Map & Measures

Start with the map: Packs 1-4 form the care cycle, Pack 5 scales that cycle across organisations, and Pack 6 is the boundary condition that keeps every deployment local, plural, and sunset-ready. The map is the framework index — pack stations open their chapters, measure links land on the audit entries below, and handoffs show what each phase gives the next.

Framework map

Each pack then answers a distinct public audit question. To preserve the public promise of one headline public measure per pack, each pack has one headline public measure plus supporting diagnostics.

These metrics are designed for sufficiency, not maximisation. Each deployment context defines a threshold — "good enough" for that community. Crossing the threshold is the goal; score-chasing beyond it risks the same metric gaming the 6-Pack warns against. A headline measure is a public test, not a totalising score.

Headline public measures

Representation gap — the Attentiveness measure: Which materially affected groups are still missing or badly under-represented in the record?

Promise fidelity — the Responsibility measure: What share of material obligations are explicitly owned, properly authorised, and kept on their published terms?

Verified execution rate — the Competence measure: What share of audited decisions or releases pass guardrails, include a usable trace, and stay inside release bounds?

Trust-under-loss — the Responsiveness measure: After a bad outcome and attempted repair, do affected people report that the system became more trustworthy rather than less?

Uncommon-ground index — the Solidarity measure: Are shared decisions showing real cross-group participation and co-endorsement, rather than separate silos?

Exit readiness — the Symbiosis measure: Could this system hand over or shut down on schedule without rights loss, continuity failure, or recentralisation?

Supporting diagnostics

• Pack 1 — Attentiveness. Coverage of affected people; voice equity between least-heard and most-heard groups. Evidence artefact: the bridging map.
• Pack 2 — Responsibility. Named-owner coverage; authority-match rate; adopt-or-explain rate.
• Pack 3 — Competence. Guardrail integrity; trace completeness; canary health; audit overturn rate.
• Pack 4 — Responsiveness. Appeal closure time; repair completion rate; harm recurrence within 90 days.
• Pack 5 — Solidarity. Portability success rate; accountable-identity coverage; federation participation — the structural conditions that keep co-action freely given rather than captive.
• Pack 6 — Symbiosis. Scope compliance; sunset compliance; ecology diversity; handover rehearsal pass rate.

Named instruments

The book's Parts List catalogues every instrument a practitioner is asked to build or run. The ones the measures above depend on:

• Perspective receipts (Pack 1) — the acknowledgement each contributor can hold: a receipt that lets every person find and correct how their words were represented in the record — distinct from the civic receipt a decision trace carries for settlement.
• Engagement contract (Pack 2) — a short, legible public document for every significant Kami deployment, recording what the system is supposed to do, who is answerable for it doing that, what happens when it goes wrong, and how the deployment will eventually end; Pack 6 renders the same contract machine-readable, so its purpose bounds, consent rules, data-retention limits, portability guarantees, and shutdown procedures are enforced by infrastructure.
• Adopt-or-explain (Pack 2) — the rule that when an Alignment Assembly produces a recommendation, the team either integrates it into the system's behaviour or publishes a reasoned explanation of why not, with the remedy offered instead.
• Obligation ledger (Pack 2) — a public, weekly, digitally signed record of what the deployment has committed to deliver and whether it is meeting those commitments, kept by a named Participation Officer.
• Shadow mode (Pack 3) — the first stage of the Apprentice Model: the system sees real inputs and proposes real actions but does not act, its proposals compared against the human or prior system so divergences can be studied before trust is granted.
• Canary release (Pack 3) — the second stage: deployment to a small, stratified, representative slice of real cases with automatic rollback triggers if drift exceeds defined bounds.
• Shared eval registry (Packs 3-4) — a public, Wikipedia-like registry of versioned, community-authored evaluations: affected people write the tests for harm and repair (Pack 4), and passing them becomes a release gate (Pack 3); Weval.org is the working example.
• Decision trace (Pack 3) — the per-action record showing which rule fired, which sources were consulted, and what uncertainty the system carried, the operational counterpart that makes the engagement contract live rather than merely written. When a Kami draws on a community's knowledge, the same trace is designed to double as a civic receipt (Pack 6), settling value back to the custodians through pre-funded escrow — a design commitment, not yet a running settlement, and distinct from Pack 1's perspective receipts, which let contributors see and correct how they were represented.
• The brake (circuit breaker) (Packs 3-4) — a single, prominent, wired-and-tested control that stops the system now, accessible at the speed of human recognition, so a person can halt a wrong action in the moment rather than only signalling for later.
• Override ledger (Pack 4) — a room's plain-text working memory recording every time a human said "no" to the Kami — evidence that its governance charter is live and the room still holds standing to correct it.

