Last spring, in one pull request, you left a short comment. I would avoid passing booleans into functions here, it makes the calls hard to read. It was a fair point about that one function. Today you see a reviewer blocking a new colleague's change, quoting the same line, and adding that it is a team rule. Someone has even linked your old comment as the source.
It is flattering for about ten seconds. Then you realise that a suggestion about one piece of code has turned into a law, without its reason, and with your name attached.
This happens more often than people think. Feedback from someone experienced carries weight, and people copy what seems to work. A comment that was a preference in context becomes a rule out of context. Nobody decided it. It just spread.
First, step in where it is happening now. Reply kindly on the new pull request. That was my comment, and I meant it for that one case, where three booleans in a row made the call impossible to read. A single flag like this one seems fine to me. That frees the new colleague, and it shows the reviewer that the reason matters more than the rule.
Then decide whether there really should be a rule. Sometimes there should. If the team agrees, write it down properly, in the contributing guide or as a linter setting, with the reason next to it and an example of when it does not apply. A rule with its reason can be questioned sensibly later. A rule without one can only be obeyed or ignored.
If there should not be a rule, say so in the team channel, briefly and without blaming anyone. People were trying to do the right thing.
And change how you write review comments from now on. Make it clear when something is your taste and when it is a real problem. Starting with a word like suggestion, or ending with feel free to ignore, takes a second and stops a preference from growing into policy.
The more senior you become, the further your words travel. Part of giving good feedback is noticing where it ends up.
– Asael Shinder
