feat(pmu): compression PICKS the word — the provenance token that makes the number mean something

Operator 2026-07-05: 'compression should pick a word that most clearly fits that position — the
intersection of the sub-categories at those coordinates. If we don't, it might just be a number, and it
can't be a vibe with a coordinate stapled on. It has to be clear all the way down.' This closes the
provenance chain: σ/off-lane% (the number) → the coordinate the walk placed it at → its axis names
(Strategy.Law × Tactics.Signal) → THIS WORD (the token compression says most distinguishes that cell's
corpus) → the snippet it was mined from. Every step recomputable; nothing invented.

scripts/pmu/anchor-words.mjs: for each of the 144 cells, score every candidate token by TF × IDF where
IDF = the surprisal a compressor pays across the lattice (log(144/df)) — high = frequent HERE, rare
ELSEWHERE = the intersection word. Axis labels excluded so it drops to the concept BENEATH the label:
Tactics × Law → 'regulation', Strategy × Grid → 'topology', Speed × Law → 'constrained', Signal × Law
→ 'message'. Deterministic (alphabetical tie-break; same corpus → same word), LLM-FREE. Written to
data/pmu/shortlex-words.json (144), wired into coordName so every encircled coordinate reads
'Strategy.Law × Tactics.Signal · regulation'. Guards: shortlex-names (word appended, deterministic, no
model). Honest: individual picks are rough but stable in rank — the naming averages out across the lattice.

Originating-Terminal: 🔨 iTerm Builder
Relevant-Rooms: 🔨 builder, 🧪 laboratory, ☕ network
Story: The number is no longer a bare coordinate — compression names the meaning under it, clear all the way down, with no model in the chain.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013BUf5M5p42Zw73VaGt33ZL

commit 4faf82be0 — 144×144 lattice triptych (on-chip)

◎ the ingest chose A2,C2 — Strategy·Goal × Operations·Loop — acting on A1,A1 — Strategy·Law × Strategy·Law (grip 0.766) · the walk starts at the ACTOR's row · 4 file(s)

⏱ ingest 469.4ms (corpus→lattice) · definer walk 1114ms (row→transpose→row, ballistic XOR per hop) · total 1929ms
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

A2,C2 — Strategy·Goal × Operations·Loop (actor) acting on A1,A1 — Strategy·Law × Strategy·Law (patient) · grip 0.766

actor seed A2,C2 — Quarterly goal targets define the objectives that the iterative feedback loop and hypothesis test cycle measure. Each iteration is a hypothesis tested against the ground truth of the world. Loops shorter than the feedback half-life waste the half-life; loops longer than it pay compounding interest on the wait. Cycles compound when each turn's exhaust feeds the next turn's intake. Closed loops compound; open loops decay. The closing of the loop IS the productivity, not the running of it. Iteration discipline: one commit per iteration, never bundled. Bundled iterations destroy the bisect, and the bisect is the future agent's only tool for finding which round of feedback produced which artifact. The target is the cell coordinate the system MUST occupy by end-of-quarter. Anything that does not drive toward that coordinate is exhaust, and exhaust does not become substrate by being labelled progress. Vision is the coordinate the lattice rotates around — the fixed bearing every iteration is measured against. Without the fixed point, motion is movement; with the fixed point, motion is convergence. Define the goal as a verifiable predicate, not a feeling. The system terminates when the predicate evaluates true, and the predicate is what the next agent inherits when this one hands off.

patient seed A1,A1 — EU AI Act Article 14 requires human-on-the-loop oversight, with the supervisor able to interrupt the system at any moment. Compliance attaches to the artifact the artifact cannot escape — provenance is a regulatory primitive, not a feature. Constraints precede objectives. The cap-table, the license text, the consent envelope — each draws the inviolate line the action must respect, and the line is enforced before the optimization runs. A rule names what the system MAY NOT do. If the agent can do X, then 'forbid X' is not the constraint — it is wishful thinking dressed as policy. Constraints live in the gate, not in the prompt. Regulation is not advisory. The fine is the price tag stapled to the omitted control; insurance is the option pricing on the failure-mode the regulation already named.

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) · ⚠ TOO MANY

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 · 6% 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: 0 in the 3×3 ABC corner · 5 in the 9×9 A1..C3 square · 1559 in the 132×132 children square · 278 cross-zone
REALITY: 0 in the 3×3 ABC corner · 1 in the 9×9 A1..C3 square · 1075 in the 132×132 children square · 121 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 41 claims @ θ 0.625 · reality 34 claims @ θ 0.625 · projection 1036ms

SHAPE-MATCH σ = -2.37 — the real intent↔reality definer walk vs random-reality (actual 0.049 vs random 0.1). >0 = the walk distinguishes real alignment from noise; aggregate over independent walks → the divergent series.
σ_drift · noise · below zero — this read is noise, not a measurement

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=FIRED — orthogonal out-of-lane 30% vs tolerance 25%.
648 green · 575 amber · 524 red · 160 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 70% (no distinctive match)

reality 77% This change operates in the blog content domain, involving section anchor og encircled needs. T

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

intent 70% (no distinctive match)

reality 70% const words = pickAnchorWords();

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

intent 78% (no distinctive match)

reality 78% (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 77% (no distinctive match)

reality 78% (no distinctive match)

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

intent 75% (no distinctive match)

reality 81% process.stdout.write(`✓ wrote ${Object.keys(words).length} anchor words → ${OUT.replace(REPO +

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

intent 67% (no distinctive match)

reality 77% // deterministic order: sort candidates so ties resolve alphabetically → same input, same word

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

intent 69% (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 78% (no distinctive match)

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

intent 67% (no distinctive match)

reality 72% // Every step is recomputable; nothing is invented.

C1,C1 Power flows where the grid lets it.

intent 77% (no distinctive match)

reality 77% if (score > bestScore) { bestScore = score; best = w; }

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

intent 69% (no distinctive match)

reality 75% (no distinctive match)

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

intent 77% (no distinctive match)

reality 77% (no distinctive match)

commit context — 4 file(s)

feat(pmu): compression PICKS the word — the provenance token that makes the number mean something

Operator 2026-07-05: 'compression should pick a word that most clearly fits that position — the
intersection of the sub-categories at those coordinates. If we don't, it might just be a number, and it
can't be a vibe with a coordinate stapled on. It has to be clear all the way down.' This closes the
provenance chain: σ/off-lane% (the number) → the coordinate the walk placed it at → its axis names
(Strategy.Law × Tactics.Signal) → THIS WORD (the token compression says most distinguishes that cell's
corpus) → the snippet it was mined from. Every step recomputable; nothing invented.

scripts/pmu/anchor-words.mjs: for each of the 144 cells, score every candidate token by TF × IDF where
IDF = the surprisal a compressor pays across the lattice (log(144/df)) — high = frequent HERE, rare
ELSEWHERE = the intersection word. Axis labels excluded so it drops to the concept BENEATH the label:
Tactics × Law → 'regulation', Strategy × Grid → 'topology', Speed × Law → 'constrained', Signal × Law
→ 'message'. Deterministic (alphabetical tie-break; same corpus → same word), LLM-FREE. Written to
data/pmu

data/pmu/shortlex-words.json
scripts/pmu/anchor-words.mjs
scripts/pmu/shortlex-names.mjs
tests/pmu-simulator/shortlex-names.test.mjs

Produced on the chip by the running pipeline (scripts/pmu/pipeline.mjs → runPipeline → walk + xor), commit-scoped · on-chip per-walk 868.55ns (this commit). Recompute: node scripts/pmu/commit-triptych.mjs --commit 4faf82be0

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.