diff --git a/evidence/CB-EV-0013-the-inert-clauses.md b/evidence/CB-EV-0013-the-inert-clauses.md index 1a3ba27..e3e06b6 100644 --- a/evidence/CB-EV-0013-the-inert-clauses.md +++ b/evidence/CB-EV-0013-the-inert-clauses.md @@ -4,7 +4,9 @@ CB-WP-0015 T03. Measured 2026-08-02 at `7e9ab22`+. Pass kind `product`, tier **S** (chaos d4=2, no override). Declaration 10 of 12. Per the rule adopted in CB-EV-0012, the cost table quotes **CB-WP-0014's** -final figure and declines to quote this pass's own. See §5. +figure and declines to quote this pass's own — and §5 records that the +rule, applied here for the first time, **did not hold**: the number moved +15% between the start of this pass and its end. --- @@ -164,19 +166,46 @@ what they always were. | pass | kind | responses | cost | $/response | |---|---|---|---|---| | CB-WP-0013 | meta | 47 | $8.26 | 0.176 | -| **CB-WP-0014** (final) | product | 34 | **$7.47** | **0.220** | +| **CB-WP-0014** | product | 48 | **$8.56** | **0.178** | | CB-WP-0015 | product | *provisional — not quoted, per CB-EV-0012 §5* | | | -**CB-EV-0009's standing prediction came due and is consistent — once.** It -predicted that a pass opening above the SH-1 hard line would cost more than -0.123 $/response. CB-WP-0014 opened above the line and cost **0.220**. +### The self-quoting rule is not enough, and this pass is the proof -That is one point past a threshold, and it is worth being explicit about -what it is not: there is **no control**. No pass has opened *below* the -line since the prediction was made, the two passes either side differ in -tier, kind and subject, and 0.220 is inside the range product passes have -shown anyway. The prediction is unfalsified, not confirmed. It needs a -pass that opens below the line to mean anything. +CB-EV-0012 adopted the rule that an evidence file quotes the **previous** +pass's final cost and never its own, because four passes running had +under-reported themselves mid-flight. This file opened by quoting +CB-WP-0014 at **$7.47 / 34 responses**, which is what `make status` said +at the time. By the time this pass closed, the same tool reported +CB-WP-0014 at **$8.56 / 48** — the figure moved by 15% *after* it was +supposed to be final. + +The cause is mechanical: a pass's window runs to the next pass's first +commit, so CB-WP-0014's window was still open until CB-WP-0015 declared. +**"The previous pass's final cost" is not final until the pass after it +starts** — the rule fixed the wrong boundary. Quoting from two passes back +would be safe; that is a change to the rule, and it is recorded here as +owed rather than made silently. + +### CB-EV-0009's prediction now has a point on each side + +It predicted a pass opening above the SH-1 hard line would cost more than +**0.123 $/response**. For three passes it was untestable because every one +opened below the line. + +| pass | SH-1 at open | $/response | +|---|---|---| +| CB-WP-0014 | above the 300,000 hard line (347,720) | **0.220** | +| CB-WP-0015 | below it (237,533, after a compaction) | **0.111** | + +Both fall on the predicted side, and the second is the **control the last +report said was missing**. That is a real strengthening — but it is n=2, +the two passes differ in tier, kind and subject, and the compaction that +supplied the control is also the thing that makes the passes different. +Consistent with the prediction; not yet a demonstration of it. + +**Meta budget: 26% `[OVER]`** the soft 25%, on CB-WP-0013's $8.26 sitting +in a trailing three with two product passes. No product work was displaced +and ADR-0006 D2 exempts instrument repair anyway. ## 6. Open @@ -198,4 +227,6 @@ pass that opens below the line to mean anything. been reproduced or instrumented.** Recorded because an intermittent positive control is worth more attention than a failing one, and because a flake in the tool that prices every pass would be quietly corrosive. +- **The self-quoting rule fixed the wrong boundary.** §5. Quoting two + passes back would be safe; the change is owed, not made here. - **Chaos: 10 of 12 declarations, 1 override.**