Hindsight is 2020
Jul. 5th, 2021 11:03 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
brin_bellway[cw: illness, (mild) discourse]
Me: *re-reading old posts*
Past!me: Even if I did the evil thing and tried to continue going about my life while sick, often you *can't*. I have very little experience with remaining functional while depressed, and even people with lots of experience often fail at it.
Me: Oh honey, experience isn't why they're more functional while sick than you are. I can see why you guessed that, but no.
---
A couple weeks later (than the re-reading me, not the writing me), I was remembering some discourse about why young people would ever agree en masse to get COVID vaccines when, for them, the side effects of the shot would be as bad as the disease.
The people arguing back were attacking the premise (and indeed, as [awareness that COVID-19 has high morbidity even in the young] increases this discourse seems to have dropped off AFAICT, though it's possible it simply moved out of sight). And while I agreed that the premise was wrong, I *also* thought that the conclusion didn't follow: being sick is inherently bad above and beyond its level of suffering, so if you're going to have to suffer anyway it's better for it to be from a vaccine than from a virus.
Looking back on that, my metacognition is going '...yeah, it makes perfect sense that we'd develop that valueset, huh. Like, to an *extent* *everyone* should think that way because "suffering + risk of spreading it to others" really is strictly worse than "same amount of suffering + no risk of spreading it to others", but also: historically, when people tell us--usually implicitly--"X disease and Y non-disease are equally bad", it's a safe bet that X is worse, and probably *much* worse. The training data we received from social reality had a consistent pro-disease bias relative to the training data we received from physical reality, and the simplest way to compensate for that hidden bias and make the decision model match up was to add a negative modifier on disease-per-se.'
Do I plan to change my values in this area? Not really. I take a fairly enlightened-self-interest approach to making the world a better place, and "making the world more like the kind of world I want to live in" necessarily implies an unusually (relative to other people) large focus on infectious-disease prevention.
(Conveniently for me, it so happens that ~all of the lowest-hanging fruit in general utilitarianism right now is in infectious-disease prevention, so I can generally safely ignore the values gap between me and GiveWell: it's possible that their recommendations aren't *precisely* the best from my perspective, but they're bound to be pretty close. Not sure what I'll do if and when there's a significant shift in the lowest-hanging cause area: probably depends on what it shifts to.)
Me: *re-reading old posts*
Past!me: Even if I did the evil thing and tried to continue going about my life while sick, often you *can't*. I have very little experience with remaining functional while depressed, and even people with lots of experience often fail at it.
Me: Oh honey, experience isn't why they're more functional while sick than you are. I can see why you guessed that, but no.
---
A couple weeks later (than the re-reading me, not the writing me), I was remembering some discourse about why young people would ever agree en masse to get COVID vaccines when, for them, the side effects of the shot would be as bad as the disease.
The people arguing back were attacking the premise (and indeed, as [awareness that COVID-19 has high morbidity even in the young] increases this discourse seems to have dropped off AFAICT, though it's possible it simply moved out of sight). And while I agreed that the premise was wrong, I *also* thought that the conclusion didn't follow: being sick is inherently bad above and beyond its level of suffering, so if you're going to have to suffer anyway it's better for it to be from a vaccine than from a virus.
Looking back on that, my metacognition is going '...yeah, it makes perfect sense that we'd develop that valueset, huh. Like, to an *extent* *everyone* should think that way because "suffering + risk of spreading it to others" really is strictly worse than "same amount of suffering + no risk of spreading it to others", but also: historically, when people tell us--usually implicitly--"X disease and Y non-disease are equally bad", it's a safe bet that X is worse, and probably *much* worse. The training data we received from social reality had a consistent pro-disease bias relative to the training data we received from physical reality, and the simplest way to compensate for that hidden bias and make the decision model match up was to add a negative modifier on disease-per-se.'
Do I plan to change my values in this area? Not really. I take a fairly enlightened-self-interest approach to making the world a better place, and "making the world more like the kind of world I want to live in" necessarily implies an unusually (relative to other people) large focus on infectious-disease prevention.
(Conveniently for me, it so happens that ~all of the lowest-hanging fruit in general utilitarianism right now is in infectious-disease prevention, so I can generally safely ignore the values gap between me and GiveWell: it's possible that their recommendations aren't *precisely* the best from my perspective, but they're bound to be pretty close. Not sure what I'll do if and when there's a significant shift in the lowest-hanging cause area: probably depends on what it shifts to.)
no subject
Date: 2021-07-19 08:49 am (UTC)