feat(lens): deterministic rule-fit self-experiment (forward net-width + reverse starvation)

Operator's LLM-free rule-selection experiment, built on the substrate that actually
exists (verified first): rules live in data/pmu/lens-reef.json at domain granularity
(the SQLite lens_rules table is empty → perimeter always 0), and routeToDomain() places
any text deterministically (keyword/anchor, no qwen). For the last N commits it places
the ASK (message) and the WORK (changed paths + code), then measures Chebyshev lane
divergence against the lens's own fence:
  FORWARD net-width  = pulled a lane the commit didn't land in (aim too wide)
  REVERSE starvation = the landed lane's rules weren't fenced (should-have-applied miss)
No model grades anything — pure coordinate arithmetic (the S≡P≡H requirement). First run:
AIM 76% / divergence 24% over 26 commits; the diverged list names each ask→did lane
mismatch so the fix is the rule coord or the routing, never an LLM reminder.

Guard: tests/ops/lens-rule-fit.test.js (emits the numbers · deterministic · LLM-free · reef+Chebyshev).

Honest limit: at domain granularity net-width==starvation (one divergence event, two readings);
separating them needs per-rule SQLite coords (mirror reef→lens_rules), which would ALSO make
the perimeter fetch live — a follow-on that changes runtime injection, flagged not flipped.

Originating-Terminal: 🔨 iTerm Builder
Relevant-Rooms: 🧪 Cursor Laboratory, 🔨 iTerm2 Builder, 🎩 Kitty Operator
Story: Operator asked to confirm rules come from the SQLite metal before building; the
check found that table empty and the data feeds absent, so the harness was built on the
reef + deterministic placement that do exist — measuring, never grading.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013BUf5M5p42Zw73VaGt33ZL

commit 1ef810325 — 144×144 lattice triptych (on-chip)

◎ the ingest chose C,C1 — Operations × Operations·Grid — acting on C,C — Operations × Operations (grip 0.766) · the walk starts at the ACTOR's row · 2 file(s)

⏱ ingest 156.1ms (corpus→lattice) · definer walk 154ms (row→transpose→row, ballistic XOR per hop) · total 416ms
ingest here = commit-scoped SENSING only (this commit's message + changed files → lit anchors) — deep seed authoring lives in the reef-self-loop, off the commit path

rows = ACTOR (lens) · cols = PATIENT (object) · diagonal = self-reference; ◎ at actor-row × patient-col — every panel, same orientation. Heat above the diagonal = actors ranking ShortLex-earlier than their patients (the walk follows definers uphill in rank).
The 12 canonical lanes (rows top→bottom, cols left→right): A·Strategy · B·Tactics · C·Operations · A1·Law · A2·Goal · A3·Fund · B1·Speed · B2·Deal · B3·Signal · C1·Grid · C2·Loop · C3·Flow

PAIR LATTICE (the map we normally see)

PRE-WALK SENSE · INTENT grid — the snapshot, no point of view

the no-point-of-view snapshot — what the sensor lit before any walk; the leaf walk below is what bridges the two grids

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PRE-WALK SENSE · REALITY grid — the snapshot, no point of view

the no-point-of-view snapshot, reality side — same sensor law

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PRE-WALK Δ · straight comparison of the two ingested grids (green both · cyan intent-only · amber reality-only) · 5% overlap · ◎ = where the competence pixel sits in the raw overlap

good = green overlap without being a copy — near-100% is self-similarity, not alignment

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

◎ the pixel this commit belongs to

C,C1 — Operations × Operations·Grid (actor) acting on C,C — Operations × Operations (patient) · grip 0.766

