Sovereign 8s.is mirror of standardgalactic/admissibility-lab
  • TeX 88.8%
  • Python 8.2%
  • HTML 1.6%
  • Shell 1.4%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
standardgalactic 3358fb479e Minor Revisions
2026-10-23 16:55:26 -03:00
adm_archaeology Theory Development 2026-06-13 20:16:55 -03:00
adm_core Theory Development 2026-06-13 20:16:55 -03:00
adm_generator Theory Development 2026-06-13 20:16:55 -03:00
adm_graph Theory Development 2026-06-13 20:16:55 -03:00
adm_metrics Theory Development 2026-06-13 20:16:55 -03:00
adm_repair Theory Development 2026-06-13 20:16:55 -03:00
adm_visual Theory Development 2026-06-13 20:16:55 -03:00
examples Theory Development 2026-06-13 20:16:55 -03:00
experiments Theory Development 2026-06-13 20:16:55 -03:00
processing Clean generated artifacts and compress audio 2026-09-21 07:02:18 -03:00
tests Theory Development 2026-06-13 20:16:55 -03:00
A_amnésia_da_IA_quebra_sistemas.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
A_amnésia_da_IA_quebra_sistemas.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
A_amnésia_da_IA_quebra_sistemas.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
admissible-degradation.pdf Theory Development 2026-09-18 21:50:13 -03:00
admissible-degradation.tex Theory Development 2026-09-18 21:50:13 -03:00
continuations_before_objects.pdf Theory Development 2026-06-13 20:16:55 -03:00
continuations_before_objects.tex Theory Development 2026-06-13 20:16:55 -03:00
corrections.log Minor Revisions 2026-10-23 16:55:26 -03:00
cost_of_future_admissibility.pdf Minor Revisions 2026-10-23 16:55:26 -03:00
cost_of_future_admissibility.tex Minor Revisions 2026-10-23 16:55:26 -03:00
eight_letter_keyboard.pdf Theory Development 2026-06-13 20:16:55 -03:00
eight_letter_keyboard.tex Theory Development 2026-06-13 20:16:55 -03:00
essay_recombine.py Corpus Recombination 2026-10-16 21:33:37 -03:00
experiment-results.txt Theory Development 2026-06-13 20:16:55 -03:00
fix_transcription_errors.sh Minor Revisions 2026-10-23 16:55:26 -03:00
Free_Cloud_Storage_and_AI_Synthesis.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
Free_Cloud_Storage_and_AI_Synthesis.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
Free_Cloud_Storage_and_AI_Synthesis.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
From_Corpus_to_Argument.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
From_Corpus_to_Argument.tex Corpus Recombination 2026-10-16 21:33:37 -03:00
How_Compressed_History_Breaks_AI_Safety.mp3 Minor Revisions 2026-10-23 16:55:26 -03:00
How_Compressed_History_Breaks_AI_Safety.txt Minor Revisions 2026-10-23 16:55:26 -03:00
How_Compressed_History_Breaks_AI_Safety.vtt Minor Revisions 2026-10-23 16:55:26 -03:00
How_Memory_Compression_Breaks_AI_Safety_Rules.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
How_Memory_Compression_Breaks_AI_Safety_Rules.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
How_Memory_Compression_Breaks_AI_Safety_Rules.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
How_motor_gestures_reconstruct_our_reality.mp3 Theory Development 2026-06-13 20:16:55 -03:00
How_motor_gestures_reconstruct_our_reality.txt Theory Development 2026-06-13 20:16:55 -03:00
How_motor_gestures_reconstruct_our_reality.vtt Theory Development 2026-06-13 20:16:55 -03:00
How_Systems_Decide_What_to_Sacrifice.mp3 Theory Development 2026-09-18 21:50:13 -03:00
How_Systems_Decide_What_to_Sacrifice.txt Theory Development 2026-09-18 21:50:13 -03:00
How_Systems_Decide_What_to_Sacrifice.vtt Theory Development 2026-09-18 21:50:13 -03:00
index.html Theory Development 2026-06-13 20:16:55 -03:00
memory-compression-visualizer.html Minor Revisions 2026-10-23 16:55:26 -03:00
