# v0.14.0 Research Methodology and Verification Boundary Validation ## Validation scope This record validates the v0.14.0 public Research Methodology & Verification addition and the repository state packaged from `/mnt/data/mi_v0140_work`. The release adds `/research/methodology/` as a first-party explanation of how supplied research is curated, how source and curated SHA-256 identities are used, what automated discovery features mean, what they do not prove, and when fresh external verification is required. It does not change the authoritative bodies of the 25 curated supplied reports, their manifest membership, or the accepted glossary definitions. ## Executed checks - All 102 PHP files passed `php -l`. - All 83 canonical public routes returned HTTP 200 from the local PHP test server. - All 83 routes had exactly one H1, a non-empty unique title, a non-empty unique meta description, the expected clean self-referential canonical URL, and parseable JSON-LD. - `/research/methodology/` rendered the expected public methodology boundary, breadcrumb, `AboutPage` structured data, and five-stage `ItemList` workflow. - The methodology page was linked from `/research/`, `/research/library/`, `/status/`, all six research topic hubs, and all 25 report-reader pages. - The primary navigation remained intentionally unchanged: three disclosure groups plus the direct Share action. - Three stale research-fragment links on Terminology, Identity, and Stewardship were replaced with their current research topic hubs. - 178 clean-URL redirect assertions passed: root `/index.php`, missing-trailing-slash and route `/index.php` variants for all non-root canonical routes, plus the retained legacy public root PHP endpoints. - 1,077 unique rendered first-party targets and fragments resolved successfully. - `sitemap.xml` contained 83 canonical page URLs, 8 image entries, and 1 video entry; `robots.txt` retained the expected sitemap pointer. - The research manifest remained byte-identical to the v0.13.0 baseline: SHA-256 `7b5388fd9b0a1a99c266aa204a0575c170787ec146ce91ee6a2b5b1e26ca2dcb`. - `includes/glossary-data.php` remained byte-identical to the v0.13.0 baseline: SHA-256 `85787f020d86c48a0d7497f50973008b1ca10b61cdca10378fa3f4206fbd3a66`. - All 25 curated report files matched the curated SHA-256 values recorded in the manifest. - All 25 original supplied attachment files available in the conversation workspace matched their recorded source SHA-256 identities. - All 25 curated reports retained both durable anchors: `#curation-boundary` and `#research-body`. - The exact UAIX wizard/update URL remained present exactly once in each required owner file: `AGENTS.md`, `.uai/readme.human`, and `.uai/agent-pickup.uai`. - TABOO-001 leakage was zero outside its owning `.uai/taboo.uai` record. - The curated-report paternalistic-catchphrase scan returned zero findings. - Lightweight raw-secret marker checks returned zero findings. - Active handoff intake buckets contained only inert `.keep` placeholders. - Repository-document validation resolved 193 local Markdown links, 59 durable long-term-memory pointer records with 83 section/review destinations, and 329 path-like UAIX references with zero errors. - The root-deployment archive clean-extracted with 331 entries, preserved root dotfiles/directories, re-linted all 102 extracted PHP files, re-smoked all 83 canonical routes, and re-ran the rendered-link/research/UAIX validation checks with zero errors. ## Evidence boundary validated The new methodology page correctly distinguishes four different kinds of evidence: 1. **Repository behavior** is established by current source code plus executed local validation checks. 2. **Source identity** is established by the SHA-256 identity of the exact supplied attachment bytes when those source files are available. 3. **Curated-edition identity** is established by the SHA-256 identity of the exact durable repository edition. 4. **External truth or currentness** is not established by either hash and requires fresh primary verification when a public claim depends on unstable legal, scientific, standards, vendor, deployment, search, or other external state. The release also correctly describes topic membership, exact glossary matching, and Related Research calculations as first-party discovery/navigation mechanisms rather than proof of citation, endorsement, agreement, authority, or semantic equivalence. ## Unchanged authority surfaces No substantive changes were made in this release to: - the 25 curated supplied research bodies; - `docs/long-term-memory/research/report-manifest.json`; - the accepted 20-term glossary definitions; - the PHP/HTML5/CSS/native-JavaScript runtime constraint; - the no-database/no-third-party-runtime architecture; - clean canonical URL policy; - the project's identity-versus-key boundary; - the rule against implying consciousness, sentience, personhood, citizenship, legal rights, or human equivalence from terminology alone. ## Unavailable or deliberately unverified state This validation does not establish production deployment state, DNS, TLS, apex/`www` normalization, server-specific rewrite behavior, CDN/cache configuration, Search Console or Bing Webmaster state, search indexing/ranking, production Core Web Vitals, or independently verifiable cryptographic machine-execution provenance. No live deployment was authorized or performed. ## Result The v0.14.0 Research Methodology & Verification boundary and its clean-extracted root-deployment package are locally validated, subject to the external-state limitations above.