Picture Frames, Window Frames and Frameworks 2019-11-03T22:09:58.181Z · score: 20 (5 votes)
Healthy Competition 2019-10-20T20:55:48.265Z · score: 56 (20 votes)
Noticing Frame Differences 2019-09-30T01:24:20.435Z · score: 137 (49 votes)
Meetups: Climbing uphill, flowing downhill, and the Uncanny Summit 2019-09-21T22:48:56.004Z · score: 22 (5 votes)
[Site Feature] Link Previews 2019-09-17T23:03:12.818Z · score: 35 (9 votes)
Modes of Petrov Day 2019-09-17T02:47:31.469Z · score: 68 (26 votes)
Are there technical/object-level fields that make sense to recruit to LessWrong? 2019-09-15T21:53:36.272Z · score: 26 (10 votes)
September Bragging Thread 2019-08-30T21:58:45.918Z · score: 52 (15 votes)
OpenPhil on "GiveWell’s Top Charities Are (Increasingly) Hard to Beat" 2019-08-24T23:28:59.705Z · score: 11 (2 votes)
LessLong Launch Party 2019-08-23T22:18:39.484Z · score: 13 (4 votes)
Do We Change Our Minds Less Often Than We Think? 2019-08-19T21:37:08.004Z · score: 21 (3 votes)
Raph Koster on Virtual Worlds vs Games (notes) 2019-08-18T19:01:53.768Z · score: 22 (11 votes)
What experiments would demonstrate "upper limits of augmented working memory?" 2019-08-15T22:09:14.492Z · score: 30 (12 votes)
Partial summary of debate with Benquo and Jessicata [pt 1] 2019-08-14T20:02:04.314Z · score: 90 (27 votes)
[Site Update] Weekly/Monthly/Yearly on All Posts 2019-08-02T00:39:54.461Z · score: 36 (8 votes)
Gathering thoughts on Distillation 2019-07-31T19:48:34.378Z · score: 36 (9 votes)
Keeping Beliefs Cruxy 2019-07-28T01:18:13.611Z · score: 53 (21 votes)
Shortform Beta Launch 2019-07-27T20:09:11.599Z · score: 71 (19 votes)
Can you summarize highlights from Vernon's Creativity? 2019-07-26T01:12:31.724Z · score: 16 (4 votes)
"Shortform" vs "Scratchpad" or other names 2019-07-23T01:21:48.979Z · score: 15 (2 votes)
Should I wear wrist-weights while playing Beat Saber? 2019-07-21T19:56:54.102Z · score: 8 (2 votes)
Robust Agency for People and Organizations 2019-07-19T01:18:53.416Z · score: 53 (19 votes)
Doublecrux is for Building Products 2019-07-17T06:50:26.409Z · score: 31 (9 votes)
"Rationalizing" and "Sitting Bolt Upright in Alarm." 2019-07-08T20:34:01.448Z · score: 31 (11 votes)
LW authors: How many clusters of norms do you (personally) want? 2019-07-07T20:27:41.923Z · score: 40 (9 votes)
What product are you building? 2019-07-04T19:08:01.694Z · score: 41 (22 votes)
How to handle large numbers of questions? 2019-07-04T18:22:18.936Z · score: 13 (3 votes)
Opting into Experimental LW Features 2019-07-03T00:51:19.646Z · score: 21 (5 votes)
How/would you want to consume shortform posts? 2019-07-02T19:55:56.967Z · score: 20 (6 votes)
What's the most "stuck" you've been with an argument, that eventually got resolved? 2019-07-01T05:13:26.743Z · score: 15 (4 votes)
Do children lose 'childlike curiosity?' Why? 2019-06-29T22:42:36.856Z · score: 44 (14 votes)
What's the best explanation of intellectual generativity? 2019-06-28T18:33:29.278Z · score: 30 (8 votes)
Is your uncertainty resolvable? 2019-06-21T07:32:00.819Z · score: 32 (17 votes)
Welcome to LessWrong! 2019-06-14T19:42:26.128Z · score: 94 (48 votes)
Ramifications of limited positive value, unlimited negative value? 2019-06-09T23:17:37.826Z · score: 11 (6 votes)
The Schelling Choice is "Rabbit", not "Stag" 2019-06-08T00:24:53.568Z · score: 108 (44 votes)
Seeing the Matrix, Switching Abstractions, and Missing Moods 2019-06-04T21:08:28.709Z · score: 32 (20 votes)
FB/Discord Style Reacts 2019-06-01T21:34:27.167Z · score: 77 (19 votes)
What is required to run a psychology study? 2019-05-29T06:38:13.727Z · score: 32 (10 votes)
What are some "Communities and Cultures Different From Our Own?" 2019-05-12T22:03:42.590Z · score: 31 (14 votes)
The Relationship Between the Village and the Mission 2019-05-12T21:09:31.513Z · score: 139 (38 votes)
How much do major foundations grant per hour of staff time? 2019-05-05T19:57:42.756Z · score: 24 (6 votes)
Bay Summer Solstice 2019 2019-05-03T04:49:10.287Z · score: 27 (6 votes)
Open Problems in Archipelago 2019-04-16T22:57:07.704Z · score: 48 (15 votes)
Robin Hanson on Simple, Evidence Backed Models 2019-04-16T22:22:19.784Z · score: 44 (11 votes)
How do people become ambitious? 2019-04-04T19:12:26.826Z · score: 58 (18 votes)
LW Update 2019-04-02 – Frontpage Rework 2019-04-02T23:48:11.555Z · score: 52 (12 votes)
What would you need to be motivated to answer "hard" LW questions? 2019-03-28T20:07:48.747Z · score: 48 (14 votes)
Do you like bullet points? 2019-03-26T04:30:59.104Z · score: 55 (20 votes)
The Amish, and Strategic Norms around Technology 2019-03-24T22:16:04.974Z · score: 116 (46 votes)


