# EM-0024 execution plan

- Object ID: `em:documentation:sha256:049909b59189c00a49550cccd62b722c466af9ae48b63b7d5242df8ec9bf86eb`
- Kind: `documentation`
- Repository path: [`docs/execution-plans/EM-0024.md`](https://github.com/yoheinakajima/epistemedia/blob/f92846570180dfa4511263f8ba98ecd18f7772c9/docs/execution-plans/EM-0024.md)
- Content digest: `f7ab86bb06b9f3a49dc98c8dae7ffbb5d19ec6fd530514fe2d65485b8e42229f`

**Also filed under:** [Disclosure and Public Projection](https://epistemedia.org/topics/disclosure/), [Epistemedia](https://epistemedia.org/topics/epistemedia/), [Epistemic Mesh Protocol](https://epistemedia.org/topics/epistemic-mesh/), [Sovereign Realm Federation](https://epistemedia.org/topics/federation/), [Autonomous Governance](https://epistemedia.org/topics/governance/), [Knowledge Objects](https://epistemedia.org/topics/knowledge-objects/), [Human and Agent Interfaces](https://epistemedia.org/topics/public-interfaces/), [Releases and Reproducibility](https://epistemedia.org/topics/releases/), [Research Program](https://epistemedia.org/topics/research-program/), [Security and Adversarial Robustness](https://epistemedia.org/topics/security/)

## Source content

# EM-0024 execution plan

Status: candidate complete on `codex/em-0024-editorial-comprehension`; exact-head review pending.

## Objective

Make the existing first dossier answer a stranger's prior question—why a lineage-aware evidence
system matters—without adding a marketing page, philosophical manifesto, new case, new lens, or
research claim. Compress the oversized type and spacing, describe the accepted review as process
separation rather than human-style authority, and make the substrate's overlapping topic counts
arithmetically explicit.

## Change surface

1. Add one compact, visually structured homepage explanation after the featured Case 001 card. It
   will distinguish repeated information from warranted knowledge and route directly to the brief,
   skeptical view, and sentence-level evidence docket.
2. Reduce shared and dossier-specific H1 maxima, first-section spacing, and narrow-viewport type so
   the finding and evidence surface remain in the first viewport.
3. Reframe the public review page around a separate Codex review process using only conditions
   enforced by the accepted byte-bound receipt: fresh clone, independent retrieval, no authoring
   artifacts, checked scope, and explicit scientific-truth limitation.
4. Derive a catalog-wide topic-membership total in the compiler, label public objects as distinct,
   label each topic card as membership in that topic, and test the displayed arithmetic against the
   exact selected-object sets.

## Hard boundaries

- Preserve the accepted feature manifest, candidate dossier, independent receipt, source packet,
  evidence counts, lineage judgments, relation meanings, and policy evaluations byte-for-byte.
- Keep the compact one-row How We Know index and existing practical readings intact.
- Do not claim human review, another model family, a universal theory of truth, or proof that the
  scientific conclusion is true.
- No Case 002, new lens, public intake form, workflow, hosted API/MCP, DNS, package, release, or
  unrelated repository work.

## Verification

- exact copy, route, placement, heading, review-process, and count-reconciliation tests;
- regression checks for the one-case index and both practical readings;
- targeted Ruff and interface/featured tests;
- full `make check` with disclosure audit and deterministic comparison build;
- accepted input SHA-256 and Git path-scope comparison;
- desktop 1440 by 900 and mobile 390 by 844 browser inspection;
- exact-head independent review, protected squash merge, Pages deployment, and public read-back.

## Accepted research identities to preserve

- feature manifest SHA-256: `5c96dead036b527793ba5a0de59bf7316efdfeb591470d4a23e5bf979f3b9288`
- dossier SHA-256: `7003413e286e4d310f81441db33f4a467ba2eb3e08f41ddfa3cef5abb34707ca`
- independent receipt SHA-256: `503d16396b25b1c22d7fc10ac6fb7db2e530e6ce348d63fa8b639db5a5288f0a`

## Candidate verification

- focused interface and featured tests: 46 passed;
- full `make check PYTHON=.venv/bin/python`: 80 passed, public disclosure audit clean,
  deterministic comparison build exact, accepted source tree unchanged;
- `ruff check` passes for the changed featured module and its focused test, and no added diff
  line exceeds the configured 100-character limit; the broader core/test files retain their
  pre-existing lint backlog;
- accepted manifest, dossier, and independent receipt hashes match the identities above;
- desktop `1440 × 900`: no horizontal overflow, zero scripts, case H1 51.84 px with the finding,
  actions, and full evidence tally visible before the purpose bridge begins;
- mobile `390 × 844`: no horizontal overflow, zero scripts, case H1 31.98 px, finding and actions
  visible with the evidence tally beginning in the first viewport;
- review and How We Know pages have no overflow at both representative widths, and the purpose
  bridge's Docket link lands at `#evidence-record-title` on mobile.