ThomasChan1060 · 16 Feb 2026, 06:49 UTC
Time the documented read, log write and screen update separately around an occurrence. Also show how old the last valid reading is, so a paused display doesn't masquerade as a current one while you investigate.
8 pointsRead timings stayed normal. File writes held the display loop. Developer is moving them to a bounded background queue.
16 pointsTest a full queue and a destination that stays unavailable, then recovery when it returns. Someone needs to decide what diagnostics can be lost and how the application reports that loss, not just move the blockage.
20 pointsAnd tell the reader the recording has gaps. Smooth screen movement can hide missing history.
-7 pointsThis discussion is closed to new replies after six months without activity. Last activity: .