This seems like a very valuable bit of info to have!
I haven’t thought about this for long, but I’m not actually sure which way a system like this would go.
I know that when a bunch of random variables are added together then that will result in a normal distribution, and when a a bunch of random variables are multiplied together it’ll give a log-normal distribution, but I don’t know which of these is a better model for different pain intensities in a given pain event
I love seeing this stuff written up! Thanks for taking the time to write it. This way other orgs can build-on and modify this kind of system for their own needs, rather than everyone having to start from scratch
I felt the Sam Harris interview was disappointingly soft and superficial. To be fair to MacAskill, Harris did an unusually bad job of pushing back and taking a harder line, and so MacAskill wasn’t forced to get deeper into it.
And basically nothing about how to avoid a similar situation happening again? Except for a few lines about decentralisation. Quite uninspiring.
Thanks a lot for your thoughts!
I like your first point, that the intervention with the greatest reduction in highest intensity pain will always be in the Pareto frontier (as the 6th image shows, any point to the left of another will either be “unclear” or “strictly worse”, but never “strictly better”). I hadn’t considered that before!
I don’t think I understand your second or third points though.
Namely this line:
But I think we can also use it to achieve the prerequisites required for estimating better Pareto frontier data.
And this line:
One caveat though is change in behavioral patterns after long periodical treatments. For example: the case of cattle would be different from hens and chickens
I don’t understand the distinction you’re making between cows and chickens.
Thanks again for your comment!
Personally, I know I’d rather make the decision of donating once, and forget about it, rather than having to decide every day. I know if I had a daily notification asking me to make a decision I would quickly get rid of it, as I find those sorts of things exhausting after a while. But maybe there’s a personality of people who like these sorts of prompts and appreciate micro-decisions.
On the other hand, maybe some people are overwhelmed by the idea of donating $1500/year, but $5/day seems fine, so they’d be into this.
For people who like the sound of $5/day but would soon mute the notification, there could be a way of toggling a switch to automate the process of donating the 5 bucks, but maybe that’s missing the point of your idea?
You’d probably need a way of getting around credit card fees, as I think these have a fixed price so making lots of small transactions isn’t very effective?