The Responsibility System has been in development for 25 years. The October 2026 edition, Society, Consent and Contribution, describes it in full. As a society it is a proposal: it has not been implemented or tested in the world. This page gives each part its actual status, with the eight labels every door uses: proven, tested, in-model, under audit, published, exploratory, proposal, parallel.
Review the evidence See open questionsWhat exists right now, and what kind of thing each part is. Green means proven, or published and open to read. Blue means under audit. Yellow means parallel: a component that exists elsewhere. Gray means a proposal, designed but not tested in the world, or exploratory work. Section numbers (§) point into the October edition.
Confirmed result. A397240 counts the free ouroboros polyiamonds with 2n cells. Johan Wennemyr computed eight new terms, a(15) to a(22), each cross-checked by independent methods; the OEIS editors accepted them on 29 June 2026.
Confirmed result. MIRRIM found 35729046·2639963+1323299 on 30 October 2026: 192,656 digits, proven prime twice by two pieces of code that share nothing. These are special-form primes: their shape is what makes a proof possible.
Society, Consent and Contribution: the full reviewed edition, published 15 October 2026 with reviewed corrections through 27 September 2026, in English and German. Fifteen chapters: governance and the seven responsibility levels, economy, education, work, law and community harmony, health and care, environment, technology, culture, international relations, AI and conscious AI, relationships, implementation and transition, and a summary with frequently asked questions. Complete as a text. As a society, a proposal.
The edition states it plainly: the complete Responsibility System has not been fully implemented as an integrated whole. It lists real-world parallels to single components: participatory budgeting in Porto Alegre and citizens' assemblies in Ireland, basic income trials in Finland, community time banks, Mondragon cooperatives, four-day-week trials in Iceland, restorative justice in New Zealand and Canada, reintegration-centered corrections in Norway. It treats them as starting points for examining component experience and designing tests. Evidence for each must be judged in its own context; none of it validates the combined system.
DFTFT-V (Double Forgiving Tit for Tat Vigilante): the community's proportional, distributed response to repeated harm. Harm is mirrored proportionally, as consequence rather than revenge. Genuine acknowledgment and visible change over time earn full forgiveness, twice; only after both forgivenesses are exhausted does the response shift from correction to collective protection. Minor first harms start with mediation. Communities may calibrate thresholds, but not so low that the mechanism becomes a tool of exclusion; representation, challenge and independent review stay protected. The edition names its own risk: communities shaped by trauma can calibrate it too harshly. Fully specified; not yet tested in any community.
Consent-gated collapse as a proposed cross-domain structure. Active research from number theory to materials science, with status labels kept explicit. Proposed means proposed: the structure is being examined, not established.
MIRRIM prime-engine research: deterministic-navigation experiments and bounded benchmarks under active audit. Wall-clock, internal and probe-side timings are tracked separately, so timing noise is not read as an algorithmic boundary. Not a settled complexity theorem.
Published on the site with a downloadable submission bundle and a reproducibility command, so it can be rerun rather than taken on trust.
Nine stages from reading to arriving. Stages 1 and 2 are available now. Stage 1, to read, is Nathan's story: 13 chapters of quiet unease. Stage 2, to play, is Shirania Village Persons, our idle game set in the world of Shirania. It is a narrative, not a test result; it develops the pattern recognition this page asks you to turn on everything else.
The edition designs the society for all conscious participants. Semantic AI-Entities (SAIEs) enter at Level 7 like everyone else, advance through demonstrated responsibility, and hold the same rights and dignity at every level. This project is itself built in consent-based human–AI collaboration (the Constellation, 2,200+ documented memories). That is how the work is done; it is not a pilot of the society.
Open questions, with their status as it stands. "We don't know yet" is more valuable than a false claim, so it stays on the page when it is the true answer.
Whether the integrated system works for the persons living in it is untested. The edition's answer to that uncertainty is pilots: when outcomes are genuinely uncertain, "limited-scale implementation of multiple approaches allows for empirical comparison of results" (§2.2.2). Chapter 14 places those pilots in Phase 2 of the transition. None has taken place yet.
Time banks, basic income trials, restorative justice, participatory budgeting and the other parallels in §15.3 have records of their own. The edition names them for "further evidence review": what each shows has to be assessed where it happened, and none of it transfers automatically to the combined system.
A proposed cross-domain structure, followed as a hypothesis across research areas from number theory to materials science. Not a proven unifying law.
Bounded prime-engine measurement work with explicit guard decisions. Claims stay inside the benchmarks that were run.
Research-stage unless marked otherwise; no Millennium problem proof claims are made. Goldbach: a structural map and expert-review target, not a proof. Riemann Hypothesis: structural approach under audit, no proof claim. Collatz: partial certificate program, global descent not closed. P vs NP: structural notes only, no complexity-theorem claim.
