Trust

Every verified participant gets 100 trust allocations per governance cycle. You distribute them across people and topics — "I trust Alice on photography", "I trust Bob on identity-verification". Trust on climate science doesn't make you trusted on software security.

Square root means distributed trust beats concentrated trust. Ten people giving 10 points each produces more power than one person giving 100.

Writing "I trust Alice" isn't enough. For a trust allocation to become recognized by the network, two things must happen:

Witnesses are peers who earned witnessing competence in the previous governance cycle. They collect and sign timestamps on the envelope. Once enough witness power accumulates, the allocation becomes valid — and anyone can independently confirm it.

Trust allocations don't last forever. The year is split into three governance cycles, and every participant belongs to exactly one cohort — determined by their identity hash (1, 2, or 3).

Each cohort's trust allocations are locked at a specific date, then only taken into account four months later. This delay is deliberate — it prevents last-minute manipulation and gives the network time to verify everything.

You can update allocations between cycles, but they don't undermine currently recognized authorities. Power transitions are predictable and smooth.

Trust isn't only positive. You can allocate negative trust on bad actors — a public signal that starves them of attestation power and makes their content filterable across the network.

Negative trust expires over time. Change your behavior, and your reputation recovers. It's more forgiving than permanent bans — and more honest than shadow-banning.

The equal split of 100 trust allocations per cycle is not a mechanism to prevent centralization, nor to force it. It is a balance between the two — a lever that the community holds in its own hands.

Some tasks genuinely benefit from concentrated competence. Identity-verification, forensic analysis, security auditing — if the community feels stronger checks are required and should be done by only a few highly competent peers, they allocate trust to reflect that. Centralization becomes a deliberate choice, not an accident of who showed up first.

And when a process has drifted too far — when the same few names accumulate power across too many cycles — participants can allocate negative trust on the top trusted peers. That starves the concentrated authority of attestation power and forces the process to decentralize again.

The same tool that enables centralization provides the lever to reverse it. No committee picks the balance point — the network does, continuously, every cycle.