Metered_Memory.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
Metered_Memory.tex Corpus Recombination 2026-10-16 21:33:37 -03:00
motor_phonology.pdf Theory Development 2026-06-13 20:16:55 -03:00
motor_phonology.tex Theory Development 2026-06-13 20:16:55 -03:00
Nineteen_Million_Possible_Essays.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
Nineteen_Million_Possible_Essays.tex Corpus Recombination 2026-10-16 21:33:37 -03:00
polish_future_admissibility.py Minor Revisions 2026-10-23 16:55:26 -03:00
Por_qué_la_IA_borra_su_memoria.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
Por_qué_la_IA_borra_su_memoria.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
Por_qué_la_IA_borra_su_memoria.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
process_audio.sh Corpus Recombination 2026-10-16 21:33:37 -03:00
Publication_Decision_CoC.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
Publication_Decision_CoC.tex Corpus Recombination 2026-10-16 21:33:37 -03:00
Publishing_an_Untested_Design.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
Publishing_an_Untested_Design.tex Corpus Recombination 2026-10-16 21:33:37 -03:00
Reachability_Geometry.pdf Theory Development 2026-06-13 20:16:55 -03:00
README.md Minor Revisions 2026-10-23 16:55:26 -03:00
recombination-seed-7.md Corpus Recombination 2026-10-16 21:33:37 -03:00
requirements.txt Theory Development 2026-06-13 20:16:55 -03:00
setup.py Theory Development 2026-06-13 20:16:55 -03:00
test-results.txt Theory Development 2026-06-13 20:16:55 -03:00
The_Combinatorial_Bottleneck.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
The_Combinatorial_Bottleneck.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
The_Combinatorial_Bottleneck.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
The_Eight_Letter_Motor_Phonology.pdf Theory Development 2026-06-13 20:16:55 -03:00
The_Flyxion_Architecture.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
The_hidden_motor_grammar_of_your_hands.mp3 Theory Development 2026-06-13 20:16:55 -03:00
The_hidden_motor_grammar_of_your_hands.txt Theory Development 2026-06-13 20:16:55 -03:00
The_hidden_motor_grammar_of_your_hands.vtt Theory Development 2026-06-13 20:16:55 -03:00
The_Limits_of_Automated_Ideation.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
The_Motor_Manifold.pdf Theory Development 2026-06-13 20:16:55 -03:00
The_Reconstruction_Imperative-extended.pdf Theory Development 2026-06-13 20:16:55 -03:00
unattempted_alternative.pdf Corpus Recombination 2026-10-16 21:33:37 -03:00
unattempted_alternative.tex Corpus Recombination 2026-10-16 21:33:37 -03:00
Undeviginti_milliones_textuum_et_mens_humana.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
Undeviginti_milliones_textuum_et_mens_humana.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
Undeviginti_milliones_textuum_et_mens_humana.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
When_AI_Code_Rejects_Human_Fixes.mp3 Corpus Recombination 2026-10-16 21:33:37 -03:00
When_AI_Code_Rejects_Human_Fixes.txt Corpus Recombination 2026-10-16 21:33:37 -03:00
When_AI_Code_Rejects_Human_Fixes.vtt Corpus Recombination 2026-10-16 21:33:37 -03:00
Why_Reachability_Matters_More_Than_Prediction.mp3 Theory Development 2026-06-13 20:16:55 -03:00
Why_Reachability_Matters_More_Than_Prediction.txt Theory Development 2026-06-13 20:16:55 -03:00
Why_Reachability_Matters_More_Than_Prediction.vtt Theory Development 2026-06-13 20:16:55 -03:00
Your_Keyboard_Is_a_Civilizational_Ark.mp3 Theory Development 2026-06-13 20:16:55 -03:00
Your_Keyboard_Is_a_Civilizational_Ark.txt Theory Development 2026-06-13 20:16:55 -03:00
Your_Keyboard_Is_a_Civilizational_Ark.vtt Theory Development 2026-06-13 20:16:55 -03:00

