analysis: complete the systematic claim verification; add F-017 and ADR-0010 draft

verify_claims.py gives all 813 claim rows a mechanical disposition;
the 28 flags were adjudicated by hand (REPORT.md appendix). Two survived
as genuine drift (F-017): a closed issue still described as open in
shared/lab-netzwerk.md, and a 'pending' decision block in hosts/cfgmon.md
whose premise the same file records as executed.

Also: narrow the vendored-path filter (it silently dropped 7 tracked
icon files and produced false path-miss flags), record the confirmed
canonical author identity in F-003, verify the Gitea#48->GitLab#46
numbering shift by title in F-005, and add the ADR-0010 draft under
analysis/drafts/ for the human to git-mv into decisions/.

Branch renamed to Neckbeard-v0.1.1-analyse-1 per the human.
This commit is contained in:
Thore Cimbal
2026-08-10 12:00:00 +00:00
parent ccf83da515
commit e68b295249
13 changed files with 1265 additions and 13 deletions
+22
View File
@@ -77,6 +77,7 @@ The declaration and the application live in different places, and nothing compar
| [F-007](findings/F-007-mirror-scope-claims-contradict-each-other.md) | Mirror count stated as five and six on the same day; `game-operating` claimed mirrored, is not | Gap in the availability guarantee the two-host topology exists for |
| [F-005](findings/F-005-dead-gitea-tracker-still-referenced.md) | Live doc routes to the tracker three docs declare dead, under an ambiguous number | A high-priority open decision is reachable only through a retired system |
| [F-010](findings/F-010-stand-labels-lag-their-own-commits.md) | Hand-written "Stand" labels older than their own file's last commit | Trains readers to distrust current content; gives no signal when content really is stale |
| [F-017](findings/F-017-prose-asserts-states-already-resolved.md) | Prose asserts states the tracker already resolved: a closed issue called open, a "pending" decision whose premise was executed 60 lines earlier | Sentences a reader would act on; found only by the systematic claim sweep |
### Pattern C — Two backlogs, one rule
@@ -144,6 +145,27 @@ Listed only, not filed as issues in the neckbeard repo — that is a separate ac
---
## Appendix — the systematic claim verification
`analysis/scripts/verify_claims.py` gave every one of the 813 extracted claim rows a
mechanical disposition (`analysis/data/claims_verification.tsv`): **205 checked-ok**
(a path, issue reference or state assertion verified against tree or tracker),
**207 informational** (tokens classed as runtime/net/image/forge references — correctly
outside the analysed trees), **360 prose-or-runtime** (no mechanically checkable token;
their semantic truth remains unchecked, stated as a limit, not passed silently),
**13 in code blocks**, **28 flagged** for hand adjudication.
The 28 flags resolve as: **10** references into out-of-scope repos (homelab/wiki,
wiki-bookstack, vendor, the retired gitops wiki) — consistent with their own context;
**7** host/runtime artifacts that correctly are not in any tree (`lab.conf`,
`runner-data/`, an untracked `.npmrc` the doc itself calls untracked, npm tarball
contents); **3** historically removed files whose removal the surrounding text records;
**2** enumerations of a hypothetical revert, not claims; **1** tokenizer artifact
(`text/html`); **3** bare/ambiguous issue references; **2** genuine drift → **F-017**.
Bonus verification: all of `roadmap.md`'s gitops references (#25, #45, #46, #47, #48,
#49) match their GitLab titles — post-migration references are consistently
GitLab-numbered, strengthening F-014.
## Open questions for the human
1. **Where do the 71 open issues go?** ADR-0002 Option C (repo canonical, mirrored to