LessWrong team member / moderator. I’ve been a LessWrong organizer since 2011, with roughly equal focus on the cultural, practical and intellectual aspects of the community. My first project was creating the Secular Solstice and helping groups across the world run their own version of it. More recently I’ve been interested in improving my own epistemic standards and helping others to do so as well.
Raemon
Karma: 66,015
Oh, this was probably because I forgot to put in a street address. Looks fixed now.
My version of this is “don’t try to come up with a name until after you’ve found the venue, because the venue will have some kind of character that lends itself to some names better than others.”
One could go with Persepolis Vibes:
Huh. The combo of this + OpenAI’s Frontier Safety Blueprint in the same week is surprising.
They both are saying more of the things I’d have been wishing for Anthropic and OpenAI to be saying all along. I’d like to just believe “oh, they wanted to say these things all along, and were waiting till the political winds felt favorable.” But, man I feel like if that was the case, at least some of their previous actions would have been different somehow. (OpenAI in particular).
Curated. I’d thought about most of these in isolation before, but found it valuable to have them in one place while sizing up “what actually matters most about cybersecurity in the AI era?.” (Tracking multiple concerns but keeping your eye on the most-important-balls seems like a good habit).
Being able to pin this down exactly is kind of an open research question. (i.e. “what is an agent?” and “what is an optimizer”). But, roughly, things are “more optimizer-like” the more they successfully converge on a target no matter what starting conditions you put them in, and no matter what obstacles you throw in their way.
Some previous posts:
LLM base models in their raw form are less “optimizer-y” because, while clearly intelligent, if you change their initial prompt they will end up doing radically different things instead of converging to the same thing. (compared to AlphaGo which always tries to win games of go).
Curious if you can say more about the diff. Also, is this true across karma scores? (i.e. does the rest of the voting population seem to vaguely agree/disagree with you?)
(strong upvoted because I’d also like an answer to this question)
I don’t think this particular one is about Bay culture. Or, like, Bay Culture might be the sum-of-the-parts here, but, it’s more like I disagree fractally with you both aesthetically and logistical-preferencelly. I enjoyed the humming thing the very first time it happened to me because it’s beautiful and warm. It sounds like you don’t find it beautiful and warm, just annoying.
Have you actually events and successfully quieted 100 people via the “talk to them all individually?” way?
I think the triggeredness is a bit about “musicalness is important to me”, but also like this is disrespecting my time/effort as an organizer, and the vibe I’m trying to create when I’m running an event.
When I imagine trying to do this via talking to individually people it doesn’t just feel “it’d take longer”, it’s more like “I don’t think that would even work.” Everyone would keep talking loudly. Eventually when I’m actually ready to start I’d still need to do something loud and obnoxious to get everyone to stop and people would still keep talking and I’d have to keep doing something loud and obnoxious until they all became silent.
The event organizer is usually frazzled and busy. The humming thing takes like… 8-12 seconds? I don’t think it even really takes more time for each person than it would to talk to each person?
And, when I imagine the reason people being upset about it being because they wanted to keep a conversation going, I’m like “but, you aren’t supposed to be having that conversation anymore, that’s the point. This room is now about whatever-the-next-activity is.” (Generally it’s known that there’s an activity that’s about to start when people do the humming thing).
Fwiw if I were organizing an event I would want people ~silent for each of those, and I definitely want everyone to be literally silent for at least like 10 seconds so I can verify that people are paying attention.
I’ve never seen the humming used except in situations where the point was to explicitly change the vibe of the room from “people are doing whatever” to “the activity we were waiting for is beginning and there is now a shared intention of some kind.” And people who don’t want to be part of that intention are more like supposed to leave the room if they don’t want to be part of it.
But usually even then you can just walk around saying “I’ll be doing announcements in one minute, be ready” to people, and then a minute later it’s real easy to get peoples’ attention when you stand in the middle and ask for it.
This also seems like it takes way longer than the humming thing?
...
I’d agree “Clap once / twice / three times if you can hear” is faster, and is what I use for quick logistical announcements. The humming thing is specifically when I want to shift the energy to something more relaxed and contemplative.
[sorry I don’t know why I’m so randomly triggered by this]
aims to shut everyone up when usually you just need critical mass.
I’m a bit confused why you think this – every time I’ve ever done this it’s definitely been the goal to shut everyone up (so everyone can hear whatever the announcement is). I’m confused when this has happened to you where that wasn’t the goal?
