1 The owner chooses which proof to require and how long it stays valid. The wallet supplies it, so the visitor does nothing on the site. The gate checks it at the edge.
2 Staying undisclosed is cheap to refuse. That is the incentive to declare.
3 People and bots are treated alike: each branch lists access points, and each access point has exactly one requirement. A person's "All content" and a bot's
subscribe point at the same Subscription card.4 Free, payment, and work are each a card type, and a card holds one item. Payment is settled by the wallet, from the visitor's own funds or from an allowance a sponsor such as an institution has funded. A work card links to the board in illustration 1b.
This is a simulation. There are no real proofs, no real gate, and no real wallet. The editing controls on the cards are only styling, and the diagram is static.
Open questions
- Who runs the gate: the site, a CDN, or a third party such as Mozilla or Cloudflare?
- Which proofs count (personhood, humanity, liveness), and can a proof be reused across sites without tracking the visitor?
- How is a false declaration handled, such as a bot claiming to be a person?
- What exactly does "undisclosed" get: how limited, and is it the same for every site?
- Who counts as a trusted issuer of humanity proofs and of sponsored allowances?
- How is a sponsored allowance capped for one agent and topped up for another?
- Should there be a shared format for these policies so a bot can read them the same way on every site?