actor seed C,C1 — Managing power flow across the infrastructure grid requires a daily operations run cadence and path topology. Power flows where the grid lets it. Choose connections before topologies; the topology is what the connections compose, and a connection added later is a topology silently rewritten. Latency is geography expressed in milliseconds. The grid IS the map of where compute meets storage meets bandwidth; the map is the territory at the scale the request actually traverses. Infrastructure decisions are reversible at ten-times the cost of being right the first time. Buy the irreversibility down before you provision the easy decisions, because the easy decisions ride on the irreversible ones. A connection between two systems is a contract that outlasts the systems on either end. Wire-protocols outlive the processes they connect; the wire is the substrate, the endpoints are the exhaust. Operations is the loop that runs whether anyone watches. The daily commit, the hourly hook, the per-request fingerprint — the substrate of the substrate, ambient and unwatched. Process is what you forget to do; ops is what does itself. Convert manual moves into ambient state, because the tax of remembering compounds faster than the tax of automating. The daily friction is the diff between your intent and your harness. Close the diff or pay the tax twice — once in the slip, once in the apology to the next agent. Operations earns its keep by being boring. The day the routine is interesting is the day it has already failed and is being narrated rather than executed.

patient seed C,C — Operations is the loop that runs whether anyone watches. The daily commit, the hourly hook, the per-request fingerprint — the substrate of the substrate, ambient and unwatched. Process is what you forget to do; ops is what does itself. Convert manual moves into ambient state, because the tax of remembering compounds faster than the tax of automating. The daily friction is the diff between your intent and your harness. Close the diff or pay the tax twice — once in the slip, once in the apology to the next agent. Operations earns its keep by being boring. The day the routine is interesting is the day it has already failed and is being narrated rather than executed.

INTENT · leaf-walk clouds · BLUE→VIOLET by ply (◎ = the competence pixel: actor∩patient)

good = the clouds concentrate where the commit SAYS it works

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

REALITY · leaf-walk clouds · ORANGE→RED by ply (◎ = the same pixel — the shared perspective)

good = the same clouds as intent — shipped where declared

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

DELTA · graded cloud divergence (green agree · magenta declared-not-done · amber done-not-declared) · 3% agree

good = mostly green; magenta = said-not-done (chase it in the code), amber = done-not-said (chase it in the docs)

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

TOLERANCE · cloud topology (dim green in-shape · hot amber bleed · red orthogonal concentration)

red only matters above the flip — a few amber is normal bleed

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

SHORTLEX-3 PROJECTION (ABC · A1..C3 · the 132 children — candidate, pre-ratchet)

PROJECTED INTENT · the commit’s claims, head→axis_i tail→axis_j (cyan)

the three-length view — good = claims land inside their own zone

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

PROJECTED REALITY · same projection law (amber)

same projection law — good = reality occupies the zones intent does

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

STRAIGHT COMPARISON · green both · cyan intent-only · amber reality-only · 4% overlap

good = the zones agree; cross-zone scatter is the drift to read

	ABC A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N A1–C3
ABC A1–C3 A1A–O A2A–O A3A–O B1A–O B2A–O B3A–O C1A–N C2A–N C3A–N

shortlex-3 zone occupancy (lit cells per diagonal square)
INTENT: 1 in the 3×3 ABC corner · 16 in the 9×9 A1..C3 square · 1512 in the 132×132 children square · 359 cross-zone
REALITY: 1 in the 3×3 ABC corner · 2 in the 9×9 A1..C3 square · 1080 in the 132×132 children square · 166 cross-zone
axes: identical 144 ShortLex names both sides (symmetric); weights asymmetric
zone 3 = candidate children, pre-ratchet (repo-derived dumps not yet past the perturbation-probe gate) · intent 47 claims @ θ 0.625 · reality 23 claims @ θ 0.625 · projection 256ms

SHAPE-MATCH σ = 3.98 — the real intent↔reality definer walk vs random-reality (actual 0.1832 vs random 0.079). >0 = the walk distinguishes real alignment from noise; aggregate over independent walks → the divergent series.
σ_drift · forming · separation is forming — the direction is real, the trust is not yet

Most out-of-lane row: B1 · Speed — that lane is drifting most (read the red band in that row).

Three-colour tolerance (out-of-lane = reality fires where intent is weak, in orthogonal blocks): green=in-lane agreement · amber=a few out-of-lane (tolerated) · red=armed — orthogonal out-of-lane 17% vs tolerance 25%.
468 green · 480 amber · 0 red · 597 XOR-friction cells.

