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

Good point. To disagree so strongly when anectodal evidence is shared regardless of context became a culture in itself rather than a model for rational behaviour and is disruptive.

Academic science is an entirely different story. It's a growing body, and it grows by using previous information in its body, so any probability of disruptive/wrong data entering the system has to be reduced as much as it can with strict discipline. To apply the same standard to our daily lives is absurd - you could as well be against spoken conversation because it's too lossy and converse using pen&paper - speech recognition is tricky after all.

This doesn't take much more than common sense to get - you could formalize it for specific cases statistically, maybe using shannon entropy and such - anectodal information is still information and the loss of information by virtue of not being backed up by scientific method is in most cases insignificantly low - just think of it as an efficient compression algorithm.

If you aren't ethically against lossy audio/image/text compression, there is no reason to take such a strict stance against anectodal information. Most times it's very useful, just have your rationality filter/bs detector and the like on at all times like they should be.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

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

Search: