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

Ok, and what if those are harmful antibodies instead?

Your hypothetical is just as likely to be harmful instead of helpful, unless there is actual evidence proven otherwise.

The null hypothesis should assume that it is no more likely to be helpful than harmful.




On balance I feel that that hypothetical is more likely to be helpful than harmful. Naturally, since I supported it. But of course I could be wrong. And it's very easy I think to say "let's be cautious" because we don't know. And I think it's a valid criticism to say that position is not adding any value. It's not. But maybe it is trying to preserve some value, and just because it's easy doesn't make caution an unworkable strategy.

But in the worst case caution could be a harmful position. So in these tech matters I think it's adaptive that we disagree, since if we were (hypothetically) both on the same ecosystem-engineering team, maybe our diverse opinions would contribute to a balanced strategy that I think can have the best chance of greatest net benefit.

Interesting point about null. How can we factor in risks like 'unknown unknowns' into null hypothesis thinking? I don't know right now. But I think one issue/limitation with null hypothesis thinking is that it is not imaginative. This makes it very applicable for assessing evidence for something. But not applicable for thinking about the unknown risks that could occur for something. Different thinking frameworks for different problems, not one-sized fits all, I think. But I think it is applicable for you to use it here to advocate your position of pushing forward unless we have evidence otherwise. My position differs in that I'm saying I think we need to be imaginative about the potential risks.




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

Search: