StayC — Stability Codes for Maturity Tracking#

StayC codes track the maturity of any formal claim through seven stages, from first intuition (MM) to global community acceptance (SS). The codes are the core of the StayVS system.

Key insight: NN (NimbleNonfunctional) is not a graveyard — it is the death valley that every innovation must cross to produce its minimum viable product at OO. The actual graveyard is KK (KnownKiller) in the POST system. NN carries hope of rescue; KK does not.

The Seven Stages#

Code

Full Name

Math Stage

Pict

Definition

Analogy and Example

MM

MockupModel

Intuited

Idea exists informally as back-of-envelope reasoning or verbal sketch. No formal statement yet. The mathematical intuition is present but has not been cast into precise language. Most people tend to completely reject ideas at this microscopic stage.

“I think there’s a way to prove that any society without periodic structural reset eventually self-destructs… let me think about how to formalize that.”

NN

NimbleNonfunctional

Death Valley

×

The innovation’s death valley: the idea has been stated but does not yet function. It needs feeding — resources, refinement, and iteration — to grow into a minimum viable product (OO). Most innovations die here, not because they are worthless, but because they run out of energy before crossing the valley. NN carries hope of rescue. A claim at NN is not dead; it is starving. Feed it and it may reach OO. If the problem is proven terminal, the claim moves to KK (KnownKiller), not NN.

NN also occurs as a rejection event at any later stage: a claim at OO, PP, or QQ can fall back to NN when a flaw is found. The rejection is documented with reasons. Revised versions re-enter the pipeline.

The “valley of death” in startup funding: most startups die between seed and Series A, not because the idea is bad but because they cannot sustain operations long enough to prove it works. Newtonian mechanics was not killed (KK) by Einstein — it was shown to be an approximation valid in a limited domain (NN), and it remains useful within that domain.

OO

OperatesOddly

Conjectured / MVP

?

A minimum viable product exists. Formal statement is present, believed true based on evidence or pattern recognition, but no proof has been attempted. The MVP is somewhat wobbly — it works, but not much more. Much refinement is required. Analogous to a mathematical conjecture or a startup that has its first paying customer but no repeatability.

Goldbach’s Conjecture (1742): precisely stated, universally believed true, unproven for 284 years. In this project: a new axiom stated formally but without derivation or proof sketch.

PP

PathProbing

Proposed / Monopoly

Formal statement accompanied by a proof or structured argument. A stable core exists — successful enough to dominate its niche, like a monopoly. But monopolies generate blindness: they solve 80+% of problems while ignoring the 20% that cause great pain to everyone else. PP claims are not yet adversarially tested and may harbor hidden weaknesses invisible from inside the monopoly’s perspective. Analogous to a preprint or to IBM in the era of “nobody ever makes a mistake for buying IBM.”

ax15–ax25 and th5–th11 as first written in OOv1: formally stated with proof sketches, before any critique round. The Catholic Church in the European Middle Ages: dominant, functional, solving most questions of meaning — but blind to the 20% that would eventually drive the Reformation.

QQ

QualityQuest

Contested / Reformation

Subjected to adversarial review (quest). The monopoly is broken up by faster, better iteration — feeding innovation. Objections have been raised; some resolved, some remain open. The claim is under active defense, increasingly refined but still fluid. Like the Protestant Reformation: the breaking-up of intellectual monopoly through critique, counter-critique, and competition among formulations.

ax15–ax25 and th5–th11 after OOv1’s 3 critique rounds: C1 (bistability unproven) rebutted; C2.1 (causal gap) partially addressed; feasibility critiques remain open.

RR

ReviewedRelease

Defended

All critical objections resolved through the quest process. Proper stability for very broad use begins here. The author judges the claim publication-ready. Remaining objections (if any) are classified as non-critical or as scope limitations. Comparable to a peer-reviewed and accepted paper.

ax1–ax14, th1–th4: poster PPv1r1p1. Note: assessments differ — Claude (dv_ClaOp46Max_RR) deems ax1–ax14 defended; LLoL (iv_LLoL_PP) retains PP pending broader feedback.

SS

StableSource

Established

Widely accepted by the relevant intellectual community after broad global review and adoption. Multiple independent checks exist. The result has been tested against diverse objection traditions and survived. Comparable to an internationally stable standard — or, in the strongest case, like quoting “the Bible” on a topic: the reference is so widely accepted that citing it settles the question for the community.

The Pythagorean Theorem. In this project: no axiom or theorem has yet reached SS. SS requires extended community engagement beyond the author and AI reviewers.

Lifecycle Paths#

Forward path (the normal journey):

MM ──→ NN ──→ OO ──→ PP ──→ QQ ──→ RR ──→ SS
idea   death   MVP    proof   quest   reviewed  established
       valley         sketch  /reform  release

The three fast iteration cycles:

  1. OO (OperatesOddly): Rapid iteration on the MVP. The claim is wobbly; each cycle makes it slightly less wobbly. Many small revisions, quick feedback loops.

  2. PP (PathProbing): Iteration on the proof/argument. The core is stable but blind spots exist. Each cycle probes a new path, potentially revealing monopoly-blindness.

