session 2026-07-13 (afternoon): skill-harvest — split-cause-from-magnitude + confirm-cause-by-swap-in (ladder proposals)

2026-07-13 17:55:21 +02:00
parent bc52c520bb
commit 32dfd7a14c
1 changed files with 9 additions and 0 deletions
@@ -454,3 +454,12 @@ The single place proposed skills live so they don't evaporate between sessions.
 | **Symmetria §3 flag: soft-classification-where-a-checkable-claim-was-available** | Symmetria §3 flag | Shipping a soft label ("this is reordering", "magnitude unresolved", "apparatus") when a CHECKABLE claim (a reportable number, a substrate-verified fact, a discriminating test) was producible is contamination shape — the soft label can be true-shaped and wrong, and papering over defect. Antidote: ask what checkable claim the label stands in for, produce THAT; if it can't be produced cleanly, that inability is the finding; never force a figure onto a comparison that can't bear it. Earned HARD 2026-07-13 (steward-named "5 times in one day; not coincidence" — nagarjuna, nested-block, loss-half injection, Aeschylus reordering, matcher key). Feedback memory [[feedback-checkable-claim-surfaces-bugs]] already saved; a §3 flag makes it operational at check-time. | Symmetria §3 | **PROPOSED** |

 *(One proposal, strongly earned — recurred 5× in a day and the steward named the pattern. The feedback memory captures it for recall; the §3 flag would make me reach for the checkable claim before shipping the soft label. Surfaced for steward authorization — not applied.)*
+
+### Harvest 2026-07-13 (afternoon — the 18 body-deficit cause-analysis)
+
+| Element | Kind | One-line | Where it lands | Status |
+|---|---|---|---|---|
+| **split cause from magnitude before sizing a remedy** | verification-ladder instrument | A diagnostic bucket keyed on ONE summary axis (magnitude, holds%, a deficit size) can hold **heterogeneous causes** — a single label silently implies a single fix at a single scale. Before sizing remediation from a bucket, **read the substrate to split cause from magnitude**. Earned today: the one "18 BODY-DEFICIT" label masked THREE mechanisms → three remedies (re-match 8 / re-extract 4 / re-verify-and-reconvert 12); sizing the reprocess from the magnitude alone would reach for the wrong fix on most. Kin to census-through-truncation (that flag is about *coverage* of a scan; this is about *cause-homogeneity within a bucket*). | `reference-verification-ladder.md` | **PROPOSED** |
+| **confirm-a-named-cause-by-swap-in (don't assert from identification)** | verification-ladder instrument | When you NAME the true reference / config / cause behind an anomaly, don't assert the fix from the identification — **swap the candidate in and confirm the anomaly collapses** (ideally to zero). Today: naming augustine's true key ('Confessions' full) and philo's ('On Abraham') wasn't trusted; re-running the comparison against each drove fab+lost to exactly 0/0 — the checkable version of "this is the wrong match." The specific form of [[feedback-checkable-claim-surfaces-bugs]] for root-cause claims. | `reference-verification-ladder.md` | **PROPOSED** |
+
+*(Two, both earned in use, both recurring-shaped and distinct from the morning's checkable-claim §3 flag. The first is the session's load-bearing lesson — magnitude ≠ cause in a diagnostic map. Watch-list, not proposed: "counted-the-bundled-representation-as-items" [the ③ table row = 3 Aeschylus books counted as 1] — logged as a KG drift-pattern; may graduate to a census-through-a-pattern refinement if it recurs.)*