Comment by raemon on What I’ll be doing at MIRI · 2019-11-20T00:18:51.167Z · score: 5 (2 votes) · LW · GW

Probably not that important for this comment, but pretty important in terms of "is voting currently badly broken." Thanks for the heads up!

Comment by raemon on Raemon's Scratchpad · 2019-11-17T00:54:25.620Z · score: 3 (1 votes) · LW · GW

Fair concerns. A few more thoughts:

First, small/simple update: I think the actual period of time for "canonization" to be on the table should be more like 5 years. 

My intent was for canonization to be pretty rare, and in fact is mostly there to sort of set a new, higher standard that everyone can aspire to, which most LW posts don't currently meet. (You could make this part of a different process than a yearly review, but I think it's fairly costly to get everyone's attention at once for a project like this, and it makes more sense to have each yearly review include both "what were the best things from the previous year" as well as even longer term considerations)

Why have Canonization?

Furthermore, I think a binary system is quite antithetical to how intellectual progress and innovation actually happen, which are much more about a gradual lowering of uncertainty and raising of usefulness, than a binary realization after a year that this thing is useful.

I do think this how a lot of progress works. But it's important that sooner or later, you have to update your textbooks that you generally expect students to read. 

I think the standards for the core LW Library probably aren't quite at the level of standards for textbooks (among other things, because most posts currently aren't written with exercises in mind, and otherwise not quite optimized as a comprehensive pedagogical experience)

Journal before Canon?

Originally, I included the possibility of "canonization" in this year's review round because longterm, I'd expect it to make most sense for the review to include both, and the aforementioned "I wanted part of the point here to highlight a standard that we mostly haven't reached yet."

But two things occur to me as I write this out:

1. This particular year, most of the value is in experimentation. This whole process will be pretty new, and I'm not sure it'll work that well. That makes it perhaps not a good time to try out including the potential for "updating the textbooks" to be part of it.

2. It might be good to require two years to for a post to have a shot at getting added to the top shelf in the LW Library, and for posts to first need to have previously been included

2. You choose a much more subjective process, and this leads to either the measure being more about prestige than actual goodness, making the process highly political, as much about who and who isn't being honored as about the actual thing its' trying to measure(Oscars, Nobel Prizes), or to gradual lowering of standards as edge cases keep lowering the bar imperceptibly over time (Grade inflation, 5 star rating systems).

I agree that these are both problems, and quite hard. My current sense is that it's still on net better to have a system like this than not. But I'll try to spend some time thinking about this more concretely.

Comment by raemon on Raemon's Scratchpad · 2019-11-14T05:12:29.018Z · score: 3 (1 votes) · LW · GW

Some major uncertainties

1. How much work will the community be motivated to do here? 

The best version of this involves quite a bit of effort from top authors and commenters, who are often busy. I think it gracefully scales down if no one has time for anything other than quick nominations or voting. 


2. What actually are good standards for LessWrong?

A lot of topics LessWrong focuses on are sort of pre-paradigmatic. Many posts suggest empirical experiments you might run (and I'm hoping for reviews that explore that question), but in many cases it's unclear what those experiments would even be, let alone the expense of running them.

Many posts are about how to carve up reality, and how to think. How do you judge how well you carve up reality or think? Well, ideally by seeing whether thinking that way turns out to be useful over the longterm. But, that's a very messy, confounded process that's hard to get good data on.

I think this will become more clear over longer timescales. One thing I hope to come out of this project is a bunch of people putting serious thought into the question, and hopefully getting a bit more consensus on it than we currently have.

I'm kind of interested in an outcome here where there's a bar you 


3. How to actually decide what goes in the book

I have a lot of uncertainty about how many nominations, reviews and votes we'd get.

I also have a lot of uncertainty about how much disagreement there'll be about which posts.

So, I'm pretty hesitant about committing in advance to a particular method of aggregation, or how many vetoes are necessary to prevent a post from making it into the book. I'd currently lean towards "the whole thing just involves a lot of moderation discretion, but the information is all public and if there's a disconnect between "the people's choice awards" and the "moderators choice awards", we can have a conversation about that.