  3. QQ (QualityQuest): Adversarial iteration. External critics find weaknesses; the author responds. The fastest refinement happens here because the feedback is adversarial, not sympathetic.

The two slower cycles (RR and SS) involve broader community review and adoption. These operate on longer timescales and are not the primary concern during model development.

Rejection events (can occur at any stage past MM):

OO ──→ NN    (flaw found in MVP)
PP ──→ NN    (proof attempt fails)
QQ ──→ NN    (adversarial review finds fatal flaw)

From NN:
  ├── feed and grow ──→ re-enter at OO (revised version, incremented)
  ├── fundamental rethink ──→ return to MM (new approach)
  └── proven terminal ──→ KK (KnownKiller, via JJ if intermediate)

NN is not the end. Most successful innovations pass through NN multiple times. Each passage documents what failed and why, building the knowledge needed to eventually cross the death valley.

Iteration Cycles at Every Stage#

The iteration cycle is not specific to QQ. It applies at every stage transition where a claim is being refined. The same version/rejection/revision machinery works wherever a claim currently sits:

OO iteration (refining the MVP toward PP):

  • OOv1 — first formal statement.

  • NN_OOv1 — flaw found (e.g., formalization error, missing case).

  • OOv2_NN_OOv1 — revised statement addressing the flaw.

  • Continue until a proof or structured argument is attempted → PP.

PP iteration (refining the proof toward QQ):

  • PPv1 — first proof sketch.

  • NN_PPv1 — gap found in proof (e.g., unstated assumption, circular reasoning).

  • PPv2_NN_PPv1 — revised proof addressing the gap.

  • Continue until adversarial review begins → QQ.

QQ iteration (adversarial refinement toward RR):

  • QQv1 — first version under adversarial quest.

  • NN_QQv1 — objection succeeds (documented with specific failure).

  • QQv2_NN_QQv1 — revised version addressing the objection.

  • NN_QQv2_NN_QQv1 — further objection; chain continues.

  • Continue until all critical objections resolved → RR.

RR iteration (broad review toward SS):

  • RRv1 — first reviewed release.

  • NN_RRv1 — community reviewer finds issue.

  • RRv2_NN_RRv1 — revision addressing the issue.

  • Continue until broad global acceptance → SS.

MM is typically too informal to carry formal version numbers, though an innovator who systematically journals their intuition development could in principle track MMv1, MMv2, ....

Rules (apply at every stage):

  1. Each version at a given stage gets a monotonically increasing number.

  2. Each NN explicitly names the version it refutes.

  3. Each responding version names the NN it addresses.

  4. If refutation cannot be overcome, the claim stays at NN (and may eventually move to JJ or KK if the block is structural).

  5. If a fundamentally new approach is needed, start a new claim at MM rather than incrementing the version at the current stage.

Advancement Authority: Human Decides, Machine Advises#

Stage advancement is a human decision. An AI auditor (or any machine-based assessor) may propose that a claim has reached the criteria for the next stage, but the human author decides whether to accept the proposal. This asymmetry is deliberate:

  • The machine track (dv_ regime) provides advisory assessments. Claude may say “this claim meets PP criteria” or “this claim should fall to NN.” These are recorded as dv_ VVNs.

  • The human track (iv_ regime, or other human regimes) provides authoritative assessments. LLoL (or any human assessor) may agree or disagree with the machine assessment. Their VVN is recorded independently.

  • For publication and advancement decisions, the human track governs. If Claude assigns dv_ClaOp46Max_PPv1 but LLoL retains iv_LLoL_OOv3, the claim is at OO for publication purposes.

  • Disagreement is valuable, not a bug. When the two tracks diverge, the disagreement itself is data: it identifies claims where human intuition and machine analysis see different things. These claims deserve extra scrutiny.

  • The machine may insist. If Claude believes a claim is at a different maturity level (higher or lower) than the human assessment, Claude should say so clearly and maintain the divergent dv_ VVN. The human is free to override, but the machine assessment remains in the record. Future reviewers can see both.

The 2-track system in practice:

Claim

Human VVN (iv_)

Machine VVN (dv_)

Implication

ax15

iv_LLoL_OOv2

dv_ClaOp46_QQv2

Machine sees more maturity than human. Claim stays at OO for publication. Divergence flags ax15 for review.

th8

iv_LLoL_PPv1

dv_ClaOp46_NNv1

Machine sees a flaw human missed. Human should investigate the NN reason. May accept (→ NN) or rebut (→ maintain PP).

ax1

iv_LLoL_PPv1

dv_ClaOp46_RRv1

Machine more confident than human. Human retains PP pending broader feedback. Both assessments are valid and recorded.

Connection to POST Pipeline#

See StayVS — Stabilizing Versioning System for the full JJ/KK integration. In brief:

  • JJ (JammedJob): known problem, hope of fix. Can rescue NN claims.

  • KK (KnownKiller): proven terminal. The actual graveyard. Unlike NN, KK carries no hope — but it carries understanding of why this fails, which prevents others from repeating the same error.

The distinction matters: calling something NN when it should be KK gives false hope. Calling something KK when it could be rescued (JJ → NN → OO) kills innovation prematurely.