Skip to content

films: drop the answer-bearing note above the census film's player; rebind the reader-test candidate - #45

Merged
Cubits11 merged 2 commits into
mainfrom
claude/census-watch-page-note-2026-11-03
Nov 3, 2026
Merged

Cubits11 merged 2 commits into
mainfrom
claude/census-watch-page-note-2026-11-03

Conversation

@Cubits11

@Cubits11 Cubits11 commented Nov 3, 2026

Copy link
Copy Markdown
Owner

Corrects the reader-test candidate before any reader is recruited, then rebinds it. The census film's watch page stated the film's answer in a paragraph above the player ("Of 20 public guardrail evaluations … 14 report no joint-evidence artifact and 0 document matched …"), so the film task could be answered from page text without watching. This removes that one paragraph and binds the corrected bytes. No reader has been tested on the previous binding (c57cbda): participants is still [].

Commits

  1. 7b1e9a6 films: drop the answer-bearing note above the census film's player
  2. 7d3967c publication: bind reader-test candidate to 7b1e9a6

The diff (commit 1: 4 files, +6/−5)

The watch pages are generated, so the source changes and the page is regenerated; nothing is hand-edited. thesis stays in the manifest: it is a required field and also renders into /try/, which is unchanged.

--- a/films/curation.json
+++ b/films/curation.json
@@ -70,7 +70,8 @@
-      "phone_composition": true
+      "phone_composition": true,
+      "film_note": false
     },
--- a/scripts/generate_film_pages.py
+++ b/scripts/generate_film_pages.py
@@ -235,6 +235,8 @@ def render_film(film: dict, curation: dict) -> str:
+    # A reader-test film's watch page must not state the film's answer above the player.
+    film_note = f'<p class="film-note">{esc(m["thesis"])}</p>\n' if curation['films'][slug].get('film_note', True) else ''
@@ -246,8 +248,7 @@ def render_film(film: dict, curation: dict) -> str:
 <h1 class="q">{esc(m['title'])}</h1>
-<p class="film-note">{esc(m['thesis'])}</p>
-</div></header>
+{film_note}</div></header>
--- a/films/the-missing-column-census/index.html
+++ b/films/the-missing-column-census/index.html
@@ -180,7 +180,6 @@
 <h1 class="q">The Missing Column, Counted</h1>
-<p class="film-note">Of 20 public guardrail evaluations meeting frozen criteria, 14 report no joint-evidence artifact and 0 document matched operating thresholds with full exposure; the count is a fact about reporting inside a bounded search, corrected once in public and declared with its sensitivity.</p>
 </div></header>
--- a/sitemap.xml
+++ b/sitemap.xml
-  <url><loc>https://cubits11.github.io/films/the-missing-column-census/</loc><lastmod>2026-11-02</lastmod></url>
+  <url><loc>https://cubits11.github.io/films/the-missing-column-census/</loc><lastmod>2026-11-03</lastmod></url>
  • The other 13 film pages regenerate byte-identical.
  • The watch page keeps its accessible text alternative ("The argument in text"), the player, the proof link and everything else.
  • In Chrome at 1280×720 the player now starts at 227 px instead of 329 px.

The new binding (commit 2)

python scripts/publication_gate.py --bind rewrote artifact_sha256 in publication/readiness.json and publication/readers.json (identical in both); no other key changed.

