Set a Yoda Timer and share the most important idea you haven’t had time to express. Five minutes is all you get.
Phew, okay, you asked for it. Since I only get five minutes I'm not going to filter for niceness.
Most people, including and especially most rationalists, are disconnected from an incredibly important part of what it means to be human, which for lack of a better word I'll call the soul. The soul is the part of you that you're talking to when you're doing Focusing in a particular way. It's where great art comes from. It's where the power that comes from having something to protect comes from. The nondegenerate form of circling is learning how to communicate soul-to-soul. It may or may not have something to do with the enteric nervous system.
If I understand correctly, this is vastly different from the elephant?
Jordan Peterson speaks at length about the soul. He insists for example that if you're paying attention, you can feel in your chest whether your soul endorses what you say (it feels stronger and tighter), and you can also feel when you're being dishonest (it weakens and dissolves). I'm sure the actual sensations vary from person to person but this sounds like the same thing.
It's more specific. And yeah, that sounds right; sounds like a blend of Focusing and belief reporting.
Cool, I'm treating this soul as an oracle to consult using roughly that method. I'm a little worried that your picture of the soul corresponds to (at least) two distinct things in my models, how important in particular is the "create great art" part to the rest?
I think I would classify the set of things you described as the "truth/conscience" oracle and separately the "subconscious/right-brain/unvocalized" artist. The second thing is the main producer of dreams for example.
I'm happy to make finer distinctions but at some level of fuzziness I identify the two things. The way you write great poetry using your truth / conscience oracle is you write a line of poetry and ask if your oracle endorses it; if not, you keep fixing it until it's endorsed.
Yoda Timed (plus an extra 30 seconds at the end to finish up, and then another minute to edit):
There is a type of diversity that is super important for communities. It isn't race or gender, and it isn't even belief/ tribe. It has more to do with having lots of Different Types of People.
If your car breaks down, is there a mechanic in your community you can trust to not screw you over? If everyone is making high salaries, who in your community would be available to dog-sit in your home for twenty bucks a night?
This is one reason why I don't think the Archipelago solution will work. It is even further sub-dividing an already homogenous group. Let's say some people are more of the social sort, and some people are more of the intellectual conversation sort. The social sorts will hold parties that won't have the interesting conversations they desire. The intellectual sorts will have to be intellectual by themselves since they aren't the "arrange a gathering" type.
This also leads to the (erroneous) assumption that because the rationality community consists primarily of Type X of person, that that specific type of person must be inherently better at rationality as a whole.
This really resonated with me. My instinctive reaction was that what you mean by Different Types of People is first D&D Alignment (Lawful-Chaotic, Good-Evil), and second personality type (Big Five/Myers Briggs). Is this closer to what you mean, or are you more focusing on skill specialization? What do you think Type X of rationality people consists of, and can anything be done to increase this kind of diversity?
Separate comment disagreeing with your last sentence on "rationality as a whole":
I have the notion that rationality doesn't belong to the rationality community. Rationality is systematized winning. It isn't any single set of teachings or professions of belief. The rationality community is just one of many groups of people who explicitly aspire to win systematically - perhaps even the one that has gotten the farthest towards the Way, but nevertheless a community of aspiring rationalists. Real rationalists are people who win systematically. If you want to learn real rationality, look to the Elon Musks of the world, even if they don't wear LessWrong badges.
What I originally meant does align more with skillsets. A healthy community has people who fill all the roles a community needs. Hosting, cooking, debating, organizing, leading, etc. An ideal community ALSO has people who fill roles the community MEMBERS need (helping fix your roof, providing childcare, medical consultation for simple things, etc)
We have people who like to blog and program. Not super helpful if you want someone to help plan/ set up your garden.
Things COULD be done to increase diversity along these lines I suppose, but they probably shouldn't. It trades off against other values that are held in too high regard here. My recommendation is to get your "healthy community" needs met elsewhere.
I agree that rationality doesn't belong to the rationality community. The five-minute yoda timer meant I wasn't super-clear that I think the last sentence is an erroneous belief.
I think people often worship their problems, and this directly ties into this post. Eliezer covered this with Musashi’s proverb of how one should strike the enemy rather than their blade:
The primary thing when you take a sword in your hands is your intention to cut the enemy, whatever the means. Whenever you parry, hit, spring, strike or touch the enemy's cutting sword, you must cut the enemy in the same movement. It is essential to attain this. If you think only of hitting, springing, striking or touching the enemy, you will not be able actually to cut him.