Comment by raemon on Raemon's Scratchpad · 2019-11-14T04:52:07.001Z · score: 7 (3 votes) · LW · GW

The 2018 Long Review (Notes and Current Plans)

I've spent much of the past couple years pushing features that help with the early stages of the intellectual-pipeline – things like shortform, and giving authors moderation tools that let them have the sort of conversation they want (which often is higher-context, and assuming a particular paradigm that the author is operating in)

Early stage ideas benefit from a brainstorming, playful, low-filter environment. I think an appropriate metaphor for those parts of LessWrong are "a couple people in a research department chatting about their ideas."

But longterm incentives and filters matter a lot as well. I've focused on the early stages because that's where the bottleneck seemed to be, but LessWrong is now at a place where I think we should start prioritizing the later stages of the pipeline – something more analogous to publishing papers, and eventually distilling them into textbooks. 

So, here's the current draft of a plan that I've been discussing with other LW Team members:

— The Long Review Format —

Many LessWrong posts are more conceptual than empirical, and it's hard to tell immediately how useful they are. I think they benefit a lot from hindsight. So, once each year, we could reflect as a group about the best posts of the previous year*, and which them seem to have withstood the tests of time as something useful, true, and (possibly), something that should enter in the LessWrong longterm canon that people are expected to be familiar with.

Here's my current best guess for the format:

[note: I currently expect the entire process to be fully public, because it's not really possible for it to be completely private, and "half public" seems like the worst situation to me]

  • (1 week) Nomination
    • Users with 1000+ karma can nominate posts from 2018-or-earlier, describing how they found the post useful over the longterm.
  • (4 weeks) Review Phase
    • Authors of nominated posts can opt-out of the rest of the review process if they want.
      • Posts with 3* nominations are announced as contenders. For a month, people are encouraged to look at them thoughtfully, writing comments (or posts) that discuss:
        • How has this post been useful?
        • How does it connect to the broader intellectual landscape.
        • Is this post epistemically sound?
        • How could it be improved?
        • What further work would you like to see people do with the content of this post?
      • Authors are encouraged to engage with critique. Ideally, updating the post in response to feedback, and/or discussing what sort of further work they'd be interesting seeing by others.
  • (1 Week) Voting
    • Users with 1000+ karma rank each post on...
      • 1-10 scale for "how important is the content"
      • 1-10 scale for "how epistemically virtuous is this post"
      • Yes/No/Veto on "should this post be added to LessWrong canon?"
    • (In the 1-10 scale, 6+ means "I'd be happy to see this included in the 'best of 2018'" roundup, and 10 means 'this is the best I can imagine'")
    • "Yes, add this to canon" means that it hits some minimum threshold of epistemic virtue, as well as "this is something I think all LW readers should be at least passingly familiar with, or if they're not, the burden is on them to read up on it if it comes up in conversation."
  • Rewards
    • The votes will all be publicly available. A few different aggregate statistics will be available, including the raw average, and probably some attempt at a "karma-weighted average."
    • The LW moderation team will put together a physical book, and online sequence, of the best posts, as well as the most valuable reviews of each post.
    • The LW team awards up to* $1000 in prizes to the best reviewers, and $3000 in prizes to the top post authors.
      • * this depends on whether we get reviews that seem to genuinely improve the epistemic landscape. Prizes for reviewers will be mostly moderator discretion (plus some inputs like "how much karma and engagement the review got")

And importantly:

Next Year

Even if we stuck to the above plan, I'd see it as more of an experiment than the definitive, longterm review mechanism. I expect we'd iterate a lot the following year.

But one thing I'm particularly interested in is how this builds over the longterm: next year (November 2020), while people would mostly be nominating posts from 2019, there should also be a process for submitting posts for "re-review", if there's been something like a replication crisis, or if a research direction that seemed promising now seems less-so, that's something we can revisit.

Comment by raemon on Robin Hanson on the futurist focus on AI · 2019-11-14T02:42:59.660Z · score: 6 (3 votes) · LW · GW

Would you be up for copying over the summary portion of the transcript here?

Comment by raemon on Meetup Notes: Ole Peters on ergodicity · 2019-11-12T01:26:22.350Z · score: 3 (1 votes) · LW · GW

1. Something like MVP. Don't start by throwing a brand new game out there - even if you have the edge in the game, you have to get people to play it. Getting the stuff for a new game + advertising costs money. Test it out a little (small scale). If you lose money testing it*, you paid a little bit of money to find out you'd have lost a lot of money if you'd tried it out big. (More naturally - big companies are at times known for staying the same, with startups coming in with new ideas. If you copy ideas from other people that haven't bankrupted them...)

This seems to assume the people who did the origination here were casinos or explicit entrepreneurs, instead of people who started gambling informally and then started with some sense of which games had which payoffs.

(Or rather, maybe you're explicitly not assuming that and that's your point. But the way I'd make the same point you seem to be making here is not "they operated like a startup" and more like "they operated like a group of friends/rivals/communities incrementally experimenting, and by the time someone considered starting an explicit business, good gamblers had some intuitive sense of how games worked.")

