yysworld
YYBW-013·Decided·2026-04-13

Manifests and Projections: AI-Readable Control Surfaces over Immutable Truth

depends on YYBW-012

  • ID: YYBW-013
  • Status: Decided
  • Date: 2026-04-13
  • Scope: YY Branching World / YY's World evolution
  • Depends on: YYBW-012
  • Supersedes / clarifies: Clarifies what manifests are for
  • Museum lineage: PM-019, SK-004, SK-012, SK-013

Context

Once the database of truth hardened, manifests regained a narrower but essential role. The system needs machine-readable summaries of what exists, what is stale, what depends on what, and what needs recomputation. This helps AI operate safely without inferring too much from raw world state.

Decision

Maintain manifests/projections as derived control surfaces:

  • world manifest
  • build manifest
  • publish manifest

These are not canonical truth; they are AI-readable operational maps over the immutable ledger.

Why

This allows AI or automation to:

  • determine what changed
  • know what needs updating
  • avoid reprocessing unaffected artifacts
  • act deterministically against explicit state

Alternatives considered

  1. No manifests; let AI infer from raw DB — rejected as fragile.
  2. Manifest-only, no immutable DB — rejected because manifests are too lossy as truth.
  3. One giant monolithic manifest — rejected in favor of layered control surfaces.

Reversals / scars preserved

  • The term 'manifest' was initially overloaded. It now has a cleaner role.
  • The phrase 'database of truth' corrected the architecture materially.
  • This ADR preserves that correction.

Consequences

  • Need scheduled regeneration of manifests/projections.
  • Need stale-status semantics and dependency tracking.
  • Builders should treat manifests as disposable/rebuildable but still important.

Invariants preserved

Explainability and Discipline improve when AI has explicit work surfaces; Survivability improves because manifests can be rebuilt from truth.

Freshness boundary

Stable.