What each measure refuses to reward

Each headline measure names not only what it rewards but what it must refuse, so that a good number cannot be mistaken for good care.

• Representation gap (Pack 1). Counts only when the least-heard groups gain real standing in the record, never when a gap narrows by averaging dissenters into the middle.
• Promise fidelity (Pack 2). Rewards promises that are owned and honoured, never promises that are merely well-worded, so a high score requires authority and funding that match the duty rather than language that gestures at it.
• Verified execution rate (Pack 3). Rises only when traces are reconstructable by an independent auditor and canaries are genuinely representative, never when the system passes on easy cases and calls the ceremony proof.
• Trust-under-loss (Pack 4). Counts only when tied to accountable identity and corroborated by an independent signal — a return to the service, a withdrawn appeal, a third party who can attest — never a sentiment a campaign can manufacture, since self-reported trust after a repair can be astroturfed by the very actor who caused the harm.
• Uncommon-ground index (Pack 5). Rewards groups that listen across lines, never groups that fall silent, and never a room curated into agreement — the index is read against Pack 1's representation gap, because the cheapest way to raise co-endorsement is to exclude whoever would withhold it. A floor protects standing opposition, so the measure does not pathologise legitimate, sustained disagreement by treating a persistent minority as a failure to engineer away.
• Exit readiness (Pack 6). Counts only when a system can hand off or shut down on schedule without rights loss, continuity failure, or recentralisation, never when exit exists only on paper.

How to read them

• Trust is decomposed, not collapsed. Pack 1 asks whether people were heard. Pack 2 asks whether promises were real. Pack 3 asks whether delivery held up under inspection. Pack 4 asks whether repair worked after harm. Pack 5 asks whether groups can act together fairly. Pack 6 asks whether stewardship can remain bounded over time.
• Artefacts are not metrics. Bridging maps, engagement contracts, repair logs, and exit drills are governance evidence. They matter because they support the headline measures; they are not scores by themselves.
• The map is not the index. Pack 1's bridging map charts where crossings are possible — who agrees, who clashes, and why, among those present in the record. Pack 5's uncommon-ground index counts the crossings actually made: shared decisions co-endorsed across group lines. Terrain first, then traffic. Composition precedes co-action, so a high index over a wide representation gap measures selection, not solidarity.
• Training signals are not the public audit. Cross-group endorsement may inform model tuning or routing, but its public, conference-facing role in this framework is the Pack 5 uncommon-ground index. Pack 4 owns whether affected people could correct the system and whether repair restored trust.
• The record is small and named. Most strong cases come from places with unusually strong civic infrastructure — thresholds are earned room by room, not inherited from Taiwan or California.

Before you call it Civic AI

The measures above audit a running deployment. These six questions gate whether a deployment should carry the name at all:

1. Human standing (Pack 1) — can affected people prove standing without surrendering unnecessary identity?
2. Adversarial load (Packs 1, 4, 5) — what stops synthetic publics, bots, or paid campaigns from flooding the channel?
3. Decision force (Pack 2) — where does the output bind actual decisions, and where is it only advice?
4. Repair path (Pack 4) — who can appeal, who pays for harm, and where is the repair logged?
5. Exit capacity (Packs 5, 6) — can the community migrate, fork, pause, or shut down without losing service continuity?
6. Boundary honesty (Pack 6) — what problem is out of scope, and who is carrying that problem instead?

If the answer to any of these is vague, the deployment is not ready. The point is not to defend the 6-Pack. The point is to make the room answerable.