Comment by matijsvzuijlen

The idea is that --amend would be used before pushing the commit that it amends, and so before a reviewer ever sees the original commit.