Aglet

Learn from repeated upload recovery state confusion

Recurring recovery questions usually mean that connection state, upload state, and page state use different vocabularies. Learning should preserve the interruption paths that exposed the mismatch and make the evidence required for each action part of the file workflow contract.

Keep the lesson for the next incident

  1. Cluster the interruption cases

    Group findings by active request, acknowledged progress, page revisit, stale session, duplicate attempt, and discarded partial data. For each group, record the state users saw, the action they chose, and the artifact outcome.

  2. Document recovery semantics

    Define what resume, restart, wait, and discard mean, which evidence enables each, and how file identity changes. Add timing examples for a pending response and an acknowledged boundary so recovery rules cover ordering.

  3. Trigger future review

    Require the interruption fixture when changing retry policy, persistence, page initialization, session expiry, or recovery copy. Assign an owner for state semantics and record any boundary where only a restart can be promised.

What to carry forward

The learning record is useful when every recovery action has a stated precondition and outcome, with interruption fixtures that expose timing differences. If the workflow cannot recover a case, make the restart boundary and artifact disposition explicit before calling it a user error.

Technical background: tus protocol documentation.

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