I noticed this today at martial arts when sparring someone with four times my experience; I was so intimidated that all my strikes were aimed at their guard! There was no way I was going to score a point unless they moved into my kicks. This was ridiculous. I fixed it and actually landed a few blows (to be sure, I would have gotten rolled if they actually tried).
It’s been said before, it’s been said by this very post, and yet there is still value in repeating the idea once again: do not worship your problems. You may have read this post and thought, "gee, I’m not like alkjash, there’s no way I’d win that competition". Maybe so, but I am confident that there are problems you could easily decimate if you let yourself win.
Strike true, or don’t strike at all.
Most important idea huh?
Here goes nothin'..
I think that while the community in general has a large amount of declarative information. A very small amount of it is actually put into procedural format. We have repertoires of information and expertise, we can do some really cool things with our minds when we properly use them. However, we are missing a curriculum. We are missing a centralized place where a repository of literally all the most useful information lies. We have no textbook. We have pieces like Hammertime which try to amend some of this with daily exercises and motivational blurbs to reinforce actual practice. We have the sequences, which do a good job of laying down epistemic hygiene and probabilistic thinking. But even the sequences still failed in certain ways. There are few practice problems, there are few guided opportunities to attempt what you just read. We have certain groups like CFAR's rationality checklist (imo only really covers epistemic hygiene) and the applied cannon. But these are not sufficient.
Jeez that was stressful. I went about 32 seconds over the time limit, but cutting off mid-sentence sounded like a bad time
Also, links were only added after I finished.
Also also, this is not a literal endorsement of my beliefs about the state of rationality. That was an attempt for me to push as much of this idea out of my head as possible in five minutes, so there's probably [75%] going to be some bad wording or misrepresentation or straight up falsity. But that was my best shot.
I think despite your disclaimers it was a great idea well worth expressing (although it's essentially applause lights in the comments section of Hammertime). Hopefully this updates you towards pushing out your ideas first and asking questions later. =)
The episodic (Diachronic Done Right) link points to a 404. I tried searching via google and couldn't find anything -- any chance it was moved or otherwise still exists? It sounds pretty interesting and I'd like to read it if possible.
I searched using the site searchbar, searched through Duckduckgo (with 'site:lesswrong.org'), and also looked in the Internet Archive, but could not find it. I am also interested in the article if you find it, but I am now curious as well how you found it.
One of the mantras that’s been very useful to me recently is ‘Progress isn’t Linear’.I often find that when I’m trying to fix bugs, I will give up if I miss a few days or run into a problem I don’t know how to solve. ‘Progress isn’t Linear’ reminds me that it’s okay to have problems and I can try again.
I was lazily walking home from the dentist's today, reading this post and mulling over a technical problem. I had had a few good ideas, but nothing I felt was sufficient. I made myself an ultimatum: "find a tractable solution in 5 minutes or someone you care about dies". Lo and behold, I had a good answer within 2 minutes, after having thought about it idly for two hours.
Probably too extreme for many people, but it worked for me; I was able to put myself in that situation and confident that I wouldn't suffer mental harm from doing so. Please, don't do this if you don't think the results would be good for you and your situation.
Set a Yoda Timer and share the most important idea you haven’t had time to express. Five minutes is all you get.
I really think that a lot of modern AI alignment research is being done within the academic system, but because it's done within the academic system it's fairly ignored by the independent/dedicated nonprofit research community when compared to independent/dedicated nonprofit research. On the contrary, it likely gets much more attention within academia.
I don't think the dynamic here is "each team likes their own people best." I think it's due to an unwarranted degree of skepticism of the academic system, which may be warranted in non-emergencies but which is less warranted when facing truly apocalyptic threats. The academic system has produced a lot of valuable research on climate change and nuclear risks, and I'd expect its research on AI to be similar broadly speaking.
The fact that the first few successful researchers weren't academics isn't really a point against the academic system here, any more than Priestley and Lavoisier not being academics is a point against academic chemists. The supposed pre-paradigmaticity of the field also isn't really a point against the academic system here, given that many protosciences (e.g. Freudian psychoanalysis) were able to grow into sciences within the academic system, and this pattern can be seen continuing in fields such as astrobiology.
