It seems like the main issue here is a disconnect between how 80k is generally described, including in its own guide (“career advice for EAs”/”how to use your career to increase your impact”), and 80k’s own internal vision of what it is (“how to solve important problems by directing talented people at them”). It seems that the former is a misrepresentation of the latter and people including 80k should stop misrepresenting it.
tcheasdfjkl
Karma: 203
“Mikhail Yagudin ($28,000): Giving copies of Harry Potter and the Methods of Rationality to the winners of EGMO 2019 and IMO 2020”
Why does this cost so much?
This is great, thanks.
There are not currently a bunch of openings (probably because eligibility just expanded).
How far in advance would you expect US officials to warn the public of the possibility of nukes? (i.e. how much time would we have between such a warning and needing to have left already?)
Huge thanks for spelling out the specific allegations about SBF’s behavior in early Alameda; for the past couple days I’d been seeing a lot of “there was known sketchy stuff at Alameda in 2017-18” and it was kind of frustrating how hard it was to get any information about what is actually alleged to have happened, so I really appreciate this clear point-by-point summary.
I think there’s at least some difference between [money you’ve already been paid for work you’ve already done] and [money you’ve been granted for future work that you have not yet done / money you haven’t yet spent on hiring other people / etc.]; I very clearly agree with you in the first case and think it’s at least murkier in the second case.
That’s very surprising!!
Do you know if anybody attempted to propagate this information to any of the EAs who were promoting SBF publicly? (If so, do you know if they succeeded in conveying that information to them?)
And just to check, did any of the people who warn you privately promote SBF/FTX publicly?
I ask because it seems weird for a lot of EAs to be passing around warnings about SBF being untrustworthy while a lot of (other?) EAs are promoting him publicly; I very much hope these sets were disjoint, but also it’s weird for them to be so disjoint, I would have expected better information flow.
To be fair sometimes people make accusations that are incorrect? Your decision procedure does need to allow for the possibility of not taking a given accusation seriously. I don’t know who knew what and how reasonable a conclusion this was for any given person given their state of knowledge, in this case, but also people do get this wrong sometimes, this doesn’t seem implausible to me.
Strongly disagree with most of this but upvoting because I am actively in favor of people laying out their reasoning for unpopular positions so that people can engage with the reasoning directly, rather than only operating only on the level of “this is wrong and you should feel bad”.
I really love the upvote:agreement ratio on this comment. There is some disagreement as to whether this is a good comment, but everybody agrees that what the fuck.
But Kelsey said in her email that she was going to write about their conversation, and he didn’t object. What do you think his epistemic state was, if he knew she was writing about the conversation but objected to the actual damning things he said being included? It seems like for those things to both be true, it would have to be the case that he expected her to write a piece that somehow left out the most damning things, i.e. to write a weirdly positively distorted piece.
I think we should definitely not be loyal to people who commit massive fraud, or praise ambitious destruction! I know “stand fully by my people no matter how right or wrong they are” is a common moral stance but I think it’s enormously wrong and destructive. It’s an important virtue to support things that are good and not things that are bad, even if we’re very attached to them. (Also, like, I think SBF betrayed “us” first.)
(Sorry this is more of a skeleton of an argument than an actual argument, I keep meaning to write out more of my thinking here and not finding time)
I guess he could have also not been reading carefully and missed that somehow?
Yeah I wouldn’t upvote every post that makes an unpopular argument. But I upvote posts that I want to encourage on the margin, which includes posts that are wrong but better to make than not to.
or like, if you’re close with someone who did a significant bad thing and is now facing significant consequences for it, it can make sense to be loyal in the sense of—trying to help them make it through this time, trying to not make things worse for them. but not in the sense of denying or defending their wrongdoing.
Yes! I have independently discovered this exact same thing for myself, though my terminology is almost the opposite of yours—I think I had always thought of “rest days” as days mostly spent in front of the TV or online (corresponding to your Recovery Days), whereas what I now realize I need to do regularly is “doing whatever I want” days, which I also sometimes call “desire days”—I end up doing a lot of things I don’t really think of as rest but which are much more restorative than just resting.
(If I’m really tired or anhedonic on a designated “desire day”, I end up just doing “rest day” stuff anyway; sometimes this is the right choice, other times it doesn’t really help but it’s better than doing “rest day” stuff on a day I had meant to get stuff done, so it’s possibly still better than not setting that day aside.)
Some things I’ve ended up doing on desire days:
cooking, baking, making jam
putting on some music, opening the windows, and dancing around my living room
reading French poetry
learning new songs on guitar
having a somewhat useful emotional crisis
doing some self-therapy, like writing down my thoughts / working out what I care about
figuring out how much I owe in donations this year
reading and taking notes on Nate Soares’ Replacing Guilt series
reading the 80K career guide
playing a typing game
doing things that I’m annoyed about having put off for a long time
fiddling with spreadsheets to make a useful graph of my mood-tracking data
proofreading things for people
laundry