feat(lab): measure RAVNOVES00 R1 quality baseline

docs/16_ARCHITECTURE_AUDIT_EXECUTION_ROADMAP.md

@@ -239,6 +239,15 @@ mount identity is bound to that source session. A9 transfer therefore remains
 unexecuted without publishing an empty LAB. ADR 0029 retains the audit.
 ADR 0030 now defers TEST007/E36 and makes RAVNOVES00 source-scoped quality and
 product closure the active critical path.
+R0 is now closed by E37 immutable result
+`e37-ravnoves-acceptance-01b1efd586f747341c712d82f0907b39436a6f91ae92b1dfae987eca05fd8344`.
+The first R1 baseline is E38 immutable Worker 006 result
+`e38-perception-baseline-a272f82988cd9a7e071fad94c3e9fb49daf804fdcca523f853445fd3113a62b1`.
+Its sealed 146-item validation measures presence at 82.2%, geometry
+association at 81.5% and freshness at 95.2%, with 100% accounting, zero
+false-free claims and 14 high-severity failures. R1 therefore remains open;
+the next bounded iteration targets detector/background presence and
+geometry-only association without changing the E37 validation set.
 
 - [x] Reproduce all 4,489 immutable E29 frames with the exact frozen profile
   before applying E31/E30 changes.
@@ -256,7 +265,7 @@ product closure the active critical path.
   contract without creating an empty LAB.
 - [x] Defer TEST007, E36 and new collection until the RAVNOVES00 reference
   release candidate is accepted.
-- [ ] Freeze the RAVNOVES00 task ontology, reviewed denominator and
+- [x] Freeze the RAVNOVES00 task ontology, reviewed denominator and
   development/validation split.
 - [ ] Reach `>= 90%` separately for presence, geometry association and
   freshness decisions while retaining complete accounting and zero false-free

docs/20_RAVNOVES00_REFERENCE_SOURCE_PRODUCT_PLAN.md

@@ -61,11 +61,28 @@ failures remain blocking even when an aggregate percentage passes.
 
 R0–R4 are the active path. R5 is intentionally deferred.
 
+## Current evidence
+
+R0 is closed by immutable result
+`e37-ravnoves-acceptance-01b1efd586f747341c712d82f0907b39436a6f91ae92b1dfae987eca05fd8344`.
+It freezes 486 reviewed RAVNOVES00 items as 340 development and 146 sealed
+validation cases, with separate presence, geometry-association and freshness
+references.
+
+The first R1 measurement is immutable result
+`e38-perception-baseline-a272f82988cd9a7e071fad94c3e9fb49daf804fdcca523f853445fd3113a62b1`.
+It was trained only on the development split and executed on Worker 006.
+Freshness passes at 95.2%; evidence accounting is 100% and false-free claims
+remain zero. Presence is 82.2%, geometry association is 81.5% and 14
+high-severity validation cases fail, so R1 is not accepted. The next R1
+iteration is bounded to detector/background presence, geometry-only
+association and the camera-only freshness tail without changing validation.
+
 ## Experiment rules
 
 1. RAVNOVES00 remains immutable and is never relabelled as a LAB result.
 2. Every iteration creates a new LAB identity, method profile and conclusion.
-3. Existing E29–E35 results remain historical evidence and are not overwritten.
+3. Existing LAB results remain historical evidence and are not overwritten.
 4. Metrics are computed from server-owned artifacts; the UI does not invent
    acceptance state.
 5. The compact LAB summary explains the task, method, relevant models,