Most developers expect a test to fail. Ted M. Young argues that the real value comes from predicting exactly how it will fail. When the result doesn’t match your expectation, you’ve uncovered a misunderstanding before it becomes a bug.
3 reposts2 likes