Aglet

Prioritize upload resume gaps by lost transfer effort

Resume failures cost more when retrying requires transferring a large file or when a user cannot tell whether a partial object is safe to continue. Prioritization should use observed interruption cases and the workflow's recovery promise, rather than assuming every restart has the same consequence.

Decide where the work belongs

  1. Measure the user cost

    Record the file class, approximate transferred portion, interruption trigger, and whether recovery required a full restart. Separate a short retry from a long transfer that repeatedly loses progress, and note any work that must be repeated after upload.

  2. Assess recovery confidence

    Mark whether the client and receiver expose matching checkpoints, whether a partial object has a known identity, and whether a failed resume leaves a clear restart path. An ambiguous partial state deserves attention even when the example file is small.

  3. Choose the queue

    Queue offset correctness when evidence shows a deterministic range error, recovery-state design when the boundary is unclear, or progress-copy repair when transfer recovery works but the user cannot understand it. Attach a representative interruption trace to the decision.

What to carry forward

Prioritize the gap when a documented resume promise repeatedly converts an interruption into avoidable full transfer work or leaves partial data ambiguous. If only status wording is unclear, queue that separately so an offset repair is not delayed by a copy change.

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