Based on this, I decided to see how fast I could write. I set 5 minute timers, with the initial goal of writing 400 words in each 5 minute block. I got closer to 200 words in 5 minutes, but with four 5-minute timers, I wrote a 900-word post in half an hour
I grew up in Oakland, but didn't return after college. Looking at California as it was then and as it's changed since I've left, I keep feeling like housing policy is really the nexus of so many different problems (and that this is true nationwide as well). We're struggling with homelessness, high costs of living, climate change, wildfires etc., and housing has such a big role to play in all of those. We make new housing construction so hard, not just with zoning but with lot size and coverage restrictions, parking requirements, and arduous consultation and permitting processes. So many of the most desirable neighborhoods in the country, Greenwich Village, Rittenhouse Square etc. would be illegal to build today! Hard to build housing means rising rents, rising rents mean high cost of living and people being pushed into homelessness. People moving to exurbs add carbon emissions, and low density inside cities chokes public transit systems that can't get a critical mass of riders. As we push farther into urban-wilderness boundaries, we put more people at risk in wildfires. I'm not by any means the first person to say this, but I just wish more people could see the whole web of connections.
Phew that was hard, I ended up taking about an extra 30 seconds to get it done, but it's a good exercise in just putting words down and not spending too much time searching for perfection.
Idea I haven’t had time to express:
My brand of social anxiety (so I’m assuming others have it too) might come from the combinatorial explosion of more people in a group - thus more paths of information flows. I find it very easy to talk to someone 1 on 1, and I can usually have a good conversation with anyone. But once the group gets a bit bigger, I think I get a bit scrambled by trying to simulate all of the potential flows of information and feedback loops.
I think this is where politics (at a very basic level) can come from - once you hit 4 people in a group, you have 24* different permutations/factions/sub-groups.
So my thinking here is that I might be using an inefficient “theory of mind simulator” for more complicated scenarios, but I’m also interested in the System Theory of small groups (like bands).
*(I think?)
That was tough! I took 7 minutes. Thank you for prompting me to think about that, I want to learn more and flesh it out. I’m circling around something but haven’t quite captured it yet.
You're actually hitting on an academically acknowledged hard limit to the number of people whose mind states you can keep track of in conversation. Typically, when there are more than 4 people, you'll see smaller groups form of 2, 3, and/or 4. It's very noticeable at big parties. (So no need to be anxious!)
Dunbar talks about it in one of my favorite papers of all time The Anatomy of Friendship. Happy to send you a copy if you're interested :D
Replying to add a link to the full paper: https://www.ericz.one/aof.pdf <- note, he only mentions it in this paper.
There's a more detailed explanation in this one!
P.S. Thanks for your patience! Been a bit since I was on here.
There is a very helpful model of the human:
= a person has needs
there are approximating lists - https://www.cnvc.org/training/resource/needs-inventory
- whenever need is met, we generate pleasant feelings
- whenever need is not met, we generate unpleasant feeling
- paying attention to feelings can help us discover unmet needs, even if we aren't initially conscious of them
= there is an important difference between 'a strategy' and 'a need'
understanding of which can change behavior significantly
"need" is something that can be satisfied in many ways.
model proposes, that if we don't fixate on the strategy and satisfy the need in a different way
it will be just as good.
with clear understanding of these things lots of conflicts can be solved.
conflict is often when people try to satisfy their needs with strategies that don't satisfy needs of another.
focusing on needs and not on strategies can help us find different ways that satisfy all included.
This is applicable to sub-minds or sub-agents within one person.
Some needs are overlooked by society.
E.g my need for being seen, for friendly acknowledgement of my (light) struggles is unsatisfied at this moment.
To the extent that I initially thought to dedicate idea of "me wanting more understanding, and being afraid to ask for it" as the most important.
(NVC in general turned out to be a very important tool for me)
One important idea that I have not yet had time to share widely, explained in five minutes:
What happens if we assume that a comfortable life and reproduction are inviolable priviledges, and imagine a world where these are (by the magic of positing) guaranteed never to be violated for any human? This suggests that the number of humans would increase exponentially, without end, until eventually some point is hit where the energy and resources available in the universe, available at the reach of mankind, is less than the resources needed to provide a comfortable life to every person. Therefore, there can exist no world where both reproduction and a comfortable life are guaranteed for all individuals, unless we happen to live in a world where there is infinite energy (negentropy) and resources.
---
The explanation might not be perfect, and the important implications that I believe follow may not be clear from this, but this is a principle that I often find myself meditating upon.
Were long-standing research problems actually ever presented at IMO? AFAIK, problems featured there already have solutions.
This is part 12 of 30 of Hammertime. Click here for the intro.