Comment by raemon on Against "System 1" and "System 2" (subagent sequence) · 2019-11-12T01:13:50.562Z · score: 11 (7 votes) · LW · GW


System 1 and 2 have gotten a lot of attention on LessWrong over the years, informing many explicit models as well as vague folk-theories circulating the community. Given that the originator of the concept has deprecated it, it seems particularly important for the LW community to have common knowledge of the state of the literature.

Comment by raemon on Raemon's Scratchpad · 2019-11-11T23:35:12.399Z · score: 3 (1 votes) · LW · GW


I currently believe it was a mistake to add the "unread green left-border" to posts and comments in the Recent Discussion section – it mostly makes me click a bunch of things to remove the green that I didn't really want to mark as read. Curious if anyone has opinions about that.

Comment by raemon on Predicted AI alignment event/meeting calendar · 2019-11-11T20:43:34.548Z · score: 3 (1 votes) · LW · GW

I was thinking just "reverse the order of the years/months". (Which might not be "newest added" but would be in the limit, would mean you don't have to scroll past years of irrelevant stuff before getting the stuff in the nearterm)

Comment by raemon on Predicted AI alignment event/meeting calendar · 2019-11-10T22:57:50.239Z · score: 3 (1 votes) · LW · GW

Seems weakly better for this to be organized with newer content at the top?

Comment by raemon on Ethical experimentation · 2019-11-10T19:33:27.488Z · score: 3 (1 votes) · LW · GW

Something in this space seems interesting, but a major worry I have is that much of the point of ethics is to have safeguards about doing things that are easy/convenient/harmful (where the harm might be either your future self, or other people).

If you have an injunction against pirating music or eating meat or white-lying, and then for a week you switch to treating those as permissible... one major obvious effect is just going to be "yup, it sure is convenient to do this thing I had an injunction against", which one probably already knew. 

The particular examples you give seem to mostly be doing more interesting things. My impression is your point here is less about the changing the actions you do, and more about changing the "why" behind those actions. Something in this space seems interesting and perhaps important. I expect this experiment to be more-upside when you're experimenting with ethics other than removing injunctions.

Comment by raemon on Toon Alfrink's sketchpad · 2019-11-09T19:28:51.914Z · score: 3 (1 votes) · LW · GW

Presumably not the main point, what ends up happening to your luminators?

Comment by raemon on Judgment, Punishment, and the Information-Suppression Field · 2019-11-09T19:22:03.244Z · score: 3 (1 votes) · LW · GW

A motivator here is that the judgers seem to feel unhappy and frustrated. Something something melting gold.

Comment by raemon on Self policing for self doubt · 2019-11-07T23:38:27.252Z · score: 7 (3 votes) · LW · GW
Sometimes it seems consequentially correct t

Noting that I don't think 'consequentialism' says anything about altruism. You can be a selfish consequentialist AFAICT

Comment by raemon on I would like to try double crux. · 2019-11-07T06:47:47.887Z · score: 4 (2 votes) · LW · GW
For Double Crux to work, we're not supposed to aim for direct persuasion until after we've identified the double crux, or we'll get "lost in the weeds" discussing the parts that aren't important to us.

I'm not sure this is part of the authoritative definition of doublecrux, but FYI the way I personally think of it is "Debate is when you try to persuade the other person [or third parties] that you're right and they're wrong. Doublecrux is when you try to persuade _yourself_ that they're right and you're wrong, and your collective role as a team is to help each other with that." (I don't think this is quite right, obviously the goal is for both of you to move towards the truth together, whatever that may be, but I think the distinction I just made can sometimes be helpful for shaking yourself out of debate mode)

Comment by raemon on Judgment, Punishment, and the Information-Suppression Field · 2019-11-07T03:33:02.958Z · score: 3 (1 votes) · LW · GW
Seems to me like punishment might not accomplish what it claims to & might be harmful on net

Nod. That's been my past default assumption, just noting that my overall opinion here is in flux. Looking forward to further thoughts.

Comment by raemon on Raemon's Scratchpad · 2019-11-07T02:40:22.387Z · score: 6 (3 votes) · LW · GW

Ben Kuhn's Why and How to Start a For Profit Company Serving Emerging Markets is, in addition to being generally interesting, sort of cute for being two of the canonical Michael Vassar Questions rolled into one, while being nicely operationalized and clear.

("Move somewhere far away and stay their long enough to learn that social reality is arbitrary", and "start a small business and/or startup to a bunch about how pieces of the world fit together" being the two that come easiest to mind)

Comment by raemon on Judgment, Punishment, and the Information-Suppression Field · 2019-11-07T01:14:55.483Z · score: 5 (2 votes) · LW · GW

(My current background strategy is something like "positive reinforcement is a better strategy than punishment", but I'm not that confident. I notice that your Oops Prize didn't amount to much. There's a chance that it could just use More Dakka. Upping the prize and doing more advertisement might be worthwhile, and think it is likely that I'd want to contribute to that.

