How do you deal with inconsistency in code reviews
If I submitted the same code for review 10 times I would get 10 different sets of feedback, based on who's reviewing it, time of day, how busy the reviewer is etc. Code reviews are often super helpful, but this aspect of them seems flawed. Anyone else feel this way?
1 comment
[ 3.1 ms ] story [ 7.2 ms ] threadcode reviews are not substitute for a design review; there is a limited amount of mental power that a reviewer can expend. the purpose of code reviews, in my opinion, is to
a) catch any obvious problems
b) propagate a vague understanding of the change among the team members
anything more requires much more work on the part of your team mates and probably needs to be explicitly scheduled (design review, implementation review), possibly in a different format, for example 1:1 conversation or during a lunch.