Both axes = the 144 ShortLex anchors, cell=row⊕col, on the chip. INTENT=docs+rules · REALITY=code · DELTA=friction.

tile dump — what the SimHash matched on the diagonal (inspect the sensor; tap to expand)

A,A The lattice we build today is what next year's underwriter inherits.

intent 66% (no distinctive match)

reality 70% // REVERSE starvation: same divergence, read from reality's side — the landed lane's rules were

B,B Tactics is the choice of which leverage to pull at which beat.

intent 72% (no distinctive match)

reality 73% function pct(x) { return x == null ? 'n/a' : `${(x * 100).toFixed(1)}%`; }

C,C Operations is the loop that runs whether anyone watches.

intent 78% (no distinctive match)

reality 75% (no distinctive match)

A1,A1 EU AI Act Article 14 requires human-on-the-loop oversight, with the supervisor able to interrupt the system at

intent 83% (no distinctive match)

reality 81% !/usr/bin/env node. scripts/pmu/lens-rule-fit.mjs — THE DETERMINISTIC RULE-FIT SELF-EXPERIMENT

A2,A2 The target is the cell coordinate the system MUST occupy by end-of-quarter.

intent 73% (no distinctive match)

reality 75% (no distinctive match)

A3,A3 Every dollar builds either the floor (substrate that lifts future bets) or the ceiling (cap on the next round)

intent 69% (no distinctive match)

reality 75% // (At domain granularity net-width and starvation are the two readings of the same divergence

B1,B1 Latency is the price of a missed beat.

intent 70% (no distinctive match)

reality 72% (no distinctive match)

B2,B2 The deal sets the exchange rate between two value substrates.

intent 70% (no distinctive match)

reality 75% (no distinctive match)

B3,B3 Signal-to-noise is the inverse cube of bandwidth — narrow the channel, the message arrives; broaden it, recipi

intent 72% (no distinctive match)

reality 75% diverged: diverged.slice(0, 12).map((r) => ({ sha: r.sha, subject: r.subject, intent: r.intent,

C1,C1 Power flows where the grid lets it.

intent 73% (no distinctive match)

reality 73% function cheby(a, b) { return (a && b) ? Math.max(Math.abs(a[0] - b[0]), Math.abs(a[1] - b[1]))

C2,C2 Each iteration is a hypothesis tested against the ground truth of the world.

intent 69% (no distinctive match)

reality 72% const N = Number(arg('--n', '40'));

C3,C3 Flow is the rate at which committed work crosses the finish line.

intent 77% (no distinctive match)

reality 72% (no distinctive match)

commit context — 2 file(s)

feat(lens): deterministic rule-fit self-experiment (forward net-width + reverse starvation)

Operator's LLM-free rule-selection experiment, built on the substrate that actually
exists (verified first): rules live in data/pmu/lens-reef.json at domain granularity
(the SQLite lens_rules table is empty → perimeter always 0), and routeToDomain() places
any text deterministically (keyword/anchor, no qwen). For the last N commits it places
the ASK (message) and the WORK (changed paths + code), then measures Chebyshev lane
divergence against the lens's own fence:
  FORWARD net-width  = pulled a lane the commit didn't land in (aim too wide)
  REVERSE starvation = the landed lane's rules weren't fenced (should-have-applied miss)
No model grades anything — pure coordinate arithmetic (the S≡P≡H requirement). First run:
AIM 76% / divergence 24% over 26 commits; the diverged list names each ask→did lane
mismatch so the fix is the rule coord or the routing, never an LLM reminder.

Guard: tests/ops/lens-rule-fit.test.js (emits the numbers · deterministic · LLM-free · reef+Chebyshev).

Honest limit: at domain granularity net-width==starvation (one divergence event, two readings);
separating them ne

scripts/pmu/lens-rule-fit.mjs
tests/ops/lens-rule-fit.test.js

Produced on the chip by the running pipeline (scripts/pmu/pipeline.mjs → runPipeline → walk + xor), commit-scoped · on-chip per-walk 512.56ns (this commit). Recompute: node scripts/pmu/commit-triptych.mjs --commit 1ef810325

Commit artifacts (open on a computer — Gmail mobile won't follow file:// links; the inline images above are the phone-viewable form): reef grid · this triptych · the HDL — all three are also attached to this email.