docs(board): correct my own count — 14 stale uploads, not 1

The first version of this amendment said one OPAL track on SoundCloud was
stale. Then I read the account instead of reasoning about it, and it is
fourteen: the entire Aug-12 first cut is up, thirteen of it under junk
permalinks (/opal26-wap, /12-mafia-12, /ceci-nest-pas-une-bombe) that roll
call could never have found, because roll call only looks where it expects a
track to be. Same error the board made before me, one size smaller.

The finding that explains PLN's ear exactly: eight of the fourteen stale
uploads carry cover art and not one of the six correct v4 tracks does. The
wrong cut is the one that looks finished.

Also corrected: 'every other track matches to the second' was never a verdict,
it was a tolerance. SoundCloud reproduces full_duration to the MILLISECOND, so
at the ±1.5s a transcode seems to deserve, the Aug-12 and v4 cuts of /wap
(361.0 vs 361.6) and /perfect (315.0 vs 314.6) are the same track. That is how
six stale uploads passed a duration check in August.

And two coincidences caught before they could become deletions: /saria-demo
(2025) and /au-revoir-lord-toyota (2020) match OPAL to 50ms by chance on a
132-track account. Duration is a strong discriminator and still not provenance.

    the ones with the cover art were the wrong ones.
    six months of a set played once, and the tell
    was never in the sound -- it was in which
    of the twenty ghosts looked ready.

TASKS_DUMP.md

@@ -1659,3 +1659,57 @@ audio ships from.
   `piment_bresilien.tidal`, `sunny_side_up.tidal`.
 - **No gig is scheduled.** OPAL 08-08 and CosmicSet 08-22 are both past; `armada/escales/`
   holds no future escale. Release is the only shipping thread, so it gets the priority.
+
+### 🔴 CORRECTION to this amendment, same session — it is 14 stale uploads, not 1
+
+The account was read (`master sc pull`, read-only, existing token — no login
+needed) and the lens built (`armada/tide-table/sc_generation_lens.py`, `fe8ec16`).
+The first version of this amendment said one track was stale. That was the same
+mistake the board made before it, in a smaller size: **roll call only looks where
+it expects a track to be.**
+
+`/home/pln/Work/Sound/Prod/sc_pull_2026-09-05/` now holds `catalog.json`
+(132 tracks, 10 playlists), **101 artwork originals**, and
+`opal26_generation_verdicts.json`. State of OPAL-26 on SoundCloud:
+
+| | count | detail |
+|---|---|---|
+| ✅ correct v4, right permalink | **6** | `/wap` `/there-s-something-about-drums-3` `/you-my-sunshine` `/piment-bresilien` `/eh-ouais-je-funk` `/ghosts-in-the-toilets` — all match v4 to the millisecond. **None has artwork.** |
+| 🔴 stale at the right permalink | **1** | `/take-five-drops` = 417.000 s = the **Aug-12** cut (v4 is 440.70). **Has artwork.** |
+| 🔴 stale strays, other permalinks | **13** | the *entire* Aug-12 cut, under junk slugs: `/opal26-wap` `/opal26-perfect` `/opal26-gimme-acid` `/opal26-piment-bresilien` `/opal26-something-about-drums` `/opal26-eh-ouais-je-funk` `/ceci-nest-pas-une-bombe` `/you-my-sunshine-5` `/desire` `/12-mafia-12` `/13-vague-de-crime-13` `/11-ghosts-in-the-toilets-11` `/15-la-revolution-sera-samplee-15`. **8 of them have artwork.** |
+| ⬜ never uploaded | **7** | `/ceci-n-est-pas-une-bombe` `/am-i-doing-it-right` `/perfect-3` `/gimme-acid` `/mafia` `/vague-de-crime` `/revolution` |
+| ✅ the full set | 1 | `/live-opal-2026` = 4361.355 s = the v4 nodrop mix exactly. Correct, and it has artwork. |
+| ❓ PLN's call | 2 | `/opal26-remix` (4273.548 s, unrecognised) · `/opal26_master_raw` (5207.454 s, the raw master, 2026-08-10) |
+
+**So PLN's ear was reporting something sharper than "the cuts are bad".** Browsing
+his own account he sees ~20 OPAL items; the ones wearing cover art are almost all
+the Aug-12 cut, and the six correct ones look unfinished. *The wrong cut is the
+one that looks finished.* Two permalink bugs are visible in that stray list and
+both are fixed upstream (`a6da6f5` one-function-decides-a-permalink): the doubled
+number `12-mafia-12`, and `ceci-nest-` vs the current slugify's `ceci-n-est-`.
+
+**Two false positives were caught and must never reach a delete manifest:**
+`/saria-demo` (2025-02-06) and `/au-revoir-lord-toyota` (2020-10-18) match OPAL
+durations to 50 ms by pure coincidence on a 132-track account. The lens now hard-
+gates strays on a date floor (nothing older than the oldest render on disk) and
+prints what it discarded. A duration collision is not provenance.
+
+### ✅ Also ruled out — the tolerance was the instrument, again
+
+The old note *"every other track matches to the second"* was not a verdict, it was
+a loose lens. Measured: SoundCloud's `full_duration` reproduces the source **to the
+millisecond** (477.600 uploaded → 477.600 reported). At the ±1.5 s a transcode
+seems to deserve, Aug-12 and v4 are indistinguishable on `/wap` (361.0 vs 361.6)
+and `/perfect` (315.0 vs 314.6) — which is precisely how six stale tracks passed a
+duration check in August. The lens runs at ±0.05 s.
+
+### 📋 The re-upload sequence, once PLN approves the delete list
+
+1. **Delete 14** (13 strays + `/take-five-drops`) — artwork already saved.
+2. **Upload 8** from `opal26_plan.json` (the 7 never-uploaded + take-five once its
+   permalink frees). Pass `--state` this time; no `sc_state.json` was ever written.
+3. **Re-attach the artwork** to the v4 tracks via `scwrite.py`'s `sc album` PUT
+   path — `scform` has no artwork verb, and the covers currently live only on the
+   cuts we are about to delete.
+4. Then OPAL's `audio:` field in `www/content/lives/2026/opal-festival-2026.md`
+   can finally be filled, and the catalog's release links behind it.