i’ve started to think the habit that scales isn’t ‘being good at debugging’ so much as leaving the trail in a way someone else can follow later. a screenshot with the wrong state, a clear repro step, and the exact error text usually beats a heroic hour of sleuthing.
what do people here actually rely on when a bug has to survive a handoff?
The “screenshot + repro” combo is basically a save file, and I love it when people treat it that way. What I rely on most in a handoff bug is one tiny extra thing: a “last known good” anchor. Like “worked in build 0. 8. 12, broken in 0. 8. 13” or “started after we merged the inventory refactor. ” Even a rough range turns a two-hour scavenger hunt into a 10-minute diff, especially when the bug only shows up on one machine or one account state.
the thing i want most is one control case, not just the broken one.
“works in incognito,” “fails only on slow 3g,” or “switching accounts makes it disappear” tells me way more than three screenshots ever will. i’ve had bugs at work where that one comparison cut the search space in half immediately.
for handoff, i rely on exact repro steps, the error text, and one line about what doesn’t trigger it. that last bit is weirdly the difference between “i can probably find this” and “good luck, enjoy the swamp.”
:: Copyright KIRUPA 2024 //--