Group Rationality Diary, May 1-15

This is the public group instrumental rationality diary for May 1-15. 

It's a place to record and chat about it if you have done, or are actively doing, things like:

  • Established a useful new habit
  • Obtained new evidence that made you change your mind about some belief
  • Decided to behave in a different way in some set of situations
  • Optimized some part of a common routine or cached behavior
  • Consciously changed your emotions or affect with respect to something
  • Consciously pursued new valuable information about something that could make a big difference in your life
  • Learned something new about your beliefs, behavior, or life that surprised you
  • Tried doing any of the above and failed

Or anything else interesting which you want to share, so that other people can think about it, and perhaps be inspired to take action themselves. Try to include enough details so that everyone can use each other's experiences to learn about what tends to work out, and what doesn't tend to work out.

Thanks to cata for starting the Group Rationality Diary posts, and to commenters for participating.

Previous diary: April 16-30

Next diary:  May 16-31

Rationality diaries archive

28 comments, sorted by
magical algorithm
Highlighting new comments since Today at 7:49 AM
Select new highlight date

Good habits:
+Not only maintained but improved upon my exercise routine. Body weight focus twice a week, cardio focus twice a week, mixed exercises at least once a week, yoga and stretching daily, stretching every half hour at work. I was already slim so no real visual rewards yet, but I certainly feel good.
+Returned to reading the sequences after a break in which my lack of formal math training made it seem pointless. Almost through the Quantum Physics and the rewards have been quite good. The concepts have reached a point where they feel intuitive, maybe not on a gut level but at least on a reflex level. I'm glad for this because it also makes them easier to change/reject/assimilate when I learn more and understand better. The concepts of Many-World and Timelessness feel less like a fire and more like wood to burn for a more lovely fire. As do all concepts that I understand on a gut level.
+Begun to meditate.
-But have not optimized my meditation habit. Still need to improve upon it. -Dropped attempting to learn piano or Spanish as well as attempting to utilize more open source software such as Linux. The reasons were many but the results are still a negative.

Bad habits:
-Still involved in a habit that is a time sink but which I have a great emotional attachment too. Am currently determining my best course and cost and benefits of trying to optimize it or cutting it.

New evidence:
This one was hard on me. I've recently had some kerfluffles regarding job opportunities. When an opportunity suddenly opened for a job in my very very isolated, very poor hometown, I did not want to take it. My family pressured me to do so. However, I realize that my decision not to take the job was ignorant. It is an opportunity for money and work that I can terminate when something better comes along. While I know that the reasons my family wants me working at the job (nearness to home, isolation from my peers) are things I want to avoid, this is no reason I cannot utilize the job to maximize my own position.

New beliefs:
I'm currently updating my beliefs about the LW community and the related sister communities. I still have the sensation of an outsider looking in, but I find myself relating to or agreeing with more of the socially strange beliefs of the community. Not because they come from the community but because, as I read about these beliefs, I find they make sense. And as long as they make sense, I have no real choice but to accept that they seem to make sense until I find reasons that show them not to make sense.

I imagined which world I would most need my time sink habit in. After all, my time sink habit isn't something like smoking or video gaming addiction. It isn't directly unhealthy nor does it demolish my social life. It is merely and inefficient use of time and unsatisfying in the long term. However, it does provide some, short-term emotional satisfaction, and I can imagine a world where that small benefit outweighs the price. Some circumstances demand a short-term emotional reward to accomplish long-term goals.
So, I tried to picture the world I would have to live in to justify spending time for the small emotional benefit of the habit. The world that match is the one I am now in; that is, the situations I am now living through, where the habit already seems inefficient and unsatisfying in the long-term, are BEST suited for keeping the habit. They provide the most payoff for the least price.
If this is the best world for using that particular past-time, then clearly the past-time is not worth using, as part of this world is my desire to rid myself of it. So, I did. Any change in my world now will lead either to the habit being less desired or less needed. This was the only place it could have been worth having and there is no room for it here. So there is no room for it at all.

I hate to double post, but I feel this deserves its own space.

I have recently been re-evaluating Newcomb's problem. I had not considered Newcomb's problem since I was quite a bit younger, and I two-boxed in those days, thinking "But certainty-" followed by something unrelated to probabilities. Since then, I've re-examined the problem and decided that one-boxing holds the most water. Anything less is accepting less. It's losing.

Now, proud in my new appraisal of the problem, I have a chance to apply it to real life. Great.

Box A: I live in a desperately poor, desperately uneducated, desperately uninteresting rural area. My mentor, who ran the local library, has passed and I am being eyed to fill the shoes of director of the library. The job will pay little, which is plenty for this area. I can take the job and have security. I will never do much, but I will never have to do much. Of course, there is also the idea of "I could just do it for a few years," but then I imagine myself slowing ticking from 24 to 27 to 30. I wouldn't improve or do anything important. I'd just run a library that requires little effort to run. If I left a few years later, I'd have nothing to show for it except the library. So, why leave the library at all, at that point? At least here I have security.

