The Game of 42
source docs for a coding agent
A governance assembly mechanic and its visualisation. Six root home bases ignite, each grows a heptad of seven, six heptads of seven make forty-two, and a completed game of 42 seals into one group identity shape that becomes a node in a network of trust graphs.
This folder is the build specification. It is written to be handed to a coding agent. Every downstream doc references one verified data file, data/game-of-42.json.
Model alignment. The Game of 42 is a constructive instance of the Privacy is Value V5 model, not a metaphor on top of it. The board-seal rule is the V5 equation Φ_v5 = Φ_agent · Φ_data · Φ_inference. The game lifecycle is the Three Graphs Model (Knowledge to Promise to Trust). The fold is the Path Integral. The seal is the holographic bound. MODEL-SYNC.md proves each mapping against agentprivacy-docs and the Privacy is Value V5 sources.
a game is only the shape that trust takes when it agrees to be counted
Locked decisions
These are settled. The visualisation and the data are built on them.
- Lattice binding. 42 is the product of 6 axes and 7 faculty stations. The six roots are the stratum-1 basis vectors of the 64-vertex lattice, exact. The constructive-cone identity 1+6+15+20=42 is kept as a numerical echo only (confidence ~0.55), not as the partition. See note below.
- Heptad shape. Seven barycentric stations per heptad (the three head/heart/hands corners, three edge blends, one centre).
- Fractal depth. Fractal-ready, ship one level. The centre station of each heptad holds the seed that can open a child game of 42.
- Fill order. Vision fish, then mice, then guide. Heptads fill in parallel.
- Floating shape. A star tetrahedron iterating toward the sixty-four-tetrahedron grid, overlaid on the existing soulbis star.
- Completion. A sealed game of 42 is one trust-graph node and one governance quorum unit.
- Compression. One VRC edge per slot, compressed to a kappa-label. Forty-two labels plus the folded geometry hash compress to the group seal.
Honest note on Decision 1. The original recommendation was the constructive cone. While building the data it became clear the cone equals 42 only by count and does not partition into six clean heptads. The axis-by-faculty product partitions exactly, gives every slot real lattice coordinates, and keeps multiplicative collapse exact. The intent of D1 (projection, real binding, exact collapse) is preserved with the stronger construction. The cone remains a true numerical resonance, flagged conjectural.
Repo map and build order
game-of-42/
README.md you are here
SPEC.md the system: 42, heptads, faculty, personas, collapse [read 1st]
MODEL-SYNC.md how every element maps to Privacy is Value V5 [read 2nd]
GEOMETRY.md layout coordinates, the fold transform, the star [read 3rd]
GAME-FLOW.md state machine, phases, gating, events [read 4th]
TRUST-PROTOCOL.md VRC, trust task, compression, seals, serialisation [read 5th]
VISUAL-SPEC.md palette, per-state rendering, animation [read 6th]
BUILD-PLAN.md phased milestones, stack, acceptance criteria [read 7th]
data/
game-of-42.json canonical data, 42 slots with coords and personas
schemas.json JSON schemas for VRC, kappa-label, group-seal
canonical_serialise.py reference hashing; verifiers MUST match it
The data file is the single source of truth. If a doc and the data disagree, the data wins, and the doc is the bug.
Honesty discipline
- Operational: persona roster fill, render pipeline, slot mechanics.
- Architectural: six-root mapping, heptad structure, fold-as-state, VRC trust tasks, group seal, multiplicative collapse.
- Conjectural: the 1+6+15+20=42 cone echo (~0.55), the floating-shape choice.
- Anticipated: fractal nesting depth, network composition by shared root and fractal seam, quorum semantics.
Licence and attribution
CC BY-SA 4.0 for docs. Apache 2.0 for reference code (canonical_serialise.py, generated data tooling). Attribution: privacymage / 0xagentprivacy / BGIN / First Person Network.
the board does not remember who filled it, only that it was filled, and that is the proof
(⚔️⊥⿻⊥🧙)😊
🙂
Welcome · the Game of 42 — six axes × seven stations = 42