Hacker News Reader

Comment by coldtea | Hacker News Reader

>Facts beat theory.

Not necessarily. Facts (in the sense of "anecdata") are simple observations.

The only make any sense against or in favor of something when we combine them with a theory why it is so.

OK, students faring better on exams when they are taught with tau is an argument in favor of tau only if we theorize (assume) that it was the tau/pi thing that made the difference.

Whereas it could just be that the second group of students was just better naturally (e.g. a good school vs a mediocre one), or just a chance outcome, due to the small sample we picked, or the teacher going for tau was also better at explaining and would have fared just as good if he also was the one to teach the pi group.

View on HN

It is one of the great errors of the "SCIENCE!" attitude that it convinces people that if we do not have peer-reviewed papers, we must necessarily claim full and total ignorance and berate anybody who attempts to do otherwise.

I'm not obligated to stop using my brain just because we have no SCIENCE! papers to work with. In fact, you can't. You have no choice. You must do your best with the data you have, because we do not have "science" for all the questions we encounter every day.

Given how standardized education is nowadays, as I complained in another comment, it isn't exactly a far-out theory that when the same person, in the same context, switches to using tau, and sees a difference in results, that it may have something to do with using the more mathematically-sensible concept for teaching.

> It is one of the great errors of the "SCIENCE!" attitude that it convinces people that if we do not have peer-reviewed papers, we must necessarily claim full and total ignorance and berate anybody who attempts to do otherwise.

In fact, I think quite the opposite. Its one of the great errors that people assume peer-reviewed papers are always correct, a lot of the time they're not.

We should be sceptical of people's claims when not supported by evidence. That doesn't mean we should ignore said claims, just that we shouldn't blindly accept them. This is the point where we should go and get some evidence, try to get a few teachers to use tau from the beginning and see if that helps.

Though I do agree that we shouldn't be scared of doing things just because we don't categorically know that it is right.

> Given how standardized education is nowadays, as I complained in another comment, it isn't exactly a far-out theory that when the same person, in the same context, switches to using tau, and sees a difference in results, that it may have something to do with using the more mathematically-sensible concept for teaching.

Or thinking about the same concept in different ways help you to understand the concept better. Its certainly something I believe has helped me a lot previously.

Reply on news.ycombinator.com