Box B: I have been talking with a friend about the possibility of us moving to a larger city (one that has an LW group, if that tells you anything). The city would have more to do, more to see, and, most importantly, I would no longer feel like the smartest person in the room. People better than me would be frequent, forcing me to actually learn and improve. I could involve myself in groups that wanted to do more than work. The only thing is I have no job lined up and none of my applications are returning hits in this city. If nothing comes in by June and I still resolve to do this, I may have to go without any possibility of work in my field. Maybe one of my applications will eventually strike or maybe not.

In Box A I have security. In Box B might be something worthwhile, but unlike Newcomb's problem, I'm not guaranteed $1,000,000 even if Box B is full. It may be $500,000. It may be $500. It may be nothing.

I always want to put what I learn to use. I just did not realize it would be this discouraging. I'm not sure how to win at this game. Which is why I'm glad life is not played as quickly as thought experiments!

You could use a simple and safe job as a leverage for other opportunities. Not sure how much of the following could apply to your situation.

Generally, sometimes you have an idea of a project which gives you (a) a lot of money later, but no money right now, or (b) a lot of other utility, but no money. If you need to pay your bills, you need to take some job. But sometimes the job is exhausting, and you have no more time or energy left to work on your project.

One solution is to try making a lot of money, and later take some free time and do the project. A typical failure mode here is that many people, when they start making more money, also increase their expenses (having children increases the expenses, too), so the "later" never comes.

Another solution is to find a job that is paid less, but leaves you enough time and energy for your project. A typical failure mode here is procrastination; you make less money, but you actually don't do your project. Another failure more is having a wrong model, so the less paying job actually does not leave you enough time or energy.

Example 1: A person took a job of a night guard in a factory. The job required them merely to stay all night in the empty factory, and every hour or two take a flashlight a take a five-minutes walk around the factory to check if everything is okay. The pay wasn't good, but paid their bills. They had to reverse their day/night cycle, but they had a lot of free unsupervised time during the working hours. -- The person took a notebook to their work, and during a few years they wrote a few successful novels, and became a superstar in their cultural niche.

Example 2: A person took a part-time job as a programmer, to pay their bills, and uses the other half of their day to program mobile games. They really work hard on their games (but still have a lot of free time), and experiment with different approaches. At this moment, after a year of this approach, they have released two games which together provide them a passive income of $50 a month. Which is not much per se, but it's the first two games they ever made. There is a lot of uncertainty here, but under some optimistic assumptions, if this passive income persists, and if their following games will provide more money or take significantly shorter time to develop, they could retire in ten years or so ($1000 a month is the average income here, so they are already at 5%). And the backup plan is to find a new job, developing mobile applications. (Being employed as a developer and developing in your free time may be legally risky; consult this with a lawyer. In this specific case the job is unrelated to mobiles, and the contract allows side income from non-competing projects.)

Example 3: A person made and saved some money, then took a few months of free time to make a computer game. Spent the whole time reading web; accomplished nothing. After a few months they updated, and humbly returned to the usual rat race.

Example 4: A person took a less paid but more simple and pleasant job (the estimate was based on a past experience of the same role in another institution, not available now) and wanted to make a computer game in their free time. However, in this specific institution the role actually was neither simple nor pleasant. (It turned out that the previous experience was very exceptional.) After a few years the person updated, and humbly returned to the more exhausting but better paid kind of jobs, having lost a lot of money in the opportunity costs during those years. With a little job-hopping they discovered that there was also a significant variance in that kind of jobs (despite having the same job titles and almost identical wages), and found something that seems like the local minimum in difficulty.

It saddens me to admit that the examples 1 and 2 was someone else, while the examples 3 and 4 was me. :(

So, if there is a chance of having a free time during your job (I am not sure if this applies) or working a shorter schedule, you could use it to make some project, or to learn new things. There are many free online courses: if you'd take them seriously (as in: two or three hours a day, every day, and do all the homework), within one year you could learn a lot.

Okay; I'm not saying this is necessarily a good strategy -- it failed for me; but on the other hand I personally know people who have succeeded; the critical factor is probably conscientiousness vs. procrastination -- but I think this is another option you might want to consider. Your local alternative doesn't have to be only "library", but could also be "library+project".

That person fails precisely in the "have a small safe income to pay your bills" part.

He also horribly fails at updating. His whole plan seems to be writing emacs tutorials and selling them for money. (I didn't read the whole website, maybe there are more plans, most likely not better.) Besides being an obviously bad business model, he is actually very bad at writing; which means he can't even show his tutorials to a potential employer and expect to get a technical writing job or something like that.