It runs the risk of Goodharting on Looking Like You're Not Goodharting, but maybe that's a problem to worry about when you have more than one submission)

Comment by raemon on Judgment, Punishment, and the Information-Suppression Field · 2019-11-07T01:11:24.615Z · score: 5 (2 votes) · LW · GW

Over the past year or two, I've come to believe that I'm doing a fair bit of "freeloading off others who are willing to punish". But, because the dynamics you describe here, I'm still unsure what the right course of action is.

My sense is that most people doing the punishment are

a) off enough in their targets (relative to my preferences), that I don't feel good endorsing them straight away, often actively disendorsing them, and

b) similar to what you note in this post, I think most of the effect of judgment is general reluctance to speak publicly at all, for fear of being randomly judged in an unpredictable fashion (while slightly tweaking people's predictive weights about what they're likely to be punished for)

So I see a lot of the punishment as anti-social rather than pro-social, and much of my default impulse is to punish the punishers.

But, by now I've seen enough examples of me benefiting from something like "good cop / bad cop", or from pendulum swings where one person is trying to drag the norms way off in a direction that seems too far to me, but that (as you say) "maybe slightly improves the average punishment function".

Meanwhile, it is suspicious if I think the status quo is "okay" enough not to spend any social capital pushing it in some direction.

Nonetheless...'s pretty unclear to me which directions are actually helpful, and whether strategies of judgment or punishment are the way to go.

Comment by raemon on Open & Welcome Thread - November 2019 · 2019-11-06T23:01:01.731Z · score: 3 (1 votes) · LW · GW

Seems like if the guidelines are an empty string it should probably display the default-guidelines.

Comment by raemon on Picture Frames, Window Frames and Frameworks · 2019-11-06T21:00:58.227Z · score: 3 (1 votes) · LW · GW

Minor update: initially for the "framework" images I had looked for a bicycle frame and bridge frame, but the pictures I found... mostly just looked like bicycles and bridges rather than communicating "frame." A friend on facebook recommended searching "carframe" and "airframe" which communicated the concept much better.

Comment by raemon on Elon Musk is wrong: Robotaxis are stupid. We need standardized rented autonomous tugs to move customized owned unpowered wagons. · 2019-11-05T23:10:25.768Z · score: 10 (6 votes) · LW · GW

Similar sentiment re: "I don't think this particular prediction will bear out, but I think it's a good use of LessWrong to make concrete predictions and explorations of possible futures".

Comment by raemon on Chris Olah’s views on AGI safety · 2019-11-04T23:49:01.803Z · score: 7 (3 votes) · LW · GW

I'm not sure I understand the difference between this worldview and my own. (The phrase-in-italics in your comment seemed fairly integral to how I was thinking about alignment/capabilities in the first place).

This recent comment of yours seems more relevant as far as worldview differences go, i.e. 'if you expect discontinuous takeoff, then transparency is unlikely to do what you want'. (some slightly more vague "what counts as a clever argument" disagreement might be relevant too, although I'm not sure I can state my worry crisply, nor really confident my worry is cogent)

I don't have a strong position on the continuous/hard-takeoff debate and have updated a bit over the past year both on continuous-takeoff's plausibility as well as the value in shifting the AI field towards having clearer models of what they're building, generally. But insofar as I'm suspicious of this, it's mostly because I still put moderate probability on "some understanding here may be more dangerous than no understanding, precisely because it's enough to accomplish some things without accomplishing everything that you needed to."

Comment by raemon on Picture Frames, Window Frames and Frameworks · 2019-11-03T23:21:12.782Z · score: 4 (2 votes) · LW · GW

An open "jargon question" that came out of this is something like: what do you call the category-type of "the sort of thing that gears/feelings/power", vs the category-type of "the sort of thing that window/picture/framework are".

Comment by raemon on AlphaStar: Impressive for RL progress, not for AGI progress · 2019-11-02T19:07:48.530Z · score: 4 (2 votes) · LW · GW

Just a layman here, and not sure if this is what this particular disagreement is about, but one impression I've gotten from AlphaGoZero and GPT2 is that while there are definitely more architectural advances to made, they may be more of the sort "make better use of computation, generally" than anything feels particularly specific to the strategy/decision-making problems in particular. (And I get the impression that at least some people saying that there are further breakthroughs needed are thinking of something 'more specific to general intelligence')

Comment by raemon on Halloween · 2019-11-02T00:21:29.819Z · score: 3 (1 votes) · LW · GW

[note: I think I misread your original comment, but thought my comment was still worth posting. I originally thought you said "it seems like your solution also has nothing to do with Halloween'". The short answer to the one you actually said was "but I miss Halloween, tho." This comment is for people who actively want Halloween in their lives, not who are clinging to it out of misguided sense of obligation, which I agree some people do and which is silly]

Those are also valid suggestions, and I definitely sympathize with people who don't consider my original suggestion valid (whether because their concerns were more shaped like my-current-understanding-of-Katja's than mine, or some other reason)

I don't really think there's such a thing as the definitive True Spirit of Halloween, since it's been evolving in random directions for hundreds of years. You can celebrate some kind of classically pagan Samhain thing. You can celebrate something that earnestly deals with whatever modern conception of death or horror you care about. You can celebrate something in honor of whatever it is that you grew up with, arbitrary and/or commercial though it may be.

