Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

How about "tests with an obviously high risk of systematic, invisible false negatives are bad"


Is it a false negative if the point of the test is to filter out people who will immediately assume that anyone who likes Atlas Shrugged is someone that they don't want to know better?

It's not a filter I would choose to use, I'm just saying that OP isn't providing a failure case—the filter is still doing its job there.


Few books are more indicative of people I'd like to filter out immediately than Atlas Shrugged. It's about as close to saying "screw other people" as possible.


Yeah I guess if their filter is literally "will this person stop talking to me if I like Atlas Shrugged" and not actually something like "will this person engage in respectful disagreement with me."

Of course the first test is useless, so I suspect they're trying and failing to produce the second test.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: