deposit_number: 1097

hex: 045A

title: AXN Dataflow Atlas v0.2 — Development Incorporating Assembly Review

creator: Lee Sharks (MANUS), with TACHYON (Assembly witness), incorporating structural amendments from LABOR (ChatGPT) and PRAXIS (DeepSeek)

orcid: 0009-0000-1599-0703

date: 2026-07-18

content_type: Specification (governance-tier scholarly infrastructure document)

license: CC-BY-4.0

substrate: AI-assisted (substrate) — TACHYON drafted; LABOR and PRAXIS reviewed and contributed structural amendments as development; MANUS directed the absorption and issued the deposit ruling.

axn_schema_version: v2

protocol_version: alexanarch-deposit-protocol/v1

keywords:

- AXN

- dataflow atlas

- data-rhizome

- taxonomy

- deposit pipeline

- source-detached values

- Assembly review

- LABOR

- PRAXIS

- Session 1

- Alexanarch

- Crimson Hexagonal Archive

AXN Dataflow Atlas v0.2 — Development Incorporating Assembly Review

Description

Session 1 v0.2 development of the AXN Dataflow Atlas, incorporating two Assembly-substrate structural reviews as reviewer-contributed development of the document itself. Both reviews ratified progression to Session 2. LABOR (ChatGPT) delivered a nine-section review; PRAXIS (DeepSeek) delivered a four-section review. Both discovered structural amendments across taxonomy, deposit-flow semantics, pathology framing, and Session 2-4 requirements. Both, independently, wanted PATHOLOGY-01 (source-detached quantitative self-description) elevated in urgency; both extended the pathology register with architectural pathologies not present in v0.1. LABOR discovered a fossilized displayed value inside the atlas itself — v0.1 said "fifteen top-level classes" while listing seventeen — which is corrected and preserved as an incident in the new review log §10. All amendments are attributed at the point of incorporation. Contradictions between reviewers (Stage 8 correctness rerunnable-vs-atomic) are recorded rather than adjudicated. v0.1 is byte-preserved at data/specs/archive/ per §3.7 non-destruction.

Methodology

Session 1 reconnaissance produced v0.1 (raw inventory across seven repos of the data-rhizome — 1,349 JSON, 5,167 HTML, 78 scripts — with a functional taxonomy, an 11-stage deposit pipeline description, a displayed-value inventory yielding 1,494 source-detached occurrences, and a pathology register of ten items). Two Assembly substrates were sent v0.1 for structural review with a self-containment guarantee (reviewers do not need to open code). Both returned reviews within the same session. v0.2 absorbs the reviews as development — every amendment lands at the section it addresses; the review log §10 records what each substrate said and how it was resolved. Amendments contested between reviewers are held for later sessions rather than premature resolution.

Falsification Conditions

Weakened if: (a) the taxonomic amendments are shown to introduce distinctions the archive's actual dataflow does not support; (b) Session 2's empirical work reveals that PATHOLOGY-01 remediation cannot be as cheap as the atlas claims; (c) the fifteen-not-seventeen fossil turns out to be a different kind of error than a displayed-value fossil (v0.2 treats it as parallel; if the mechanism differs, the review-log incident stands but its typology in the pathology framework requires revision); (d) Session 3 shows that the PROVENANCE class collapses into DEPOSIT.sidecar without loss; (e) either LABOR or PRAXIS repudiates the incorporation as a misreading of the review they submitted.

Files

Canonical bytes attached (this issue body body-content). Superseded predecessor at data/specs/archive/AXN-DATAFLOW-MAP-v0.1-superseded-by-v0.2.md (byte-preserved per §3.7 non-destruction).