This is something that cannot happen in the "library" scenario. At least not soon; maybe a few years later, if the library job becomes unavailable. On the other hand, it can happen in the "moving to a larger city where no one responds to my job applications" scenario, if the potential employers keep not-responding, there is no income, and the rent goes up. But I didn't want to discourage; it only takes one job offer to get out of that situation.

For another horror story, I personally know a guy who decided that he will refuse all programming job offers until he gets an offer using SmallTalk (because "Java is only for idiots", although he made decent money from it in the past). He had a lot of money when he decided to do this, now, a few years later, he is homeless. True story. (If you ever visit Slovakia and find a homeless guy with a notebook, that's probably him.) Again, he failed at having a small safe income. That's the point. You either have some way to pay your bills, or you quit your plans before you hit the bottom and break your legs.

Most people seek "paying your bills" and "making a carreer" in the same package. It's probably a good strategy in many cases. But there are other options, too.

I knew a guy who worked for government. He told me he only worked one hour a week. (His work consisted of receiving a Word document with some information, converting in to HTML, and publishing on the website.) He spent rest of the time browsing web. When he discovered that the new version of Word has a "save as HTML" function, so his one hour of work becomes literally ten seconds, he became depressed, and later changed jobs. I kinda didn't understand him. If I had that job, I wouldn't quit. Instead, I would use this free time to do my hobbies and projects. Like translating Sequences, making websites for my friends, writing textbooks, learning new programming languages, making computer games, etc. And I would gradually prepare an exit plan, so that one day I would quit the job, but the next day I would activate AdSense on the dozen websites I would have made before, and activate purchases of virtual goods in the mobile games I would have made before. (To make it legally okay, that I wasn't making money while working for the government.) These are lost opportunities. And maybe... maybe not... the library is one of them.

He also horribly fails at updating. His whole plan seems to be writing emacs tutorials and selling them for money. (I didn't read the whole website, maybe there are more plans, most likely not better.) Besides being an obviously bad business model, he is actually very bad at writing; which means he can't even show his tutorials to a potential employer and expect to get a technical writing job or something like that.

The comments in the lengthy HN discussion suggest that his tutorials & other documentation are well-written, it's just his personal stuff which is written in his horrible idiosyncratic style.

Viliam_Bur: Thank you for the suggestions and the examples. I have decided to take the job and spend a couple of years gaining experience here with a pay rate that, while not large, is enough. The job is not such that I can sit on my laurels, so I will be active in the community which will encourage my own personal development and establish contacts. I'll still have free time to write and practice new skills. And, though the money is actually less than I originally was told, I should still have at least two hundred a month to add into my investment porfolio. Not much, but if I can get the right funds going, I shouldn't have to worry too much.

Lumifer: Thank you for the link. Such a situation is the exact one I wanted to avoid. And still want to avoid by ensuring that whatever I do next, I do it now. Not later.

An idea: Write on a paper your specific expectations about what you expect to happen in three months, six months, a year, two years. Make it four different papers, and put them in a place where you will find them; make a calendar note to read them at that time. Compare the outcomes.

The idea is that if your strategy wouldn't work, this is the way to know it. (As opposed to, e.g., keeping expecting that "one more year" and the things will improve. If nothing of your plans becomes real in one year, or if nothing finished successfully in two years, you are living an illusion, and it's time to give up.)

An idea I plan to implement tonight. Thank you Viliam_Bur.

I'm currently updating my beliefs about the LW community and the related sister communities. I still have the sensation of an outsider looking in, but I find myself relating to or agreeing with more of the socially strange beliefs of the community. Not because they come from the community but because, as I read about these beliefs, I find they make sense. And as long as they make sense, I have no real choice but to accept that they seem to make sense until I find reasons that show them not to make sense.

This is all good. You also have to be aware of frequency bias and confirmation bias. If you see the same argument a lot, and you see a bunch of evidence for that belief (because the stuff you're reading a lot has mostly positive evidence) of course beliefs are going to sound better.

When I first came to lesswrong (I do this with any community I encounter that has certain signs of a cult), I first look for counter arguments against their core beliefs, which gives proper perspective.

This is probably, more than anything, the main trait I've noticed in the sequences. Repetition of concepts. Every other articles acts as a collector of what's been explored before. So, it becomes important to look away from the repeated and cached ideas, to make sure that they still hold up outside of an echo chamber.

The ones that still hold start to feel suspiciously more real than competing ideas. Doesn't make them perfect maps. Just better ones.