Anyone who can muster their willpower for thirty seconds, can make a desperate effort to lift more weight than they usually could. But what if the weight that needs lifting is a truck? Then desperate efforts won’t suffice; you’ll have to do something out of the ordinary to succeed. You may have to do something that you weren’t taught to do in school. Something that others aren’t expecting you to do, and might not understand. You may have to go outside your comfortable routine, take on difficulties you don’t have an existing mental program for handling, and bypass the System.
~ Make an Extraordinary Effort
I don’t know if I’ve ever made an extraordinary effort (that’s probably evidence I haven’t), but I’ve certainly made desperate efforts. The philosophy of Yoda Timers is that it might be enough to make desperate efforts all the time: to do the known thing as well and quickly as can be done. Past that is the realm of rare genius.
CFAR calls Yoda Timers “Resolve Cycles,” a sub-skill of Resolve – the ability to make a desperate effort. Least glamorous of all rationality techniques, Resolve deserves its own book. How much could you accomplish just by more brute force all the time?
Previously: Day 2.
Resolve is the main skill being trained by Yoda Timers, but there are a number of other useful reasons to build timers and deadlines into your life. Today I’ll share three ideas to make the most out of Yoda Timers.
Sometimes, you surprise yourself with what can be done in five minutes. But sometimes, there are things that can’t be done in five minutes. In this case, the generalization of Yoda Timers is to set absurdly short deadlines for these tasks.
How long does it take to write a novel? NanoWriMo is a Yoda Deadline for one month.
How long does it take to solve long-standing research problems? The IMO says: sometimes, only four and a half hours.
How long does it take to turn your life around? How many people waste away for years or decades before accelerating back through life in the span of weeks, tipped by a single conversation or book or trip?
The short answer to all these questions is: you have no idea how fast you can be without practicing for speed.
The Harvard-MIT Math Tournament (HMMT) has an easier version, the Harvard-MIT November Tournament (HMNT), which is run for local and less experienced (middle and early high school) students. HMNT is composed of several individual and team-based rounds, the most exciting of which is the Guts Round. Teams of 4-6 students work together on problems that come in sets of three to solve a total of 36 questions in 80 minutes.
A handful of older students, myself included, helped out at the HMNT of 2011. The coach of the IMO team challenged us to participate in the Guts Round, except instead of working in teams of 6 we would work alone, and without scratch paper.
And so it came to pass that, behind an auditorium full of teenagers loudly whispering ideas and trading scratch paper, the five of us sat silently in a row, staring problems down and writing down answers.
Tallying our scores at the end, each of us individually beat all of the actual teams by a wide margin.
From that day on, I did HMMT practice problems with half the time and only mental math. I won the thing twice.
“My dear, here we must run as fast as we can, just to stay in place. And if you wish to go anywhere you must run twice as fast as that.”
~ Alice in Wonderland
There’s a common failure mode with writing projects: if you work too slowly, ideas become stale before you’re even close to finishing.
How many unfinished thoughts get the backspace because they don’t stand up to reflective endorsement?
A half-finished blog post rusts overnight.
After a week, the first chapter of your novel reads like a child’s writing.
That proof you jotted down months ago? You haven’t a clue how to fill in the details.
I give examples in writing because I’m preoccupied with writing, but staleness and motivation decay apply to all creative pursuits, especially for episodic people. One solution is to try to solve the control problem, build trust with your future self, and otherwise learn to plan for the long term. That we covered on Days 8, 9, and 10. But another solution is simply to do things faster.
Murphyjitsu should have no trouble detecting these failure modes. There are ideas that you know you won’t follow through on if you don’t finish them immediately. If you put something off for months, even when you end up doing them, they’ll take twice as much effort.
Set Yoda Timers and Deadlines. Motivations and values drift – make the most of those you have today.
Usually, five minutes is an absurdly short time to try something. But sometimes five minutes is an eternity. The secondary use of Yoda Timers is to draw your focused attention to tasks that you normally spend seconds on.
How much time do you spend planning your day? Set a Yoda Timer and move things around on your schedule to maximize efficiency.
How much time do you spend expressing gratitude? Set a Yoda Timer and searching for the perfect gift, or writing a thoughtful note, for a loved one.
Are there muscles you never exercise? Set a Yoda Timer and train that one muscle group (see Sore in Six Minutes to learn how). Notice what flexing and relaxing it feels like. Explore the full range of motion. Accept the lovely burn of lactic acid.
Do you dive into things without enough planning? Set a Yoda Timer to slow down and Murphyjitsu.
Today’s exercise: set a Yoda Timer for five minutes and build a plan to incorporate timers and deadlines into your life.
Set a Yoda Timer and share the most important idea you haven’t had time to express. Five minutes is all you get.