I often phrase code review comments as questions with specific goals in mind, such as "Would this be simpler/faster/clearer if [refactor idea]?" or "How about [

I often phrase code review comments as questions with specific goals in mind, such as "Would this be simpler/faster/clearer if [refactor idea]?" or "How about [alternative]?" This makes it clear that I, as a reviewer, do not have all the context that the author does, and that it's a good chance they did consider those ideas but rejected them based on previous experience. This may be as simple as a broken third party thing that they had to work around, or prioritizing differently. The only thing this misses is code style, but holy crap automated code formatting tools are such enormous time savers for reviews. Black + isort on Python code and Prettier for everything else makes code reviews easily three times faster.