC-17` (DM, 23 October 2026, 16:21 to 16:26 BST): status line records that Section 3 rows 1 to 6 and Section 4 items 1 to 5 are included and that Section 5 stands as drafted; reference 11 cites queue v23. No other change. Build 10 (SHA-256 `48df8c2144b8c75478e45d85bfc0350e8b0b625a121d31c195bdb60a43e8756a`) superseded. For a narrow re-check. Not approved for publication. |
| 6 (build 10) | 23 Oct 2026 | After the independent check of builds 7 to 9 (FAIL, finding F1 blocking), under `APPROVE CHK-3 (a)` (DM, 23 October 2026, 16:16 BST): Section 6 item 5 says the evidence checks and commits have run and that one filing step remains (finding F1); Section 4 items 7 and 8 say what the 20 and 23 October checks found rather than that nothing changed in the interval (finding F3); status line updated. No other change. Build 9 (SHA-256 `589343d39496e75e9862d907a423d4134b08622ace447d53a909df37c4dc440a`) superseded. For a narrow re-check. Not approved for publication. |
| 6 (build 9) | 23 Oct 2026 | Under `APPROVE PF5-1` (DM, 23 October 2026, 16:02 BST): Section 4 item 9 states the decision to close `publish-005` late and that the close is scheduled after this record's publication; reference 15 added; status line updated. No other change. Build 8 (SHA-256 `7af07172b1c7d72e8010d903b889c95975d71c9c5329801377f031075aa86efe`) superseded; builds 7 to 9 go to one narrow independent check. Not approved for publication. |
| 6 (build 8) | 23 Oct 2026 | Under `APPROVE INC-5` (DM, 23 October 2026, 15:26 BST): Section 4 item 9 (Q-11) added and its lead changed from eight places to nine; reference 11 cites queue v22; reference 14 added; status line updated. No other change. Build 7 (SHA-256 `0f3032761987dc092037ca765568f42156bc1b364f34898f7610c90280642de9`) superseded before any check; builds 7 and 8 go to one narrow independent check. Not approved for publication. |
| 6 (build 7) | 23 Oct 2026 | Under `APPROVE INC-4` (DM, 23 October 2026, 15:20 BST): Section 4 items 7 (Q-9) and 8 (Q-10) added and its lead changed from six places to eight; P-D7 added to Section 5 as not included; Section 6 item 5 states the completed evidence steps of `publish-006`; reference 11 cites queue v21; reference 13 added; status line updated. No other change. Build 6 (SHA-256 `e3277deb58aaf0e1e5477646318037010acf681b053a59048d076b378f70e3f8`) superseded. For a narrow independent check. Not approved for publication. |
| 6 (build 6) | 22 Oct 2026 | After the independent narrow check of build 5 (PASS WITH FINDINGS, no blocking finding), under `APPROVE CHK-2` (DM, 13:15 BST): Section 3 row 4 states that GETV6-7 authorised the served Reason wording, as that check determined, with its reliance on the founder's quotation; status line updated. No other change. Build 5 (SHA-256 `c4a685c4812313000e07691ab092a74f6bdbf89552636098a74f777a156ad423`) superseded. This change not separately checked, by decision. Not approved for publication. |
| 6 (build 5) | 22 Oct 2026 | After the independent check of build 4 (FAIL; findings 2 to 6 blocking), under `APPROVE CHK-1` and `APPROVE INC-1, INC-2, INC-3` (DM, 13:03 BST): Section 3 row 3 distinguishes v5 build 8 as approved from build 10 as published (finding 2); row 4 states the served Reason wording verbatim and leaves open whether GETV6-7 authorised it (finding 6); Section 5 states that the typed decision for -009-j7bham is not established (finding 5); Section 4 item 6 gives the 20 October publication as 14:40 BST (finding 3); the assessment markers on Section 3 rows 7 and 8 and Section 4 item 6 removed as included (finding 4); status line, Section 4 lead and Section 7 updated to match; reference 11 cites queue v20 for row Q-8 (finding 7). Build 4 (SHA-256 `7260e4071413ad9d49aa764be5caec9aec2e6e1083a7f3e3e12fbd8571a8b5d6`) superseded. Changes unchecked. Not approved for publication. |
| 6 (build 4) | 22 Oct 2026 | Under `APPROVE D-DRAFT-2` (DM, 12:35 BST): the D-DRAFT-2 marker in Section 3 row 7 replaced by the statement that the discovery document was served until 22 October and corrected by publication at 12:29 BST; status line updated. No other change. Build 3 (SHA-256 `0c3a277f13857578f648387b0f8c15c28d4d6f5db5407a4a4c7edaf903bb8d26`) superseded before any check. Not checked. Not approved for publication. |
| 6 (build 3) | 22 Oct 2026 | Under `APPROVE W1` (DM, 12:27 BST): Section 4 item 6 records that the discovery document and `llms.txt` were corrected and published at 12:29 BST on 22 October; status line updated. No other change. Build 2 (SHA-256 `68beb9aa57b9316dc338fa2dddca3bd305e855a5e397aefb3b86fbf5303054fa`) superseded before any check. Not checked. Not approved for publication. |
| 6 (build 2) | 22 Oct 2026 | Under `APPROVE D-DRAFT-1` (DM, 11:41 BST): the D-DRAFT-1 marker removed and its text kept as drafted in Section 3 row 6; status line updated. No other change. Build 1 (SHA-256 `f160999d559ae80589575efee9e7134504e20c5427ca421dc1ad2b77c1635fec`) superseded before any check. Not checked. Not approved for publication. |
| 6 (build 1) | 22 Oct 2026 | Prepared under `APPROVE R2` from queue v19. Eight entries (all proposed, two to assess); five control failures and a sixth proposed (the discovery document's statement on v6, found still served on 22 Oct); twelve items considered and not included. Drafting decisions D-DRAFT-1 and D-DRAFT-2 open. Not checked. Not approved for publication. |

*AKR AIPS · v6 · APPROVED FOR PUBLICATION · Layer 1, no house identity.*