Admissibility Lab

From Corpus to Argument

Nineteen Million Possible Essays

Metered Memory

The Unattempted Alternative

The Cost of Future Admissibility

Publishing an Untested Design

Publication Decision CoC

Admissible Degradation

Intelligence as Perspectival Convergence

The Eight-Letter Keyboard

Motor Phonology and Symbolic Reachability

Continuations Before Objects

Audio Overviews

A Python laboratory for observing admissibility geometry.

This is not a visualization of existing theorems. It is an instrument for generating observations — a place where you discover properties of the theory you did not anticipate.


What this is

The Admissibility Program has produced extensive mathematical ontology. What it currently lacks is a laboratory: a place where the theory is pressured to produce definitions by being repeatedly observed.

adm-lab is built on the insight that the most important concepts in any framework tend to be discovered archaeologically — you run experiments, recurring structures appear, and only afterward do you prove theorems about them.

The generative kernel is:

State → AdmissibilitySet

A state is not a record. It is a locus from which continuations project. An admissibility set is not a static list. It is the live set of what can come next, under current constraints.

Everything else follows from watching this structure respond to damage, constraint accumulation, repair, and traversal.


Project structure

adm-lab/
├── adm_core/           # Irreducible primitives
│   └── primitives.py   # State, Continuation, Constraint, Trajectory, AdmissibilitySet
├── adm_graph/          # Continuation graphs, reachability, fiber computation
├── adm_repair/         # Repair simulator and strategies
├── adm_visual/         # Terminal and matplotlib rendering
├── experiments/        # Executable thought experiments
│   ├── factories.py            # Canonical graph factories
│   ├── exp01_repair_corridor.py
│   ├── exp02_sink_formation.py
│   ├── exp03_trajectory_archaeology.py
│   └── run_all.py
└── tests/
    └── test_core.py

Installation

pip install matplotlib networkx   # optional, for plot rendering

No other dependencies. The core and repair modules use only the standard library.


Quick start

from adm_core import State, Continuation, Constraint
from adm_graph import AdmissibilityGraph
from adm_visual import TerminalRenderer

# Build a small graph
g = AdmissibilityGraph(label="demo")
states = [State(label=f"s{i}") for i in range(5)]
for i in range(4):
    g.add_continuation(Continuation(states[i], states[i+1]))
# Add a branch
g.add_continuation(Continuation(states[2], states[4]))

# Inspect
renderer = TerminalRenderer()
renderer.render_graph_summary(g)
renderer.render_fan_out_map(g)

# Damage and observe
record = g.damage_random(fraction=0.4, rng_seed=42)
print("After damage:")
renderer.render_graph_summary(g)

# Repair
record.restore()
print("After repair:")
renderer.render_graph_summary(g)

Running experiments

# All experiments
python -m experiments.run_all

# Individual
python -m experiments.exp01_repair_corridor
python -m experiments.exp02_sink_formation
python -m experiments.exp03_trajectory_archaeology

# Tests
python -m tests.test_core

Core concepts

State

A position in the space of distinctions. Identity is by id, not content: two states with identical content that arose through different histories are distinct. This models irreversibility as an architectural principle.

Continuation

A possible next state. Not a stored record — a potential. The framework is forward-facing throughout.

Constraint

A predicate over continuations. Constraints narrow admissibility sets. Projection is not in adm_core — it emerges one level up when constraints destroy distinctions, collapsing multiple states into one.

AdmissibilitySet

The live set of continuations available from a state under current constraints. An empty set is an admissibility sink — the formal signature of agency collapse.

Trajectory

A sequence of states with full admissibility history. The key diagnostic object: you can look back and ask when fan-out narrowed, when sinks appeared, and where bottlenecks were.

Repair

Not error-correction (restore prior state) nor exception-handling (branch to alternative). Repair is ontologically primary: the system is always already in a repair-eligible condition. The repair simulator runs damage+recovery experiments to discover structural invariants that the current mathematics does not yet describe.


Phenomena to look for

Running experiments, you may observe:

  • Continuation bottlenecks — states where fan-out drops to 1 before recovering
  • Admissibility sinks — states from which no continuation survives constraints
  • Repair corridors — paths that repair consistently routes through
  • Projection singularities — points where many states collapse to one image
  • Agency collapse regions — zones where navigability drops to near-zero
  • Lamphron ridges — gradient structures in the reachability field (adm.field, forthcoming)

The goal is that these become important because they keep appearing, not because they were posited in advance.


Relation to the broader Admissibility Program

adm-lab is the laboratory layer of a larger ecosystem:

Module Status Description
adm.core ✓ this repo Primitives: State, Continuation, Constraint, Trajectory
adm.graph ✓ this repo Continuation graphs, reachability, fibers
adm.repair ✓ this repo Repair simulator and strategies
adm.visual ✓ this repo Terminal and matplotlib rendering
adm.field planned RSVP field simulations, lamphron dynamics
adm.path planned Continuation-preserving pathfinding, geodesics
adm.memory planned MEM
adm.social planned Agency collapse in social systems
adm.learn planned Balbach–Zeugmann learning experiments

The mathematical foundations are documented in the RSVP-Monograph and associated papers. adm-lab exists to pressure those foundations by making them runnable.


Philosophy

The visualizer would pressure the theory to produce definitions.

Right now many concepts exist because they are mathematically plausible. A phenomenological instrument forces them to exist because they are repeatedly observed. A continuation bottleneck becomes important because it keeps appearing in experiments. A repair corridor becomes important because systems keep healing through the same structures.

The theory becomes less deductive and more archaeological. You start uncovering recurring objects in admissibility space. Only afterward do you prove theorems about them.


Flyxion / Admissibility Program