Hacker News new | comments | ask | show | jobs | submit login

Most disciplines where correctness is important seem to end up having some adversarial component. It is explicitly how the justice system in the US works [1]. Many software companies have separate QA departments that are deliberately kept at a remove from the engineers to encourage some rivalry between them. Security issues are almost always managed in an adversarial way (though here, you could argue that's because it reflects how the system itself is [mis-]used). Markets are intended to allow fair competition between producers to find an optimal price and product for consumers.

Peer review is supposed to do this, but the fact that peer reviewers are often colleagues leads to collusion, whether intended or not.

Maybe we need a separate body of scientists whose sole job—and whose entire prestige—derives from taking down and retracting bad science.

[1]: https://en.wikipedia.org/wiki/Adversarial_system

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