I agree with Stuart on this in general but where I disagree is the utility of Lumina if it worked. There are many people in our society that are not capable of effectively cleaning their teeth and preventing cavities. For example, people with disabilities or people with Alzheimer’s.
Also, I hate going to the dentist and many people do and I have had cavities throughout my life. Not a lot but enough that I’d love for this to work. However, I’m not trying it till it goes through some good trials on humans. Not worth it.
I agree with Tom in principle, and Stuart in practice (circa 1:03 onwards). Yes, you can treat everything in a Bayesian way and attempt to put a probability distribution on the benefits for every crackpot idea someone comes up with, and that in some sense may be rational. But there are an infinite number of those and a finite amount of time/attention/resources that can be devoted to things and so it does make sense IMO to have some threshold for consideration.
(From the perspective of a consumer that is: it is of course great that there are crazy scientists who experiment with creating mouth bacteria or whatever).
I’m not steeped in it but often rationalist arguments (especially about long term stuff) sound like they have a Pascal’s Wager flavour to me: “I can think of a scenario in which this intervention saves 10^25 lives and so even if that scenario has 10^-20 probability we should still do it”, whereas IMO, you should probably just round that P down to zero. The arguments for this bacteria (e.g. 54:25) sound like that to me. Open to being told why I'm completely wrong/what I've missed.
Pascals wager and Lumina die-hardism both seem to rely on a type of tunnel vision, where you fixate on on particular god or one particular cure as the saviour. In reality there are more possibly gods than anyone has time in the day to worship, and more possible miracle cures than anyone could possibly have time to do dodgy and uncontrolled, n=1 field trials on. And then you have to worry about interactions... If you try to worship all possibly gods, surely at least some of them will get ticked off about your theological promiscuity. You certainly couldn't try multiply miracle cures at once without interfering with the validity of your data.
I agree with Stuart on this in general but where I disagree is the utility of Lumina if it worked. There are many people in our society that are not capable of effectively cleaning their teeth and preventing cavities. For example, people with disabilities or people with Alzheimer’s.
Also, I hate going to the dentist and many people do and I have had cavities throughout my life. Not a lot but enough that I’d love for this to work. However, I’m not trying it till it goes through some good trials on humans. Not worth it.
I agree with Tom in principle, and Stuart in practice (circa 1:03 onwards). Yes, you can treat everything in a Bayesian way and attempt to put a probability distribution on the benefits for every crackpot idea someone comes up with, and that in some sense may be rational. But there are an infinite number of those and a finite amount of time/attention/resources that can be devoted to things and so it does make sense IMO to have some threshold for consideration.
(From the perspective of a consumer that is: it is of course great that there are crazy scientists who experiment with creating mouth bacteria or whatever).
I’m not steeped in it but often rationalist arguments (especially about long term stuff) sound like they have a Pascal’s Wager flavour to me: “I can think of a scenario in which this intervention saves 10^25 lives and so even if that scenario has 10^-20 probability we should still do it”, whereas IMO, you should probably just round that P down to zero. The arguments for this bacteria (e.g. 54:25) sound like that to me. Open to being told why I'm completely wrong/what I've missed.
Pascals wager and Lumina die-hardism both seem to rely on a type of tunnel vision, where you fixate on on particular god or one particular cure as the saviour. In reality there are more possibly gods than anyone has time in the day to worship, and more possible miracle cures than anyone could possibly have time to do dodgy and uncontrolled, n=1 field trials on. And then you have to worry about interactions... If you try to worship all possibly gods, surely at least some of them will get ticked off about your theological promiscuity. You certainly couldn't try multiply miracle cures at once without interfering with the validity of your data.
Plaque is also bad, it’s not only about cavities.