Interface: DecisionStatus

Defined in: src/atom/types.ts:2931

ONE DECISION THAT BELONGS TO A PERSON, and whether it has been made — the rows Session.decisions serves, read at the moment you ask.

The sibling of AskStatus: that one answers "is anything waiting on a person?", this one answers "is anything a person's to decide?". They are different questions with different next moves, so they keep separate rows and share no vocabulary — no askId appears here, because a decision mints no card and there is nothing for did_it_work to be asked about.

Graph-wide, like the ask book: a decision on another page still holds a journey, so a row exists for every declaring control wherever it lives.

Properties

about?

optional about?: string

Defined in: src/atom/types.ts:2934

The app's own words for what is being decided — DATA (see HumanDecides.about).

---

affordanceId

affordanceId: string

Defined in: src/atom/types.ts:2932

---

made

made: boolean | "unknown"

Defined in: src/atom/types.ts:2950

Whether the app's own doneWhen holds RIGHT NOW, evaluated fresh on every call:

• true — it holds.
• false — it was evaluated and does not hold. Only an evaluable, failing condition may say this.
• 'unknown' — it could not be evaluated (a key absent from the state view, or holding undefined — the same rule guardUnevaluated applies to guards), or no doneWhen was declared at all.

'unknown' IS NEVER COLLAPSED INTO "not yet". They are different answers to different questions, exactly as an unevaluable guard is served with a marker rather than treated as failed.

---

madeBy?

optional madeBy?: Principal

Defined in: src/atom/types.ts:2968

WHO MADE IT — served beside made: true only, and minted from exactly the identity-bearing rungs of updateState's attribution ladder: a delta naming a fired transition, a handler's own call window, or an attributed updateState(delta, { principal }).

ABSENT IS THE HONEST ANSWER and the common one. The matching rungs — FIFO settlement, the single-cover arm, effect-signature inference, the unknown-stimulus floor — compute a JOIN, and a computed join never attributes a human decision, so each of them CLEARS whatever the book held. The decision is then visibly made and nobody is named: the library does not say the human did it, and does not say they didn't.

Never inferred, never defaulted, and never 'user' unless a door that carries identity said 'user'. A sentence typed in conversation reaches no such door.