created 14 October 2026 · last modified 14 October 2026
the framework's own repair surface
Repair.
A theory about repair cannot hide its own corrections. This page is the source of truth for what has been corrected, retired, revised, or held open in the framework. Empty lists are honest — an empty list of retired claims is the truth, not a failure to populate. When a claim is retired, it lands here before it leaves the public surface elsewhere.
Companion pages: the empirical face of falsification at /research/falsification; the maturity-tagged claim status at /canon/claims.
§01version history
v0.32026-05-06Cosmologies section launches at /research/cosmologies — 12-month engagement with 11 non-Western trust substrates; Month 1 (Buddhist Sangha) locked with predictions. Enterprise ecosystem complete at /enterprise with 6 sub-routes. Trust Force Equation canonical form stabilized at Phase 6/C701, with Phase 7 power asymmetry surfaced as a structural variable but not yet added as an explicit integrand multiplier. Concept graph at 1,440 nodes / 4,600+ edges. KTP at v1.0 Lovelace.
v0.22026-05-04Enterprise section ships: /enterprise + 6 sub-routes, MIT AI Risk Scorecard v0.2 (1,612 risks scored against 9 KTP primitives, mean pressure 41.0). KTP-RPT v0.2 hardening pass: action tuple, governability ladder, adversarial pressure-test. Agents of Chaos (Shipwreck) study landed. Falsification and Methods pages upgraded. /about updated to v1.4.
v0.12026-04-26Canon Wave 1 ships: stack, claims, limits, glossary, ethics, synthesis. Cognitive section ships: hub, DCTI, ACPP summary, the Installed Watcher. Falsification, methods, cartography land on /research as canon-adjacent.
§02corrections
2026-05-06
Pollan book title in rooms.ts: *A World Appears* corrected to *How to Change Your Mind* (the actual book the Pollan four-room structure comes from).
The wrong title was in the rooms.ts data file header comment and would have been misleading to anyone verifying the citation.
2026-05-06
Case count in AcequiaBody lede and room 08 description: 'thirty-four analyzed' updated to 'forty-nine author-scored cases.' The 34 figure referred to original Trust Force Equation structure tests; the forty-nine is the full case-analysis corpus used in the proof.
Inconsistency between the room lede and the Counterproofs section, which explicitly names forty-nine cases and refers to the full corpus matrix.
§03retired claims
No claims retired yet.
When a claim fails its own falsifier or is superseded by a sharper formulation, it lands here with the successor named. The list grows over time; that is the framework working.
§04source corrections
No source corrections recorded yet.
When a citation is updated, withdrawn, or its meaning changes, it lands here. Source-level honesty travels with claim-level honesty.
§05open questions
Questions the program is holding open in public. Each names what the framework does not yet know and what would weaken or revise the corresponding claim.
research
Can the Trust Force Equation's variables (cost, presence, reciprocity, horizon, betrayal) be scored consistently across coders and substrates?
What would weakenIf the recoding gate produces inter-rater reliability below an agreed threshold, the equation must be reformulated.
framework
Does the trust integral hold any positive value across the human–AI boundary, or does C → 0 collapse it regardless of agent behavior? Shapira et al. 2026 (Agents of Chaos, Case #9 — Doug and Mira) supplies a positive existence proof inside the agent-to-agent boundary: working trust accumulates when the substrate carries cost, presence, reciprocity, and stable identity. The frontier is therefore weakened, not closed — the question reframes from species to substrate. Cross-agent dynamics at scale, and the human-to-AI direction specifically, remain open.
What would weakenIf witness and cost cannot hold meaning across the boundary in the human-to-AI direction at scale, KTP and Tamed Autonomy are sibling objects to relational trust, not the same object.
research
Does federation through nested enterprise actually compose? What makes federated trust networks bridge rather than fragment?
What would weakenEmpirical study of multi-network composition (acequia federations, Wikipedia subcommunity bridging) showing no detectable bridge effect over comparable non-federated controls.
research
Do trust-bearing edges actually require periodic renewal to remain active, or is renewal frequency uncorrelated with edge survival?
What would weakenStage 6 (DP9) field study showing renewal frequency does not predict edge survival in real systems.
operational
Can KTP red lines hold under production load, or does an enterprise that wants a person score break them in practice?
What would weakenA reference implementation under explicit governance constraints failing to deliver appeal, expiration, repair, and audit at production scale.
framework
Is the framework's response to ST8 (Foucault on power-as-constitution) sufficient, or does the constitution-of-power critique remain materially unanswered?
What would weakenA direct engagement essay or paper section showing the framework's commitments do not survive Foucault's strongest version.
research
Is there a specifiable repair function — how much repair metabolizes how much betrayal — or does Reconciliation Dynamics remain only convergent observation? Status: advanced, not closed. The direction (completed repair decrements R) and the trigger (completion, not delivery) are on the record at /canon/reconciliation-dynamics; the quantitative operator is not.
What would weakenIf the repair operator cannot be specified as a measurable parameter on the trust integral across at least two distinct substrates.
operating rule
If a claim changes, the change is documented. If a source is corrected, the correction is visible. If a metaphor is retired, the reason is recorded. The repair page is updated before the next major essay or paper ships, not after.
This is not administrative overhead. It is proof of trust infrastructure. A framework about repair that does not maintain a public repair path has failed its own claim.
How to cite
APA 7th ed.
Perkins, C. (2026, May 6). Repair. Kinetic Trust Protocol. https://kinetic-trust-protocol.net/canon/repair
Plain text
Chris Perkins, "Repair," Kinetic Trust Protocol, https://kinetic-trust-protocol.net/canon/repair (accessed May 6, 2026).
reference
pending — source not yet published