qi6: full-budget A/B for graded circulation-connectivity signal, negative result

conn_grade ON vs OFF (qpk protocol, experiments/run_qi6_ab.sh): harbor-house
(budget 2500, seeds 1-3) byte-identical output in every seed — the secondary
comparator key never fired. programme-house (budget 3000, seeds 1-5) 3/5 seeds
tie exactly; seeds 1/2 diverge to a different topology but the fail delta is
adjacency/crinkliness/width/access/size, never connectivity. Zero of 4 cases
where a not-connected fail was present got cleared by the grade.

Mechanism (b)/(c) (graded proximity as tertiary comparator key) is falsified,
not just unconfirmed. Kept default OFF (already was). Closed qi6; filed
homemaker-py-8sh for the remaining candidate (mechanism (a): an explicit
insert/relocate-circulation operator that doesn't depend on the search
stumbling onto a fail-count tie).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GDZjAATDWW1xFfc7xnJqSt

DESIGN.md

@@ -2415,7 +2415,7 @@ gated on the 9o5 landscape-flattening risk (§13 / `homemaker-py-xi7`) and its o
 Tests: `tests/test_collapse_global.py` ×6 (demand-set relabel, level hard constraint, c/o/s
 exclusion, no-op safety, keep-better/unmerged); 267 pass.
 
-## 18. Graded circulation-connectivity signal (`homemaker-py-qi6`) — in progress
+## 18. Graded circulation-connectivity signal (`homemaker-py-qi6`) — DONE (negative)
 
 **Motivation — the binary fail is flat.** After the §17 collapse, the residual fails on the
 harbor-house set are dominated by `level N not connected` (2 of the best layout's 12; also on
@@ -2458,13 +2458,39 @@ the gradient toward connected topologies.
 (env `HOMEMAKER_CONN_GRADE`, default OFF); the grade is read off the optimised tree, one extra
 native eval per child.
 
-**Status / next.** Signal, fitness wiring, CLI, and 9 tests landed (`tests/test_conn_grade.py`:
+**Build.** Signal, fitness wiring, CLI, and 9 tests landed (`tests/test_conn_grade.py`: pure-graph
-pure-graph fraction contract, non-circ cells ignored, monotone under (dis)connection, and the
+fraction contract, non-circ cells ignored, monotone under (dis)connection, and the score/fail-
-score/fail-count-invariance of the flag). The A/B — does the gradient actually pull evolve runs
+count-invariance of the flag). 276 tests pass.
-toward connected circulation and clear `not connected` fails — needs full-budget runs and is
+
-pending (short 60-eval smoke run confirms the plumbing only). If the graded key alone is
+**A/B verdict (measured, 2026-07-22, qpk protocol, `experiments/run_qi6_ab.sh`) — NEGATIVE.**
-insufficient, the follow-on is an insert/relocate-circulation mutation operator (mechanism (a),
+Equal-budget `conn_grade` ON vs OFF, both arms finished with the standard finish-time `--collapse`
-still `homemaker-py-qi6`) that now has a gradient to climb. 276 tests pass.
+(94g), 4 workers, canonical `homemaker-fitness` re-score for the `.fails` breakdown:
+
+- **harbor-house** (`init.dom`, budget 2500, seeds 1–3): **byte-identical output** in every seed
+  (dom, fail list, fitness all diff-clean ON vs OFF) — the secondary comparator key never fired,
+  i.e. the search trajectory never actually hit a tie at fail-count that the grade could break.
+  This is the programme §18 was motivated on (2 of 15 fails on the best layout are `not
+  connected`), and the signal moved nothing.
+- **programme-house** (`init.dom`, budget 3000, seeds 1–5): 3/5 seeds tie exactly (byte-identical
+  `.fails`); seeds 1 and 2 diverge to a **different topology** with one fewer total fail (8→7
+  each) — but the diff is entirely adjacency/crinkliness/width/access/size fails, not
+  connectivity. In all 4 seed-arms across both programmes where a `not connected` fail was
+  actually present (harbor 1&3, programme 3&4), the fail is **unchanged** in both arms — zero
+  cases of the grade clearing one.
+- **Conclusion: the grade does not do what §18 designed it to do.** It occasionally perturbs
+  tie-breaking among equal-fail-count neighbours (programme-house seeds 1/2), which can
+  incidentally shift the total fail count, but that perturbation never targets circulation
+  connectivity specifically — consistent with a comparator key that is either too weak relative
+  to the primary `(-n_fails, fitness)` keys to steer topology choice, or whose grade values are
+  rarely distinct enough between the actual neighbours the search compares to break a tie in the
+  intended direction.
+
+**Status / next.** Kept default OFF (already was). Mechanism (b) (graded proximity as a tertiary
+key) is falsified by this A/B, not just unconfirmed — do not re-attempt without a different
+mechanism. The remaining candidate from the original issue is mechanism (a): an explicit
+insert/relocate-circulation mutation/repair operator, which does not depend on the search
+stumbling onto a fail-count tie to act. Not started; low priority per DISCOVERED-FROM epic
+`homemaker-py-94g`'s framing (fitness fidelity, not search capability).
 
