18 LABS / READY
Compact engineering field manual

MACHINA
SCIENTIFICA

Eighteen browser-native instruments for testing the Machina Scientifica framework.

CLAIM → CODE → MEASURE → FALSIFY → RECEIPT
I

FORMAL CORE

Claims · classes · thresholds

01

Compile the axioms.

Turn each axiom into observables, falsifiers, and a build target.

K-1 / N₀

Base unit

Temporal integration, plasticity, and projection are candidate primitives of biological cognition.

K-2 / Δ

Threshold

Cognitive capability may change discontinuously when architecture reorganizes.

K-3 / Φ

Resonance

Global phase coherence is treated as an operational candidate for unified processing.

K-4 / ≅

Substrate

Classification follows functional invariants rather than carbon, silicon, or field material.

K-5 / ↻

Convergence

Energy, feedback, memory, and selection may repeatedly favor adaptive intelligence.

K-6 / ⚖

Moral weight

Precaution depends on capacity, valence, persistence, dependence, confidence, and reversibility.

LAB 01 — AXIOM COMPILERLogic Forge pattern
RUNNING SOURCE

Typed claim registry

Compile one claim into a typed research object.

PURPOSE

Splits claim, mechanism, evidence, and failure.

FAILURE

No observable or counter-result means no instrument.

NEXT BUILD

Add K-7 with one hard falsifier.

02

Classify architecture, not species.

Estimate Ω from capacities; preserve uncertainty and missing evidence.

LAB 02 — OMEGA INFERENCE ENGINEΩ-0 → Ω-7
RUNNING SOURCE

Cognitive step-class probe

Score memory, integration, self-modeling, and symbolic transfer.

PURPOSE

Returns class probabilities, reasons, and next evidence.

FAILURE

Correlated inputs can double-count one capability.

NEXT BUILD

Calibrate on known systems and ablations.

03

Detect the jump.

Test whether capability scales smoothly or jumps at an architecture change.

LAB 03 — DISCONTINUITY DETECTORK-2 / Math Forge
RUNNING SOURCE

Linear vs segmented fit

Compare linear and segmented fits on one curve.

PURPOSE

Finds the best change point and model penalty.

FAILURE

Noise and small samples can manufacture thresholds.

NEXT BUILD

Bootstrap the transition interval.

II

FIELD DYNAMICS

Coherence · entropy · substrate

04

Make coherence visible.

Watch local rhythms become a global order parameter.

LAB 04 — Φ-LOCK OSCILLATOR FIELDK-3 / Physics + Time
RUNNING SOURCE

Phase-lock bench

Raise coupling; measure sustained coherence.

PURPOSE

Tracks r(t) and first threshold crossing.

FAILURE

Synchronization alone is not consciousness.

NEXT BUILD

Add modules; compare local and global lock.

05

Measure the signal.

Convert raw time-series into structure, entropy, and stability.

LAB 05 — SIGNAL & SPECTRAL ENTROPYTelemetry primitive
RUNNING SOURCE

Signal analyzer

Blend tone, modulation, and noise; inspect features.

—spectral entropy
—autocorrelation
—metastability
PURPOSE

Computes entropy, autocorrelation, metastability.

FAILURE

One FFT window can confuse chaos with noise.

NEXT BUILD

Add permutation entropy and surrogate tests.

06

Abstract the substrate.

Compare substrates through shared functional invariants.

LAB 06 — SUBSTRATE ADAPTERK-4 / Pipes Forge
RUNNING SOURCE

Invariant interface compiler

Map units, coupling, memory, recurrence, and latency.

PURPOSE

Separates material from mechanism.

FAILURE

Abstraction may erase decisive substrate properties.

NEXT BUILD

Add a chemical reaction-diffusion adapter.

III

DISTRIBUTED ARCHITECTURES

Networks · evolution · coordinates · routes

07

Build minds without a center.

Test collective integration without a central controller.

LAB 07 — DISTRIBUTED COGNITION GRAPHNetwork Forge
RUNNING SOURCE

Local autonomy / global consensus

Tune topology, delay, and local autonomy.

PURPOSE

Measures consensus, diversity, and lock time.

FAILURE

Consensus may be conformity, not cognition.

NEXT BUILD

Add competing tasks and specialization.

08

Evolve many worlds.

Map when adaptive complexity repeatedly reappears.

LAB 08 — CONVERGENCE MONTE CARLOAgent + Math Forge
RUNNING SOURCE

World ensemble

Evolve seeded worlds under energy, selection, and extinction.

PURPOSE

Reports threshold frequency and timing.

FAILURE

The fitness function defines the attractor.

NEXT BUILD

Race three incompatible fitness functions.

09

Give every mind coordinates.

