toonalfrink
Karma: 692
NewTop
The league of Rationalists
Fundamentals of Formalisation level 2: Basic Set Theory
I think this article is a considerable step forward, but it could benefit from some examples. I think I have a pretty good idea what this is about (and share the horror of being called out by a low-decoupler for being some kind of ism), but still.
It seems to make sense to make this graph 2-dimensional, with axes {A} and {B}, and plot Adam-Edgar as points on it. Clearly this isn’t about {B}, and avoiding {X} and {Y} through adjusting {B} is hopeless. Clearly this is about {A}, and the right course of action is to figure out what {A} is.
For most donors, you really do NOT want to give an accounting of the direct value you’ve provided, and instead want to focus on the indirect/long-term/global values that you’re providing.
You’re afraid that this will lead to perverse incentives, correct? Signaling instead of focusing on real impact?
Inviting them to working sessions, though, is confusing. Either they contribute value in these sessions, and you’d want them there even without the donation, or they detract and you’re wasting your time and their money by having them there.
This looks like one of those cases where ignoring desiderata that are hard to measure leads to a skewed decision. The downsides are salient, the upsides are something fuzzy tribal romantic something. I wouldn’t be surprised if the latter is sufficiently motivating for the counterfactual to be no donor rather than an uninvolved donor. Some people want inclusion. I want to include all these excellent people, but resources/logistics don’t allow it. Their donating fixes that problem.
Probably the source of our disagreement is what an organisation is supposed to be. Here’s a helpful distinction: the tribe and the hunting party. The hunting party is lean, maximizing, exclusive and goal-oriented. The tribe is broad, satisficing, inclusive and process-oriented.
It’s easy to find a hunting party, but hard to find a tribe. It strikes me that I’m leveraging this need for a tribe to bolster my hunting party. You may have a point that this is a bad idea. On the other hand, there is evidence that hunting parties with tribal characteristics are more effective. This distinction might not be as useful as our culture might suggest.
So I still think it’s worth a try, though I appreciate your warning, and I’ll keep it it mind.
This is great data. Thanks!
What do you mean?
You shouldn’t do this for everyone
I’m not sure. In my experience, people tend to respond to exceptional treatment with exceptional treatment. That I so readily put trust in people is often perceived as a sign of high social capital. As if my prior is that they will respond in kind, which I’d only have if I was used to that kind of thing.
And the funny thing is, people actually do respond in kind, so it becomes a self-fulfilling prophecy.
One important requirement is that you apply the same kind of respect to yourself. The proper mindset seems to be “we are excellent and both of us deserve the best”, not “you are excellent” or “I am excellent”. Think win-win.
The reverse job
You represented it well. We’re currently doing 2 things at once. The prerequisites track was too good to pass up.
Fundamentals of Formalisation level 1: Basic Logic
I thought generating examples would be trivial.
Someone cooks for another, instead of not doing that. Net social capital increased. Right?
Just an example though.
Hm.
I don’t think the former is free from Goodharting either. My sense of a good community is one where we get the character judgment out of the way from the start. So indeed “people evaluating one another on fuzzy personal criteria”. In the sense of “hey we like you for the things about you you can’t change even if you tried”. So personal value is secured, meaning that the person can actually start to pursue the things they value truly for their own sake.
As I said in other places: If I got all of my needs out of the way, I would still work on AI Safety (which I value for it’s own sake), and have a lot more cognitive bandwidth to allocate to it too. All of which is now going to securing my worth. Which is essentially Goodharting, since I’m incentivized to skew everything I do to things that can be easily used for signaling.
An unsatisfied satisficer is a maximizer. I’m maximizing my status, and the useful work I’m doing is only a side effect. That doesn’t seem like a good thing. Especially with a security mindset.
You seem to be coming from the premise that there is plenty of praise out there, just not in the right places. But the point of the post is that there just isn’t enough praise out there. Gut-level appreciation, the thing I want people to have for me, isn’t zero sum. They can have it for both building things and shiny blogs.
You also seem to assume that we should be using praise as an incentive. I’m on the fence about that. Maybe praise (or let’s call it respect or personhood or appreciation here) should be the bottom level, and people can actually do things for their own worth.
I, for one, actually want things to be built regardless of social incentives, and I imagine being socially “satiated” will give me a lot more resources to actually allocate on building things (especially things that are hard to signal with).
Reminds me of project Hufflepuff. That’s about getting people to do things that are good but hard to signal with, which is impossible if those people have a status deficit.
Huh. Okay. So it’s not about recognizing your worth, but about recognizing the worth of your work, and associating that work with you. Does that capture it better?
But that’s not far from recognizing that good work can be expected from you, which is the thing you actually game-theoretically want, right? That’s what I mean with “good character”, in the sense that the community is incentivized to support you and invest in you.
Maybe that’s the Berkeley optimum, but is it the optimum from an international perspective? My intuition is that 3 dunbar size communities in different cities is still better
Quick thought: since the Berkeley community currently has about Dunbar’s number of people, shouldn’t we want this number to remain stable? Of course new rationalists want in on all the fun and all the impact, and currently it looks like Berkeley is their only option. Should we be focusing on a second hub that aims to rival Berkeley in it’s size and awesomeness?
What places should we be recommending? I’m thinking London or Berlin. Any other options?
Yes. This. Very much this. I get a sense that recognition, as opposed to positive judgment, is more durable. It’s not just that you did a good thing, it’s that the thing you did is a reflection of your good character, and we expect you to do more good things, and we want to keep you around and support you.
Uh, well I don’t know you, but it seems unlikely that anyone would deny an argument just because it’s conclusion (vaguely) implies that you should be regarded with respect.