CDN-245 · The Disjointness Audit — GCFT Dark-Sector Book Arc (Art Edition)

CDN-245 audits the CDN-244 frozen holdout sample against the CDN-240 25-cluster CLUSTERS list before any κ-proxy evaluation. The audit (cdn245_audit.py, deterministic) finds 9 exact-name overlaps and 1 alternate-name overlap (RX J1347.5-1145 matched against CDN-240's RXJ 1347-1145), for a total of 10 of 15 CDN-244 holdout clusters in conflict with the CDN-243 §3 disjointness clause. Only 5 clusters are confidently disjoint (Abell 2390, RXC J2228.6+2036, Abell 370, ZwCl 1358+62, Abell 851), well below the §3 minimum of N=15. The inability-to-test branch fires. CDN-244's sample lock (CDN244_FROZEN_SAMPLE_LOCKED) is SUPERSEDED. The 15-cluster list in CDN-244 §3 is not a valid holdout. What is NOT refuted: the CDN-242 result. The Class-P bound ≤0.29% population-averaged non-GCFT fraction, the GCFT coherence floor ≥99.71%, and the Class-F super-ceiling designation all stand. Refutation would have required a valid out-of-sample test that failed; what we have is an invalid sample that never ran. CDN-243's prediction template and decision rule are likewise unaffected. Root cause: CDN-244 cross-referenced its candidate list against the CLASH compilation only — because the single exclusion CDN-244 logged (MACS J1206.2-0847) was a CLASH cluster — and missed that CDN-240's 25-cluster sample drew from THREE catalogues: LoCuSS (Okabe+ 2016), CLASH (Umetsu+ 2014), AND CCCP / WtG. The CDN-240 source file lists every cluster with its catalogue tag; the audit script reads that file directly and cannot miss a catalogue. From CDN-246 onward, cdn245_audit.py is embedded in the build step as the gatekeeping tool: if the audit returns INABILITY_TO_TEST, the build halts before any Zenodo bundling. Instructions for CDN-246: pool the 5 confirmed-disjoint clusters from this audit as the starting set; source 10+ additional disjoint clusters from CCCP-extension (Hoekstra+ 2015) and WtG (von der Linden+ 2014) by mechanically filtering out every CDN-240 entry; re-run cdn245_audit.py before locking. Preserve and reuse CDN-242 locked physics, CDN-243 prediction bands and decision rule, and CDN-244's literature-source schema. Art Edition with star-field design system (seed 245). Lock: CDN245_DISJOINTNESS_AUDIT_LOCKED.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-06-16
DOI
https://doi.org/10.5281/zenodo.20715664
Primary Topic
Blockchain Technology Applications and Security
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

CDN-245 · The Disjointness Audit — GCFT Dark-Sector Book Arc (Art Edition)

Nicky Joseph Hubertus Catharina Hacquier
Zenodo (CERN European Organization for Nuclear Research)
Blockchain Technology Applications and Security
article

CDN-245 · The Disjointness Audit — GCFT Dark-Sector Book Arc (Art Edition)

Nicky Joseph Hubertus Catharina Hacquier
article en

Abstract

CDN-245 audits the CDN-244 frozen holdout sample against the CDN-240 25-cluster CLUSTERS list before any κ-proxy evaluation. The audit (cdn245_audit.py, deterministic) finds 9 exact-name overlaps and 1 alternate-name overlap (RX J1347.5-1145 matched against CDN-240's RXJ 1347-1145), for a total of 10 of 15 CDN-244 holdout clusters in conflict with the CDN-243 §3 disjointness clause. Only 5 clusters are confidently disjoint (Abell 2390, RXC J2228.6+2036, Abell 370, ZwCl 1358+62, Abell 851), well below the §3 minimum of N=15. The inability-to-test branch fires. CDN-244's sample lock (CDN244_FROZEN_SAMPLE_LOCKED) is SUPERSEDED. The 15-cluster list in CDN-244 §3 is not a valid holdout. What is NOT refuted: the CDN-242 result. The Class-P bound ≤0.29% population-averaged non-GCFT fraction, the GCFT coherence floor ≥99.71%, and the Class-F super-ceiling designation all stand. Refutation would have required a valid out-of-sample test that failed; what we have is an invalid sample that never ran. CDN-243's prediction template and decision rule are likewise unaffected. Root cause: CDN-244 cross-referenced its candidate list against the CLASH compilation only — because the single exclusion CDN-244 logged (MACS J1206.2-0847) was a CLASH cluster — and missed that CDN-240's 25-cluster sample drew from THREE catalogues: LoCuSS (Okabe+ 2016), CLASH (Umetsu+ 2014), AND CCCP / WtG. The CDN-240 source file lists every cluster with its catalogue tag; the audit script reads that file directly and cannot miss a catalogue. From CDN-246 onward, cdn245_audit.py is embedded in the build step as the gatekeeping tool: if the audit returns INABILITY_TO_TEST, the build halts before any Zenodo bundling. Instructions for CDN-246: pool the 5 confirmed-disjoint clusters from this audit as the starting set; source 10+ additional disjoint clusters from CCCP-extension (Hoekstra+ 2015) and WtG (von der Linden+ 2014) by mechanically filtering out every CDN-240 entry; re-run cdn245_audit.py before locking. Preserve and reuse CDN-242 locked physics, CDN-243 prediction bands and decision rule, and CDN-244's literature-source schema. Art Edition with star-field design system (seed 245). Lock: CDN245_DISJOINTNESS_AUDIT_LOCKED.

Zenodo (CERN European Organization for Nuclear Research)
Dynamique du noyau (FR)
Peace, Justice and strong institutions, Reduced inequalities
Openalex Percentile: Top 3%
Blockchain Technology Applications and Security
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.