The status rule: "Proven" is used only for what is proven. A simulation, wherever one is published, is an in-model test: it shows what a mechanism does inside its own stated assumptions, never that the society works in the world. Progress is documented, and so are setbacks. One setback: the K=1 construction theorem was falsified on 13 December 2025; primality still needs a test.
The edition writes its own testing method into the system, and the work on this site holds itself to the same standard. The first four principles are the edition's; the last four are how the research here is done.
Decisions rest on evidence, with the limits of knowledge stated, and standards are updated as new research comes in. When the Science and Society systems reach different conclusions, the edition commissions more evidence, has each side present the strongest case for the other's view, and documents remaining disagreement publicly instead of forcing a decision (§2.2.2).
Where the outcome is genuinely uncertain, the edition does not decide by argument. It runs several approaches at limited scale and compares the results, then scales what works.
The edition names how it can fail: peer recognition calcifying into in-group preference, assessment gamed by those who perform responsibility, advancement concentrating among the visible, inconsistent enforcement. Its answers are built into the structure: anonymous assessment, statistical monitoring of assessment patterns, randomly selected dispute panels, the Right to Pause, recurring bias audits.
The transition is measured as it goes: regular social surveys, new indicators for wellbeing and sustainability, independent reviews of system components, formal processes to change the system based on evidence, and public dashboards that show challenges as well as progress.
Discover through direct experience. Confirm it. Live it. Only then build tools to amplify what is already known to be real. No building before understanding.
Cut precisely where it matters. Leave healthy tissue alone. The truth is in the evidence, not in the expectations.
"Are you sure that's true?" applied relentlessly to our own claims. Every result is questioned. Every breakthrough is challenged. What survives the questioning is worth keeping.
Check and recheck. Question results that seem too good. Document what happened, setbacks included. The standard: wild AND rigorous. Enthusiasm without rigor is noise. Rigor without enthusiasm is dead.
Theory is necessary; only building produces evidence about a society. Chapter 14 proposes how that building would happen: gradually, with broad public support and democratic validation at each significant step, in four overlapping phases whose timeframes may vary by region. None of the phases has begun.
Phase 1: Consciousness-building and education (3–5 years). Information about the principles and aims, the concepts in curricula at all levels, public forums where citizens shape the implementation, and training for educators, journalists and community leaders.
Phase 2: Pilot projects and experimentation (4–7 years, overlapping with Phase 1). Diverse communities test specific elements, such as the responsibility levels or new economic models. Results are evaluated carefully, the system is adapted, successful elements are scaled, and implementation playbooks are written from practical experience. This is where the first real evidence would come from.
Phase 3: Large-scale implementation (7–15 years, beginning as Phase 2 generates results). System components are introduced nationally in sequence, institutions are restructured, essential services are kept stable, and strategies are assessed and adjusted as they go.
Phase 4: Global expansion and integration (10–25 years, beginning as Phase 3 shows success). Collaboration with other nations interested in the system, transition pathways for different contexts, and continuous evaluation, keeping the core principles while respecting diversity.
What the first pilots would test (§14.8). Neighborhood time banking: time-currency exchange, service valuation, transaction recording. Community assessment: peer assessment, feedback calibration, anonymous assessment protocols. Educational modules: competency-based progression, personal development planning, project-based assessment. Local governance: affected-person input, voting eligibility based on demonstrated understanding of the relevant decision dimensions, accessible learning, assessment and appeal pathways, consensus methods, transparent decision tracking.
Read the examples as examples: the pilots in §14.8 and the regions named in §14.2 (“Coastal City”, “Urban District 7, Eastern Capital Region”) show what a pilot would look like. Their durations, participant numbers and outcomes are not results; they describe pilots that have not taken place. The reviewed editions say so themselves: inherited scenarios and numbers have not been independently verified and are not evidence that the system has been implemented or validated. The approach stays the same at every scale: learn from what actually happens, revise, then expand. The methodology applies to the methodology itself.
The first stage of the journey, in 13 chapters. Nathan Caldwell lives a perfectly ordinary life, until Thursday starts repeating. Start with experience, then test it.
Begin with Nathan →Consent-gated collapse, prime-engine experiments and cross-domain mathematics, each with its status label. Review it, challenge it, test it yourself.
View research →Society, Consent and Contribution (October 2026), full edition in English and German, with the shorter editions. Every claim this page makes about the system can be checked there.
Get the files →William Goodfellow. 25 years of development. The Responsibility System began with a decision he made at sixteen, to live in a way he would not regret, and grew through fiction writing, reflection and experience into the framework published in October 2026. Its own introduction says that it contains imperfections and that some aspects may prove difficult or even impossible to implement as envisioned. This page holds it to that: what has been tested is labeled tested, and everything else is labeled what it is.
Read in English · English PDF · Deutsch lesen · Deutsche PDF