[pydry]

>Every change would inevitably break tests (not functionality!) and developers were spending more than half their time wrestling with the test suite.

I find that this invariably happens because the team tries to test implementation at a very low level rather than behavior at a high level.

I blame an overemphasis on unit tests over integration testing (e.g. J B Rainsberger's inane rant) and tutorials that imply that, for example, if you build a class you have to have some tests for that class.

[icedchai]

At a previous company, someone established a "rule" that every "new function or method" needed its own test. Unless you lucked out with a lax reviewer (like me!) , your code would not pass review if this wasn't followed. Meanwhile, there were basically zero integration or e2e tests. People spent more time adding and maintaining useless tests...