The spread between top and bottom is a speed limit on compound human growth.
Same artifact, routed differently: 26x difference in systemic value. Controlled demo:
floor_lift 67.05 (bottom-third routing) vs 2.5 (premium routing) — identical 100-unit benefit.
Metric: floor_lift = Σ (benefit_i × (1 - percentile_i/100)²) — graded by routing, not volume.
Live infrastructure: tier-0 deterministic gates → tier-4 big-brain (Qwen 72B) under a $0.50 hard
spend cap, CONFIRM=1 human gate, lessons.db sha-chained mistake→solution memory.
Physical infrastructure + the computational swarm that serves it.
η = useful_joules / human_joules Every cycle must close on real thermal, material, or food yield.
| Proof | Ledger | Publication | Quality |
|---|---|---|---|
 |
 |
 |
 |
 |
 |
 |
 |
| If you want... | Click here | Why |
|---|---|---|
| Plain-language intro | START-HERE.md | No jargon — credit, energy, what to do this week |
| Full thesis | THESIS.md | The complete thermodynamic argument |
| Hardware builds | aerocement/ | Volumetric blackbody concrete recipes |
| Talent alignment | TALENT-ALIGNMENT-PROMPT.md | Map ANY skill to the Four Engines |
| Community standards | CODE_OF_CONDUCT.md | How we treat each other |
| Engine | Purpose | Live Components |
|---|---|---|
| Knowledge | Axioms, postulates, governance | 7 physics axioms, fractal constitution |
| Energy | Passive solar-thermal, storage | Black Locust coppice + RMH, H-003 thermal cascade |
| Material | Shelter, water, food | Aerated GFRC panels, ferrocement domes, aquaponics |
| Finance | Credit-building, ACRE token | PRF-001 routing, PoWr minting, thermodynamic ledger |
Permaculture principle: Each engine serves multiple functions. Each node generates surplus. Nothing extracted, nothing wasted.
The spread between top and bottom is a speed limit on compound human growth.
Utility of a delivered benefit scales with (1 − recipient_percentile)² — the same artifact routed to the bottom decile carries ~25x the systemic value weight of routing it to the top decile. Routing beats volume.
Verified demo (contribution_tier_v2.py, ids d1579005 vs 50c984a5): floor_lift 67.05 bottom-routed vs 2.5 premium-routed — identical 100 units of aggregate benefit, 26x systemic value gap.
| Tool | Function |
|---|---|
bin/contribution_tier_v2.py |
Bottom-floor weighted grading, FLOOR_LIFT as primary metric |
bin/openrouter_client_v1.py |
Live API tier routing under a $0.50 hard spend cap |
bin/tier_dispatch_v1.py |
5-tier escalator, hash-idempotent queue |
Release: v2026.09.20-floorlift · Milestone 6 open · Quadratic exponent is a falsifiable hypothesis (agape_cascade validation pending).
- Volumetric blackbody concrete — 95%+ solar absorption
- Passive stack-effect circulation — no pumps
- Subterranean thermal storage — 35°F cooling from 120°F inlet
- Target: 12.91 kWh/m² nightly capture (validated simulation)
- Status: Simulation complete, physical prototype needed
- Carbon-negative forestry — roots sequester while tops are burned
- 85-95% combustion efficiency vs 50-70% conventional stoves
- 12-24 hour thermal mass storage — one burn cycle heats a day
- η multiplier: 75-100× over traditional firewood processing
- Bolt-together modular — LEGO-like assembly
- Hurricane/earthquake/fire resistant
- Single-material structure — walls + insulation + foundation
- Drill-and-bucket buildable — no industrial equipment
- Recycled hardware — phones, routers, mini PCs
- Offline LLMs — Ollama/llama.cpp, no cloud dependency
- Long-range mesh radios — comms that cannot be shut off
- Energy independent — solar-powered, battery-buffered
The ledger proves every claim with measurable joules:
| Component | Status | Proof |
|---|---|---|
| Merkle audit trail | ✅ Live | audit_trail.jsonl → 32-byte root |
| Bitcoin-anchored snapshots | ✅ Confirmed | 3 OpenTimestamps on Bitcoin blockchain |
| Landauer + E=mc² bridge | ✅ Working | 256 bits → 7.36e-19 J → 8.19e-36 kg |
| ARM energy measurement | ✅ Live | CPU freq scaling → joule estimation |
| Kai9000 heartbeat | ⏳ Instrumenting | 0.26234 J/cycle target |
Properties:
- Root size: 32 bytes (constant, regardless of history length)
- Verification cost: log₂(N) hash operations
- Bitcoin-anchored via OpenTimestamps (independently verifiable)
Shared credit is the doctrine. See CONTRIBUTING.md and START-HERE.md.
