Bỏ qua để đến nội dung

Attempt-flow deep policy for timed mode, draft TTL, and submit resilience

DomainsDOL EnglishProduct200 words1 min read
confirmedbyProduct Design

DEC-0069 - Attempt-flow deep policy for timed mode, draft TTL, and submit resilience

Phần tiêu đề “DEC-0069 - Attempt-flow deep policy for timed mode, draft TTL, and submit resilience”

DEC-0068 established attempt reliability baseline, but several high-frequency edge behaviors still needed explicit defaults to avoid inconsistent implementation.

Timed background and expiry behavior:

  • Timed countdown remains authoritative even when user is in background.
  • If timer expires, attempt follows timeout-submit path immediately.

Draft retention TTL:

  • Unfinished untimed attempts keep resumable draft for 7 days from last activity.
  • Timed attempts keep draft only within active timer window.

Submit review format:

  • Before submit, show lightweight total answered/unanswered summary.
  • For multi-section attempts, allow expandable section-level breakdown.
  • Submit remains non-blocking.

Submit retry and support:

  • On transient submit failure, auto-retry up to 3 attempts.
  • After auto-retry exhausted:
    • keep manual retry available,
    • show support CTA.

Timed pause policy:

  • No manual pause in timed mode.
  • Timer continues during tab/app background.
  • Removes ambiguity in the most failure-prone part of the learning journey.
  • Preserves low-friction behavior while improving reliability and learner trust.
  • Keeps implementation deterministic and scalable.

These defaults capture the minimum behavior needed for robust attempt execution without introducing complex branching or strict user penalties.