Change Sequence

Use this sequence when a change affects shared records, validation meaning, serialization, diagnostics, identities, time, units, or support-matrix contracts. Core changes are expensive because every higher crate may interpret the same type differently if the contract is vague.

Decision Flow

flowchart TD
    change[proposed shared change] --> shared{must multiple crates read the same meaning?}
    shared -- no --> owner[keep it in receiver, signal, nav, infra, or command]
    shared -- yes --> family{which core contract family moves?}
    family --> artifacts[artifact payloads and validation]
    family --> observations[observation, acquisition, tracking records]
    family --> nav[navigation solution records]
    family --> foundation[identity, time, units, coordinates, diagnostics]
    family --> support[support matrix and configuration validation]
    artifacts --> proof[contract docs plus focused core tests]
    observations --> proof
    nav --> proof
    foundation --> proof
    support --> proof

Change Sequence

1. Decide whether multiple crates need one shared interpretation.
2. Identify the contract family in crates/bijux-gnss-core/docs/CONTRACTS.md and CONTRACT_MAP.md.
3. Inspect the owning source family before changing downstream callers.
4. Update crate-local docs when serialized meaning, validation, or public API changes.
5. Update this handbook when a reader-facing package route changes.
6. Run the focused core test that protects the moved contract.
7. Only then inspect receiver, nav, infra, signal, or command fallout.

Proof Selection

change family	first proof
public exports	cargo test -p bijux-gnss-core --test public_api_guardrail
navigation artifact payloads	cargo test -p bijux-gnss-core --test nav_artifact_validation
tracking artifact payloads	cargo test -p bijux-gnss-core --test tracking_artifact_validation
timekeeping and units	cargo test -p bijux-gnss-core --test prop_timekeeping
workspace guardrails	cargo test -p bijux-gnss-core --test integration_guardrails

Skipping ownership review is the expensive mistake. A downstream owner is often cleaner than turning a local convenience type into a permanent shared contract.