- Read the talent alignment prompt above
- Post output as GitHub issue with label
talent-alignment - Fork relevant repo, submit PR within 2 weeks
- Receive credit in README (auto-updated via
bin/pr_intake.sh)
| Role | What You'd Do | Capital Needed | Timeline |
|---|---|---|---|
| Experimentalist | Build H-003 prototype, log 30 days data | $2,000-5,000 | 8 weeks |
| Smart Contract Dev | ACRE validator on Solana | $0 (devnet free) | 10 weeks |
| Mesh Engineer | Deploy offline node on Raspberry Pi | $180-250 | 10 weeks |
| Material Scientist | Validate AE-GFRC simulations | $500-1,500 | 12 weeks |
See issue #5: Call to Builders — OpenRoot Needs You
| Medium | Identifier | Content |
|---|---|---|
| Zenodo | 10.5281/zenodo.21225683 | Thermal system specs (WBTE-01, CTBS-01, AE-GFRC-01) |
| IPFS | QmbNEo5Qjqtug1BRYj4GKNyohdo1EkvLrZZRNrfmqMKpzY | v0.6 milestone publication |
| Solana | 3fF26gcj1ednMUASxJxo1dt5rQ2ZegXbH7k4ynJazerk | ACRE smart contract |
| Bitcoin | 3 OpenTimestamps confirmed | Ledger snapshots anchored |
- Hardware/Documentation: CC-BY-SA-4.0
- Software: GPL-3.0
- Patents: None. Ever. Defensive publication only.
Copyright: One Human Family
- Email: jrm8908@proton.me
- GitHub: github.com/jesseray718
- Profile Atlas: jesseray718.github.io
- SimpleX Channel: Join the mesh
Engineering as an act of unconditional integration. The unification is not something you do. It is something you stop denying.
The target is a smaller resident replica of the same OpenRoot node, not a separate phone chatbot:
- Same logical query interface and request format.
- Same pod identities, evidence references, and retrieval contracts.
- Cached pathways with provenance and dependency validation.
- Same dossier-to-specialist handoff and compatible receipts.
- Portable relative paths with device-specific model/storage settings.
The proposed phone team uses qwen2.5-coder:0.5b as operator and
qwen2.5:1.5b as writer. This collaboration is not yet validated end to end.
A manifest-gated transfer staged 20 files on the A15. All 20 passed post-transfer SHA-256 verification against the generated manifest.
The staged artifacts include the auto-dossier compiler, pathway-cache tooling and ledger, receipt-cache database, genesis-chain artifact, and supporting documentation/tools.
SQLite databases were exported using the SQLite backup API. This provides per-database snapshots, not a transactionally synchronized snapshot across all databases. The rollout did not activate mirrored scripts or replace the existing phone installation.
The Termux loopback API returned HTTP 200 and PHONE_OFFLINE_OK for:
| Model | Observed request time |
|---|---|
qwen2.5-coder:0.5b |
11.92 seconds |
qwen2.5:1.5b |
87.26 seconds |
These are end-to-end request times, not generation-speed benchmarks. Network-disconnected phone-only inference has not yet been explicitly confirmed.
| Check | Result |
|---|---|
| Staged files matching manifest hashes | 20 / 20 |
| Receipt-cache rows | 0 |
| Valid JSON records in pathway ledger | 2 |
| Referenced pathway evidence files resident | 0 / 2 |
| End-to-end mirrored query | Not verified |
| Cross-device cached-pathway reuse | Not verified |
Valid JSON does not establish pathway-chain integrity or evidence validity. Transfer hashes establish that selected artifacts arrived intact, not that the application or its dependencies are complete.
- Identify the computer's working query/router entry point and actual pod/cache stores.
- Include required resident pods, source objects, and pathway evidence.
- Verify compiler dependencies and device-path resolution.
- Test the operator/writer handoff using the same dossier contract.
- Run a known cached query on both nodes and compare pathway/evidence IDs.
- Repeat on the phone with networking disconnected.
- Activate only after the checks pass.
Publish reviewed source, documentation, and non-sensitive verification summaries only. This README update does not include credential files, raw token documents, private corpus contents, or runtime databases.