> > There is no way to predict where these bugs will be. Why are the tests > > slow to run? Is it due to small sleeps in the test themselves that > > gradually add up? If so, what can be done to remove those? > > They're not "slow"; they take time.
OK -- but I think it's still important that we be able to justify each and every sleep or delay in the test suites. (Not only for performance reasons -- I have seen many cases where these delays were hiding bugs or deficiencies in the thing being tested.) -- meem