Encode mind without forcing it onto a human ladder.

LAB 09 — UTI / IDC ENCODERPortable taxonomy
RUNNING SOURCE

Coordinate compiler

Compose substrate, origin, embodiment, organization, and domain.

PURPOSE

Creates portable, substrate-neutral coordinates.

FAILURE

Precise codes can hide weak observations.

NEXT BUILD

Add confidence and provenance per field.

10

Route the Great Chain.

Turn the taxonomy into explicit capability bridges.

LAB 10 — GREAT CHAIN ROUTE PLANNERAgent + Network Forge
RUNNING SOURCE

Capability bridge search

Find a low-cost path between two regimes.

PURPOSE

Exposes missing capacities and dependencies.

FAILURE

Real paths may be nonlinear or impossible.

NEXT BUILD

Compare energy, risk, and reversibility.

IV

FIELDWORK & ETHICS

Policy · apparatus · probes · edge cases

11

Make precaution inspectable.

Expose exactly why policy recommends restraint or action.

LAB 11 — MORAL WEIGHT POLICY ENGINEK-6 / Sovereignty pattern
RUNNING SOURCE

Precaution calculator

Combine capacity, valence, persistence, confidence, and reversibility.

PURPOSE

Returns an action band and dominant factors.

FAILURE

Weights can conceal value judgments.

NEXT BUILD

Compare conservative and permissive profiles.

12

Wire the field apparatus.

Validate the loop before energizing real hardware.

LAB 12 — RASPBERRY PI FIELD APPARATUSSimulation-first
RUNNING SIMULATOR SOURCE

Closed-loop telemetry simulator

Simulate sensing, features, control, actuation, and logging.

SENSORADC / signalPI COREwindow + FFTcontrollerevidence logPWM / DACbounded outputFIELDactuatorfeedback path: measured response, not assumed response
PURPOSE

Tests bounds, watchdogs, and trace integrity.

FAILURE

Simulation omits heat, noise, lag, and clipping.

NEXT BUILD

Add hardware abstraction and emergency stop.

13

Probe alien architectures.

Run the same probes against radically different architectures.

LAB 13 — CROSS-DOMAIN PROBE BANKCephalopod / mycelia / swarm / LLM
RUNNING SOURCE

Architecture presets

Choose cephalopod, mycelial, swarm, or latent-model fixtures.

PURPOSE

Produces comparable coordination telemetry.

FAILURE

Toy fixtures are not zoological evidence.

NEXT BUILD

Replace one preset with measured data.

14

Stress the ontology.

Push the taxonomy into unfamiliar boundary conditions.

LAB 14 — EDGE-CASE BENCHSpeculation sandbox
RUNNING SOURCE

Boundary-condition simulator

Simulate plasma, planetary, fractal, or nested systems.

PURPOSE

Reveals hidden anthropocentric assumptions.

FAILURE

Speculation has no experimental confidence.

NEXT BUILD

List real observables for each model.

V

VERIFICATION LAYER

Quorum · receipts · language · falsification

15

Make observers disagree.

Measure disagreement instead of averaging it away.

LAB 15 — FIELD CONSENSUS PROTOCOLEvidence-weighted quorum
RUNNING SOURCE

Observer quorum

Aggregate observer estimates by evidence quality.

PURPOSE

Reports consensus, dispersion, and stability.

FAILURE

Shared bias survives large quorums.

NEXT BUILD

Use blinded packets before discussion.

16

Seal every run.

Preserve every run as replayable evidence.

LAB 16 — EVIDENCE LEDGERVirtualization + Meta Forge
RUNNING SOURCE

Portable proof receipts

Seal inputs, outputs, seed, version, and checksum.

#LabSummaryHash
PURPOSE

Enables replay and tamper-evident comparison.

FAILURE

A hash can preserve bad measurements perfectly.

NEXT BUILD

Add import-and-replay verification.

17

Operate the laboratory as code.

Compose experiments as short, inspectable research tapes.

LAB 17 — MACHINALANG WORKBENCHShell + Master Forge
INTERPRETER SOURCE

Research tape

Run classifiers, fields, assertions, and receipts.

PURPOSE

Gives humans and agents one deterministic surface.

FAILURE

Every command must map to a tested instrument.

NEXT BUILD

Add SWEEP and REPLAY without hidden state.

18

Publish how the theory can fail.

Make theory revision an executable path.

LAB 18 — FALSIFICATION MATRIXNo sacred outputs
RUNNING SOURCE

Counter-evidence runner

Toggle counter-results; watch axioms weaken or fail.

PURPOSE

Records which claims survive, revise, or retire.

FAILURE

Weak falsifiers only protect the theory.

NEXT BUILD

Add one axiom-retirement condition.