I think I mostly figured out this sequence wasn’t going to accomplish the implicit goal I had at the time (which was “get some particular people to stop being annoying, by highlighting to multiple they were bickering with or dismissing were approaching a different part of the coordination frontier from a different angle”).
Also, a lot of the answer to the problem here was more like “factor problems into spaces such that it doesn’t matter if people disagree”, which is a good answer but like, doesn’t really need the theory the sequence was trying to build towards.
It looks like you are tracking these sorts of considerations, but, to spell out:
It’s fairly loadbearing for how Lighthaven works to have a fundamental vibe of “be willing to be weird”. You can copy a lot of the surface features for a DC-Area Lighthaven, but, I think it’d fundamentally a different product. Either you need to find a way to make that weirdness work in DC, or you need to find a different “source of soul.” Naively mashing the vibe into DC probably wouldn’t work.
Or, different framing than “weird.” Lighthaven is about enabling people to think original good thoughts. Original thoughts will tend to be outside current cultural/overton windows – that’s why they’re original. It’s hard to optimize for thinking good thoughts, and for interacting with power structures, at the same time. And I think it’s also hard to optimize a space for having a vibe of being intellectually generative while also playing well with politics.
None of this means this project can’t work, but, idk it might be helpful think of this more like “building the Anti-Lighthaven” than building Lighthaven East.
Good event spaces don’t need to be a clone of Lighthaven, but, event spaces as good as Lighthaven need to have some kind of soul. I think a key to whether this project succeeds will be whether you have a good sense of taste and vision for how to build a soul that is appropriate to your local situation.
If you’re not aiming for “being Lighthaven, or Lighthaven-tier”, and just being “a pretty good event center”, that also is fine, but, if that’s the case I think calling it “Lighthaven East” will be misleading both to yourself and other people about what you’re trying to achieve.
(I say all this from the perspective of framing the problem to solve, not as like a criticism. As I said, the vibe of this post seems to be at least somewhat tracking the above concerns. But they might be more important and higher-magnitude than you’re realizing.
(I find myself wishing I has a shorter version of Subskills of “Listening to Wisdom” to link to, but, like, I think there is some tacit-soulful-knowledge here that is difficult to grok if you haven’t had a particular set of experiences)
...
(Separate from “having a particular soul”, I also just want to reiterate “Every piece of Lighthaven had a ton of effort put into making it beautiful and functional.” We spent a year on construction, rebuilding it to be the nook shaped, Christopher-Alexander-vibed space it ultimately was, and we’ve continued to focus on “Project Delight-haven” to push it further)
I think if he had better taste in claims being disruptive to LW as a whole, this would have been good or neutral instead of bad.
blrg no, that’s a bug affecting events for a “global” audience. Hopefully will fix soon.
In addition to this, I’d add: there is a valid role for pointing out that people’s claims are confused in a way that is more-central-than-they’d-like-to-admit. But, Said didn’t have good enough taste in that for it to seem net-valuable to me.
(I have only skimmed the OP and don’t expect to engage much since moderators have already spent far too many hours arguing about this)
This wins some kind of elegance-to-plausibly-useful ratio award for simple advice that I’ve heard lately. I’mma give this a try.
Well, one thing is that last year, when “AI psychosis” became a term, the people treating the AIs as people just actually were unhinged seeming. And, it was probably incorrect to model the AIs as moral patients at that point.
Or, at least, all of people’s intuitions about what kind of moral patient they might be, were probably quite wrong, if they were treated the sorts of words they said as much evidence. (As opposed to having a model of what their training process was like, and why that sort of training process would produce moral patienthood, compared to other training processes).
Ah thanks. Don’t have further thoughts at the moment but that makes sense.
Curated.
When I first opened this post, I skimmed the intro, and thought “this is a cute idea but seems crazy and I don’t believe it can work. Mnemonics for 19,000 genes? No way!”, and I closed the tab.
When I saw it got 200 karma I took a second look. And… well okay this still seems kinda crazy and I’d like to see someone else use the browser extension and see if it actually helps.
But, I read the “gender = protein transmembrane status” bit, and had a sort of sinking feeling I was about to wrong and a rising feeling of excitement, that, it doesn’t actually take that many dimensions/gradations to get enough bits to specify one guy within 19,000. And the dimensions do seem like categories that leverage my human-racial-bonus-to-identifying-people.
It may just be a cute idea. But I feel like I learned a potentially generalizable tool for mnemonics. I don’t particularly need to memorize protein-coding-genes. But, this has me vaguely excited to try and memorize something. :P