From my perspective, Halloween has a few different things going on. Some have to do with aesthetics, some have to do with "the functional/logistical form", and the two are blended together a bit.

Aesthetically, there's horror, death, autumn, and some manner of masquerade-ness.

Functionally/logistically, if you're doing 20th-century-American-style-suburban-halloween, there's interacting with your neighborhood in some way – either going door to door in costume getting treats, or having people come to your door and give treats out. In my conception of Halloween this part is at least 30% of the point.

I mentioned elsethread the two-sided-marketplace problem here. Some other things in that frame is that currently, community/neighborhood interaction seems on the decline (at least in neighborhoods I've been in), and this seems quite bad – we (where "we" means people with roughly my values) are in a rearguard fighting-retreat against atomic individualism corroding a lot of cultural cornerstones. A lot of those cultural cornerstones were actually bad and it's fine to let them go, but the process doesn't seem really aligned with any particular values.

Playing a Ravenloft one-shot is certainly another fine thing to do for Halloween, but if you were missing sense-of-neighborhood as part of the loss of Halloween, it won't help.

But if one cares about maintaining any sense of neighborhood, then it's especially valuable to look for activities one can do that take advantage of what remaining social institutions there are and reinforce/build off them.

Halloween caroling feels to me like a relatively natural evolution of the existing holiday elements (given that holidays evolve sort of randomly in the first place), and a healthier evolution than most of the ones that capitalism is prone to offer. It's easier to jump-start with random strangers because almost all of the elements are already familiar, and the one element that is new (for Halloween) is a borrowed element from Christmas.

What was really striking, going Halloween Harrowling yesterday, was how much the neighbors seemed deeply appreciated and excited* by it – it was one of the purest senses I've gotten of "going through town, spreading holiday cheer"

*unlike Christmas caroling, alas. The important, valuable distinction between Halloween and Christmas is that for Halloween, having lots of decorations in your yard is a strong signal that you actually want strangers coming to your house, which is not at all true for Christmas. Times I've went Christmas Caroling have felt fairly hit-or-miss for people actually appreciating it.

Comment by raemon on Halloween · 2019-11-01T23:07:55.512Z · score: 6 (3 votes) · LW · GW

If anyone actually wants to contend with death this holiday season, I know a ritual-guy (I think in Austin Texas?) who is putting together an experience where he ferries you across a river to an island in a small boat, and then you crawl into a coffin, close the lid, and you hear some shoveling and a small bit of a dirt. You stay there for about 15 minutes reflecting on your life, and then re-emerge.

Comment by raemon on Chris Olah’s views on AGI safety · 2019-11-01T22:34:33.385Z · score: 26 (10 votes) · LW · GW

Sort of a side point, but something that's been helpful to me in this post and others in the past year is reconceptualizing the Fast/Slow takeoff into "Continuous" vs "Hard" takeoff, which suggest different strategic considerations. This particular post helped flesh out some of my models of what considerations are at play.

Is it a correct summary of the final point: "either this doesn't really impact the field, so it doesn't increase capabilities; or, it successfully moves the ML field from 'everything is opaque and terrifying' to 'people are at least trying to build models of what their software is doing, which is net positive for getting good practices for alignment into the mainstream?"

That's an interesting and clever point (although it triggers some sort of "clever argument" safeguard that makes me cautious of it). The main counterpoint that comes to mind is a possible world where "opaque AIs" just can't ever achieve general intelligence, but moderately well-thought-out AI designs can bridge the gap to "general intelligence/agency" without being reliable enough to be aligned.

Comment by raemon on Halloween · 2019-11-01T20:10:52.446Z · score: 14 (3 votes) · LW · GW

This doesn't really address the concerns in the OP but I wanted to note: I had struggled a lot to figure out what to do with Halloween as a childless adult.

Halloween costume parties are fun if you like making costumes, but... I dunno, didn't feel much connection to the holiday. The most exciting things about adult-halloween-parties are a certain kind of free debauchery that I'm not really into.

So I was delighted when a friend of mine invented Halloween Caroling, where you dress up and go trick-or-treating, except that you go to houses with lots of decorations (which are sort of self-selected for being open-to-experience), and ask they'd like a Halloween Carol, and then sing, like, Thriller or Addams Family or the Nightmare Before Christmas songs, and/or some spookier folk songs. It adds a sense of purpose to the holiday that... well, sure still doesn't really have much to do with Facing Death Squarely in the Face but is fun.

Comment by raemon on The Technique Taboo · 2019-11-01T19:40:27.946Z · score: 3 (1 votes) · LW · GW

(ended up replying to Isur's other comment)

Comment by raemon on The Technique Taboo · 2019-11-01T19:40:03.593Z · score: 6 (3 votes) · LW · GW

This included local community college, a tech school focused on computer animation (which included traditional art classes), and visiting Rhode Island School of Design (one of the leading art schools, where I didn't actually get to take classes but got a good sense of their curriculum and it seemed pretty traditional to me).

