Re:Modern Code Review: A Case Study at Google [pdf]
by bunderbunder on Friday September 21, 2018 @08:41AM
(#18039235)
There's a 4th that I find is, at the very least, incredibly useful for mentoring and knowledge sharing: Comments to the effect of, "Did you know about this other way to do that? It would have X effect."
With a clear understanding that it is up to the person submitting the code review to decide whether or not to make the change. 'Cuz, I agree, arguing about stylistic things after-the-fact tends to do more harm than good.
Instead I've had a lot of success limiting code reviews to three types of critiques.
1. This code is broken. It does not do what it's supposed to do in a way that will affect current users.(avoid what if scenarios)
2. The code does not meet a clearly delineated style guideline that the team has agreed to.
3. The code does not meet our clearly defined checklist for possible bugs.
The idea is to keep the code reviews objective and boring. People can fight religous wars over the style guidelines, but not over any given lines of code.