Architectural Hypotheses ​

The architectural hypotheses of the current stage include preferred architectural lines that already look defendable, but must not yet be presented as fully settled public truth.

1. High privacy with a verifiable canon ​

The chain layer must be compatible with closed, compact, and hidden contours, without turning into a totally public log of the entire past.

2. A unified canon with contour segmentation ​

The network must retain a shared meaning of canon without destroying the contour isolation of state, visibility, and data.

3. Contour-relative cross-contour semantics ​

If an object acquires a canonically significant form in another contour, this must happen not through automatic same-object semantics, but through a contour-relative projection, derived representation, or derived object with a typed relation to the source object and with a separate recognition basis in the target contour.

4. Eligibility-gated canonization with explicit conflict branches ​

The lower canonization layer must retain one current canonical winner for one canonical position, canonize only an admissible candidate with a distinguishable recognition basis, and must not reduce conflict to delivery order, local queue behavior, or implicit merge.

5. Three conceptual surfaces of token-mediated participation ​

The token layer may be further disclosed through connection, global publication, and operator participation as different surfaces of token-mediated participation in the decentralized regime, without turning them into a ladder of rights, a ladder of power, a hidden order of seniority, or final profiles of token mechanics.