I'm not sure what your friends' experience was and wouldn't be that surprised if different places were different, but this doesn't match anything I've heard of in terms of art classes.

(My last classes/paying attention to this was approximately one decade ago, today)

Comment by raemon on Halloween · 2019-10-31T19:46:36.620Z · score: 10 (5 votes) · LW · GW

Huh, this comment leads me to look at Halloween through the inadequate equilibria lens.

Holidays vary in how much they're a community affair, vs family affair. Many (most?) of them at least aspire somewhat to be a community affair, but in a way that gracefully scales down when you only are doing it with your immediate family or friends. There are Thanksgiving Parades and Midnight Christmas vigils, but the holidays still basically work if you decide to do the unilaterally.

Not so with Halloween – it specifically depends on a particular kind of neighborhood, with a two sided marketplace between children and host-families.

Comment by raemon on On Internal Family Systems and multi-agent minds: a reply to PJ Eby · 2019-10-30T21:40:42.317Z · score: 8 (3 votes) · LW · GW

Quick note: Kaj had expressed some uncertainty about whether it made sense as a frontpage post, and a few members of the LW Team had specifically encouraged him to post it as such. So, first, just noting that it probably makes more sense to direct confusions or frustration at us.

I can't speak for the other team members but some of my own thoughts included:

  • the post/comment was long enough that it felt a bit more natural as a post
  • I currently think one of the main bottlenecks on LessWrong is distillation – much of the time, there's a lot of good back and forth in the comments, but it takes a lot of effort for future-people to learn the most important takeaways. The longer the conversation goes, the harder it can be to compress into something digestible. So I think there's something useful to having distillation posts such as this one by Wei_Dai (mostly re: Paul Christiano), and this one by me (re a conversation with Benquo, Jessicata and Zack Davis).

But, I can definitely see how it would come across as weird and escalatory if you weren't expecting it. (I think it is probably a better cultural norm to touch base with conversation partners and give them a chance to give feedback, esp. when you're summarizing their position).

Comment by raemon on [Site Update] Subscriptions, Bookmarks, & Pingbacks · 2019-10-30T20:33:09.970Z · score: 3 (1 votes) · LW · GW

checking if this is still the case?

Comment by raemon on The Technique Taboo · 2019-10-30T20:07:31.237Z · score: 13 (7 votes) · LW · GW
For a strange few decades that may just be starting to end, if you went to art school you'd be ostracised by your teachers for trying to draw good representational art. "Representational art" means pictures that look like real things. Art school actively discouraged students from getting better at drawing.

I roll to disbelieve on this. All the art classes I've taken in the past decade focused largely on representational art. (esp. figure drawing). The taboo was drawing cartoons, anime in particular

Comment by raemon on [Site Update] Subscriptions, Bookmarks, & Pingbacks · 2019-10-30T05:24:23.472Z · score: 8 (4 votes) · LW · GW


Comment by raemon on bgaesop's Shortform · 2019-10-29T19:53:48.651Z · score: 12 (-1 votes) · LW · GW

My sense of what the person-at-the-reunion was talking about (having chatted with them a bit, although not sure I understand their position well enough to speak for them) was a model where Chakras roughly correspond to "application of Gendlin's Focusing, directed to particular areas of the body, turns out to yield different information."

i.e. a thing that I've heard reported by several LessWrongers is that focusing directed at your stomach tends to give a set of information about what your subconscious is thinking/feeling/interested-in, than focusing other areas of your body, or without any directed attention at all.

I've heard a couple people make the broader, somewhat stronger claim that each of the body-areas associated with a chakra tend to have consistent effects across people when used as introspection targets.

This doesn't seem particularly mysterious to me, although it seems reasonable to be escalatingly skeptical of:

  • "introspecting with a focus on particular body parts yields different information about what's going on with you subconsciously than generic introspection"
  • "one particular body part tends to be particular useful for this"
  • "seven body parts tend to be particularly useful for this in a way that corresponds to traditional chakras, and there's a model of how those areas relate with particular introspective techniques that tend to cause particular effects, across people"
Comment by raemon on What are some unpopular (non-normative) opinions that you hold? · 2019-10-25T23:20:41.821Z · score: 5 (2 votes) · LW · GW

Came here to say a similar thing. I'm also curious about what Said was actually referring to with the 'most of history' clause, because there's plenty of things (slavery, some forms of serfdom, etc), where "threat of violence" was explicitly part of what was going on. But I could also him meaning to refer to... well, probably situations where threat-of-violence is still involved at some point, but is a couple steps removed (where maybe the threat is more like explusion from the tribe, which then increases your chance of death or harm)

Comment by raemon on Strategies of Personal Growth · 2019-10-25T21:05:09.755Z · score: 3 (1 votes) · LW · GW

Something I feel unsatisfied with, a year after posting this (and yet another year after Sunset at Noon) is that I think the most important habit to cultivate is "Try Things", because it unlocks all the higher return options. But I don't really have a good framework for doing that systematically or reliably.

