On Mon, Dec 2, 2019 at 3:49 AM Jan Høydahl <[EMAIL PROTECTED]> wrote:
Then the number one thing to do is fix the tests, so that they can be
trusted and leaned on more to detect these bugs. As someone who wrestled
with these tests recently... in all honesty something drastic needs to
happen (e.g. throwing them all out and starting over).
People are always going to make mistakes. These mistakes will sometimes
slip past reviewers, too. No matter how much process and time you throw at
But i guess "policy" is sexier to work on than tests.