Given the state of the existing automated tests, perhaps the Principal Engineer is correct… Why would the new automated tests be any better?
Given the state of the existing automated tests, perhaps the Principal Engineer is correct… Why would the new automated tests be any better?
Multiple times in my career I've taken ownership of a codebase written by someone else. Some of the most pleasant ones had very little test coverage, but the code quality was high enough to make the lack of tests not really matter. On the other hand, some code is so bad covering it with tests adds nothing of value; it's just one more thing I have to analyze and decide what to do with.
I agree with most of the author's points in general, but I think that most programmers who are capable of writing good tests are also capable of recognizing situations when it's not worth the trouble.