Aglet

Prioritize API nullable field regressions

Nullable changes are risky when a client cannot tell “not applicable” from “not received.” Prioritize fields that drive writes, eligibility, identity, or workflow state above display-only fallbacks. Weigh shared mapper reach and the ability to repair values without erasing the original distinction.

Decide where the work belongs

  1. Map semantic consequence

    For each field, list resource state, representation, client fallback, stored value, and downstream action. Separate a missing label from a false state transition. Keep actual examples tied to endpoint and client revision.

  2. Compare correction choices

    Assess preserving null, modeling absence, adding an explicit state, requesting a contract clarification, or holding writes. Record migration and rollback needs. Do not coerce null to empty everywhere while the field’s meaning is unresolved.

  3. Set the field queue

    Prioritize identity, authorization, and durable-state consumers before cosmetic screens. Set owner, fixture, review date, and evidence threshold. If only one endpoint is affected, keep shared mapper work behind a confirmed contract difference.

What to carry forward

Return a nullability queue with representation, semantic consequence, consumer reach, safe containment, owner, and acceptance evidence. Escalate fields that alter durable state or access. Keep order provisional when field meaning is undocumented. Keep semantic ambiguity visible to the queue owner.

Technical background: Project documentation on GitHub.

Keep the decision with the work.

Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.

Create an account See the product workflow