Is it good enough to ship?
flush the buffersatisfice
Philosophy and Tactics #
Am I spinning in circles? #
If I'm struggling to decide "is this good enough," I have some quick rubrics:
- does it have a low "blast radius"
- what's my escape hatch(es)
- does it meet the team's rubric for quality?
Don't boil the ocean #
Keep eyes on the prize! If there's opportunities to raise the quality in aspects that aren't directly related
to what I'm trying to accomplish, I can take those on ... after I accomplish my main goal.
Sociotechnical Tooling #
Reduce the blast radius #
Did I set up a mechanism to prevent this from impacting people outside the scope I want? Not just end users but co-workers who have to fix the problem!
Feature flags are friends.
Escape hatches #
If this turns out bad, how do I easily recover?
Rubrics: What's the bar for "quality?" #
If I'm working on a team, we should agree on a rubric for "good enough". Ain't nobody got time to
constantly re-litigate quality standards, so write them down -- preferably in a machine-readable format
that you can feed into automated tooling like linters.
Can the computer enforce the quality standards?
- The linter can warn "this bit doesn't match the desired style!"
- The linter can probably automatically fix issues.
- The linter should only touch what I actively worked on, so I'm not drawn into boiling the ocean.