While I haven't contributed to the P6 discussion for a while (thus, have no comment), this pet peeve is one that's shared by CPAN developers as well. Over the years, I've received a lot of "This is busted" bug reports. As I use nearly all my modules and it was broken, it's obviously a situation I hadn't run into yet. (If I had, it wouldn't be broken now, would it?!)

Since it's a situation I hadn't run into, what's the incentive for me to write the failing test that keeps the problem from returning? I'm not the one with the bug ... you are. The code is failing your expectations, not mine. By writing the failing test, you are forcing me to look at the bug. More importantly, you've given me a repeatable yardstick by which to measure my progress.

One of the biggest reasons Pugs such a success is because the developers of Pugs don't write tests. (Audrey is actually (in)famous for not writing them, much to my dismay.) They have a legion of test-writers who contribute hundreds and hundreds of failing tests. It is so much easier to write code to pass a series of tests than it is to write code that is just wandering out into the blue.

Moral of the story: If you don't like something, write a failing test. Let that be your bitchfest.