EP: 19, part2 - the Venn diagram

[Jop van Raaij]

Hi Uncle Bob,

A question/remark on the Venn diagram at the end in episode 19, part 2. I like this kind of view on the material: getting a more abstract overview of what we're actually doing.

I don't think the explanation in the video is wrong, but in my opinion the example is unlikely. I'd expect the first test-case to at least touch the boundary of the circle of desired behavior (maybe with strong concave shape,'pointing' to the circle). With each test case, a section of the boundary is 'guarded'. A quote from the video: "You test all the boundaries and error conditions first". Or is this an other 'boundary'?
As we are writing application code only to make the test pass, I'd expect the application code to share part of the boundary with the test. In the video the application code after the first test happens to be a little bit in the desired area, but this is not forced by the test. Using the same test, it could be passed by writing code which is completely outside the circle of desired behavior.

Maybe I'm missing something?

I agree on the fact test can't specify the behavior of the application, but only constrain. If the application code took the opportunity to get into the area outside of the circle of desired behavior, but inside the test area, would that be classified as 'bug'? The application in the video is a bug-free application?