 ## 19. Geometry/topology repair for shape-intrinsic fails (`homemaker-py-7fm`) — DONE (negative)
 

experiments/run_qi6_ab.sh

@@ -0,0 +1,60 @@
+#!/usr/bin/env bash
+# qi6 A/B (DESIGN.md §18): does the graded circulation-connectivity signal
+# actually pull full-budget evolve runs toward connected circulation and clear
+# 'level N not connected' fails, or is the secondary-comparator gradient too
+# weak to matter (mirrors the qpk §20 protocol: equal-budget ON vs OFF, both
+# arms finished with the standard finish-time --collapse (94g) so the
+# comparison is apples-to-apples on the final collapsed score)?
+#
+# Authoritative metrics: total fail count and 'not connected' fail count, read
+# from the .fails file homemaker-fitness writes (evolve.py itself only prints
+# n_fails, not the fail list). Each run appends one TSV row so partial results
+# survive an interrupt.
+#
+# Usage: experiments/run_qi6_ab.sh
+set -u
+cd "$(dirname "$0")/.."
+
+WORKERS=4
+OUT=scratch/qi6_ab; mkdir -p "$OUT"
+TSV=scratch/qi6_ab_results.tsv
+[ -f "$TSV" ] || printf 'programme\tseed\tconn_grade\tbudget\tfails\tnot_connected\tfitness\telapsed_s\n' > "$TSV"
+
+run() {  # programme seed conn_grade(0|1) budget
+  local prog="$1" seed="$2" cg="$3" budget="$4"
+  local tag="cg${cg}"
+  local dom="$OUT/${prog}_${tag}_s${seed}.dom"
+  local log="$OUT/${prog}_${tag}_s${seed}.log"
+  local flag="--no-conn-grade"; [ "$cg" = 1 ] && flag="--conn-grade"
+  echo ">>> $prog seed=$seed conn_grade=$cg budget=$budget"
+  local t0; t0=$(date +%s)
+  homemaker-evolve "examples/$prog/init.dom" \
+    --budget "$budget" --workers "$WORKERS" --seed "$seed" \
+    $flag --output "$dom" > "$log" 2>&1
+  local t1; t1=$(date +%s)
+  local fitness fails notconn
+  fitness=$(sed -n 's/^best *: \([0-9.e+-]*\) .*/\1/p' "$log")
+  fails=$(sed -n 's/^best *: [0-9.e+-]* (\([0-9]*\) fails).*/\1/p' "$log")
+  # re-score with the canonical scorer to get the .fails breakdown (evolve.py
+  # only prints the count, not the list); dom lives outside examples/$prog so
+  # pass an absolute path while cd'd there for patterns.config resolution
+  ( cd "examples/$prog" && homemaker-fitness "$(realpath "../../$dom")" > /dev/null 2>&1 )
+  notconn=0
+  if [ -f "${dom}.fails" ]; then
+    notconn=$(grep -c 'not connected' "${dom}.fails")
+  fi
+  printf '%s\t%s\t%s\t%s\t%s\t%s\t%s\t%s\n' \
+    "$prog" "$seed" "$cg" "$budget" "${fails:-ERR}" "$notconn" "${fitness:-ERR}" "$((t1-t0))" >> "$TSV"
+  echo "    -> ${fails:-ERR} fails (${notconn} not-connected), fitness=${fitness:-ERR}, $((t1-t0))s"
+}
+
+# harbor-house: budget 2500, seeds 1-3 (qpk protocol)
+for seed in 1 2 3; do run harbor-house "$seed" 0 2500; done
+for seed in 1 2 3; do run harbor-house "$seed" 1 2500; done
+
+# programme-house: budget 3000, seeds 1-5 (qpk protocol)
+for seed in 1 2 3 4 5; do run programme-house "$seed" 0 3000; done
+for seed in 1 2 3 4 5; do run programme-house "$seed" 1 3000; done
+
+echo "=== qi6 conn_grade A/B complete ==="
+column -t -s $'\t' "$TSV"