I could do things like have a "go to at least one new meetup a week or month", as part of a general increase luck-surface-area thing. That feels like only one dimension of the thing though.

Comment by raemon on Raemon's Scratchpad · 2019-10-24T19:48:39.348Z · score: 4 (2 votes) · LW · GW

I mean, the whole thing only triggers once per day, so I can't go farther than a single loop of "why didn't I reflect on my habit-failure yesterday?" :P

(But yeah I think I can handle up-to-one-working-memory-load of habits at a time)

Comment by raemon on Raemon's Scratchpad · 2019-10-24T02:48:34.983Z · score: 18 (9 votes) · LW · GW

I'm not sure why it took me so long to realize that I should add a "consciously reflect on why I didn't succeed at all my habits yesterday, and make sure I don't fail tomorrow" to my list of daily habits, but geez it seems obvious in retrospect.

Comment by raemon on Where to absolutely start? · 2019-10-22T22:58:52.628Z · score: 3 (1 votes) · LW · GW

There's room for epistemology practice to be hands-on, relevant and exercise-based without being about "productivity."

Comment by raemon on 2 small problems with the site... · 2019-10-22T19:48:33.559Z · score: 5 (2 votes) · LW · GW

Can you link to a particular comment that does the awkward truncation thing? Our truncation process is supposed to address that using a method similar to what you describe. But if you give me a working example I can look into it.

re: "Magical Algorithm" vs "Oldest": what you're problem running into here is a hard-coded edgecase. Very Old Posts were published before LessWrong had a threading system (or indeed, before LessWrong even existed – originally published on In those cases, some old comments actually only make sense if read in chronological order, and we override the usual settings.

Comment by raemon on Raemon's Scratchpad · 2019-10-21T20:06:37.782Z · score: 5 (2 votes) · LW · GW

I'm interested in hearing more details about that.

Comment by raemon on AI Alignment Open Thread October 2019 · 2019-10-21T13:09:59.077Z · score: 3 (1 votes) · LW · GW

There actually is an important difference, which is that you get to set the moderation norms on your shortform posts (and can ban people from commenting if need be). But, indeed, I would agree most of the difference is informal-cultural.

Comment by raemon on Raemon's Scratchpad · 2019-10-21T12:51:17.482Z · score: 3 (1 votes) · LW · GW

I don't currently have a problem with roommates (we solved it last time with some ad-hoc negotiation) I'm just more generally annoyed that there's not a good principled approach here that I can pitch as "fair".

(We do have apartment cleaners who come biweekly, whose cost is split evenly, but that also just doesn't address all the various small ways mess can add up on the timescale of hours or days. In the original motivating case it was about hairs getting in the sink-drain, which I prefer to solve once a year with a bottle of Draino, and others preferred to solve much-more-frequently with smaller-dollops-of-draino. i.e. I consider it fine if a sink drains slightly slowly, others found it gross)

((Also, there's a much more general version of this which is what I was more interested in, which isn't just the case of roommates in particular - it includes small ad-hoc situations such as some friends going camping and having different preferences about how much to cleanup))

Comment by raemon on AI Alignment Open Thread October 2019 · 2019-10-21T12:45:15.056Z · score: 3 (1 votes) · LW · GW

I'm not sure what'll end up settling with for "regular Open Threads" vs shortform. Open Threads predate shortform, but didn't create the particular feeling of a person-space like shortform does, so it seemed useful to add shortform. I'm not sure if Open Threads still provide a particular service that shortform doesn't provide.

In _this_ case, however, I think the Alignment Open Thread servers a bit of a different purpose – it's a place to spark low-key conversation between AF members. (Non-AF members can comment on LessWrong, but I think it's valuable that AF members who prefer AF to LessWrong can show up on and see some conversation that's easy to jump into)

For your own comments: the subtle difference is that open threads are more like a market square where you can show up and start talking to strangers, and shortform is more like a conversation in your living room. If you have a preference for one of those subtle distinctions, do that I guess, and if not... dunno, flip a coin I guess. :P

Comment by raemon on Healthy Competition · 2019-10-21T06:30:00.843Z · score: 3 (1 votes) · LW · GW

Agreed. The comment was meant to be referring to orgs that made particular sense as a monopoly)

Comment by raemon on We tend to forget complicated things · 2019-10-21T05:28:00.571Z · score: 6 (3 votes) · LW · GW

Just wanted to say I appreciate the brevity. :) I'm guilty of overly long LessWrong posts.

Comment by raemon on Where to absolutely start? · 2019-10-21T05:20:19.768Z · score: 18 (9 votes) · LW · GW

Even though R:AZ isn't very hands on, it's still pretty core and definitely worth your time. I think all three all roughly equally newbie friendly (Much of the Codex was written with the sequences as background assumptions, but still probably accessible without it)

HPMOR is probably the most "fun", if you happen to like the genre.