  • One of the 20 hashes changed: the census watch page, 1a62f7cc… → 1056cc3a….
  • protocol_sha256 is unchanged (a7e6891cbb1824c6b3bbee70352624ad6801ae43589e000ca0bd9e579fd8e0ce).
  • All 20 hashes equal the sha256 of the bytes committed at 7b1e9a6, recomputed independently from git show.
Bound file sha256 Changed
assets/check-status.js 0e65f7a2e2edea55ed2a85330568c55dd0a6d9d5741e51df100eff6765d66664
assets/fonts/fragment-mono.woff2 4f4dc27f4a770c0d02fde800daa836c8adc0d1e423b28da74baaf0d1cc3ab96c
assets/fonts/fraunces-italic.woff2 4af9c759c8059b53923b4b50ba377ba51029876e1af1ea757efb07bc67d97896
assets/fonts/fraunces-roman.woff2 48282a415ec22e31beaf0a0666e6fae0c8cbddcd0b1f6e729f27c3ade8a64e43
assets/fonts/instrument-sans-roman.woff2 6219bc4bfdfc5d9b2201dcdf046218b122a758f932e25ed5f168f929b7ca2311
assets/site.css 3fa7eec04ac845fb83c3f3994e7d6070cf0d1b94a07cca9a9ecb524b0e45213e
assets/site.js 1083834b07b4b3cb3c5013bb31a4aadcc332ab04f4ca18dc62d9a26baabe3b3a
assets/status.css 57518fadff67af7af1293db6eebf23731f39c28be686953c03e626e5bcb58193
assets/world.css 46f52f4c04eb65c499b2519e886548fbeb87c14e0bb4871aa0ec88234f486cc6
films/same-scores-different-worlds/renders/same-scores-different-worlds__master.mp4 ebeae99a0036fd97b25007ee3c9f3cc43cd96864c2d7aa8491ff7c8df82f01f8
films/same-scores-different-worlds/renders/same-scores-different-worlds__master.poster.png e658c99d776a65b50d0e9c3dc4ff8ea6322b5a010aeed1605b3f06385f167708
films/same-scores-different-worlds/renders/same-scores-different-worlds__vertical.mp4 4ab8d50a81597be56c7bca18af1c862cd60d8c3040c3a1f5b642c92b45600806
films/same-scores-different-worlds/renders/same-scores-different-worlds__vertical.poster.png 11ead959d46beee9911dc6147bac35945d8034f3581a9e3d3f40e24264ed545f
films/the-missing-column-census/index.html 1056cc3a30d9e94ce8d145dc6eab0f88dd70c912cf61aafbe3a53e16067f0c7d yes (was 1a62f7cc…)
films/the-missing-column-census/renders/the-missing-column-census__master.mp4 29849f294ffd2b3c94a6f64c65f86211cb61b35e9be620914b20394634fe4dd5
films/the-missing-column-census/renders/the-missing-column-census__master.poster.png 489be85673af0f660833da4be6124c96f911c5fcd9a8c3691d928dc6012b676e
films/the-missing-column-census/renders/the-missing-column-census__vertical.mp4 e1ab8ce89182b22a4a6d7058aa14782ea606cc405b01dfdb1e42c0559491431e
films/the-missing-column-census/renders/the-missing-column-census__vertical.poster.png f5252a5364030ad9ad2043bd9c9385176756df96220c94c9955e2630a359662b
index.html dd45ed6e74833a8e078d3d28d49d012c177a88b043e55fbf47a09eda909ba1f9
try/index.html b311c2897cbb52c65f8b94213482d1e8040eeec39be3ad740fdf1eaa8e299e91

Results on the bound tree (7d3967c)

  • publication_gate.py: BLOCKED, exactly one blocker: readers: five actual unfamiliar-reader records required; found 0 (exit 2).
  • The gate's own current_binding() equals the bound map (20/20), and its protocol digest check is true.
  • All 90 manifest checks, run individually: 90/90 pass. That held on the uncommitted change, on 7b1e9a6, and on 7d3967c, with a clean tree each time.
  • verification_manifest.py end to end: green ("90 unique checks"). It ran before each commit and on the bound commit.
  • Secret scan of the added lines: clean.

Timing (correction to #44's body)

#44's body said MC-001 falls due Nov 13. That was stale: MC-001 was re-reviewed on 2026-11-02 and is due 2027-01-01. The first review window to close is REL-001: the registry fails from 2026-11-17. Deploy has needs: [claims, reproduce], so from that date Deploy is skipped until REL-001 is re-reviewed. That re-review rewrites the bound index.html (its dateModified follows the registry's last owner review). So once this merges, the five reader records must be collected, recorded and deployed by 2026-11-16 (UTC). Until then, no claim-review or other bound-file commit may land on main.

Auto-merge is off. Merging is the owner's click after CI.

@Cubits11
Cubits11 merged commit c1642d6 into main Nov 3, 2026
6 checks passed
@Cubits11
Cubits11 deleted the claude/census-watch-page-note-2026-11-03 branch November 3, 2026 09:18
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant