> If you're doing up a new port, and want it to work well, then the > absolute, there should be no unexpected failures is a good place to > aim for.
It's a wonderful absolute, but not always practical. I keep a history of test results for my ports, with a goal of "no regressions" relative to those. It still leaves the "where do you start" question unanswered. I try to balance "needs to work" with "sometime this century" ;-)
