On the contrary, after reading several articles by you, it is you that are the fanatic.

My point in testing is to try to find as many bugs as a possibly can - no more and no less.

This means that some code has very little test coverage, because it is obviously correct and has worked since the start. But other code which is difficult and has a lot of edge cases will have a huge amount of testing.

It makes zero difference to me whether that code is public or private.

In fact, a pattern I often see is this. There's a public entry point which starts off with some obviously correct boilerplate involving the database or the Internet - parts which are hard to test.

But once you get past that boilerplate, you get into gnarly logic.

Frequently, I pop out just the logic into a separate function or method, a non-public entry as you'd call it, and then write a bevy of tests for just the logic - the area I believe is more likely to fail.