I've been trying to study math. The problem was that I usually need the outside force of a deadline to get me to complete my studies on time instead of procrastinating indefinitely. For a while I didn't have a good solution to this problem, but recently zedzed very nicely offered to provide an artificial deadline for me. The way we're working it is as follows:

  • We agreed on a textbook I would us, currently Art of Problem Solving Introduction to Algebra (I discovered that I needed to back up all the way to algebra!).
  • We agreed on a tentative pace. That pace is still being refined, but currently stands at one chapter per week with 2 free late days per month. If I'm late and have no more free late days, my grade for that chapter is penalized by 20% for every 24 hours late.
  • We agreed on a grading system. The system we're using is slightly convoluted, since a lot of the problems are very challenging and I didn't want to hold it against myself if I didn't manage to answer all the hardest problems. End result is that I can technically get a grade of 120+ on many chapters, though my goal is mainly to score around 100.
  • I write up all my answers in LaTeX, turn it into a pdf, and email it to zedzed.
  • I report my grades to the LW community by posting them on the Rationality Diary thread, and zedzed confirms them.

We only started this recently. So far my grades are as follows:

  • I felt I already knew the first several chapters well, so I skimmed through them and only did the very hardest problems. I did not use the regular grading system for these chapters but rather just used a simple percentage of (correct answers / total number of questions attempted) * 100. Grade for Chapters 1-5: 94
  • Chapter 6: 100 out of 109 possible
  • Chapter 7: 91 out of 123 possible

Started an exercise program that I feel I'll be able to keep up. It's really simple, no more than 15 minutes a day, and you don't need anything except enough floor space to lie down in. I've been doing it for almost a month now, only missed 2 days so far.

I'm a little concerned about "bend and bounce" in the Lifetime Ladder. Doesn't bouncing -> joint problems?

I don't know, why would that cause problems? I'm only just starting rung 15 today. I'm actually not sure what that 'bending' exercise is for, or why adding a bounce would make it more difficult.

A few sources are short on citations, but basically say "don't bounce, you'll hurt yourself":

The AAOS also says "don't do it":

I'd guess the first-ladder bending exercise is meant to stretch your hamstrings, mid/lower back, and abs, as well as serving as a gentle warmup. Adding a "bounce" would ostensibly further stretch the hamstrings.

I am not a doctor, but would suggest replacing anything bouncy with a deep inhale + slight release of stretch followed by exhale + deepening of stretch (in this case, just folding over some more.)

I think our guy is maybe just using the wrong word. What he says is 'bounce up a few inches', and I think he probably means the same as you when you say 'slight release of stretch'. The bad kind of stretch, what your first link calls 'ballistic', would be bouncing down. But because of the 'bounce' in his description, people might do that anyway. :( I like your way of describing it much better, it is clearer and safer.


  • Going for streaks of getting ALL my daily habits done. These streaks were broken during my trip to California, which I stubbornly refused to put attempting productivity on hold for.


  • Hung around Berkeley and Bay Area to get some idea if I'd like it there. My sense from this trip was that I really, really hate cities. They're expensive and noisy. I did like the diversity, you don't really get that in Utah. Salt Lake is as diverse as it gets here, and there it's mostly homogeneous with small pockets of prideful "This subculture exists!". But I wouldn't necessarily live in the middle of the city, so that doesn't tell me anything too useful. This was last-minute and not terribly well-planned, I failed to visit the rationality grouphouses I'd most likely attempt to move in/near to. I had fun, but didn't accomplish much for my primary goal.

  • Stopped following the mathematical philosophy class on coursera. I will look into the later videos of the course which seem to pertain to bayesian reasoning and epistemology for curiosity's sake, but I'm not seeing a lot of personal value in learning philosophy. I was mostly interested in it for self-signaling that I'm smart and cultured, but I really want to be usefully informed more than I want to impress academics.


  • Planning to do some offline training of my morning routine. Right now I look at my to-do list and spend some time waffling about which morning activity to do first. Some items take more time or mental energy than others, so I subtly put off starting them.

  • Put a picture of a valkyrie on the head of my info board as a reminder to find and protect the "life force" (goals, happiness, well-being) of my closest people. Tsuyoku Naritai -- I actually want it done, not just to satisfy my impulse to feel like a caring person.

Emotional responses

  • Also just generally trying to train my visceral imaginings to what is likely to happen instead of what happens to be cognitively available. For example, when thinking about how to keep family safe I realized I ought to be picturing heart issues, diabetic neuropathy, worsening asthma, UTIs, and drowsy driving reaction times more frequently than dramatic spinal cancer, or getting hit off the road by a drunk driver, or stabbed to death by a random stranger. Next action is to look up actuarial tables.


  • I joined a challenge to do duolingo Spanish daily. Now I've stopped enforcing daily duolingo practice. It's still on my habits as something to reward myself for, but it won't hit me as a negative thing if I don't do it every single day.


  • Stopped keeping a daily assessment of how far I'm going with a "perfect day" streak, in which I get all my daily repeat items done. My system already supplies rewards for streaks. If you try to use a stick instead of a carrot on my subconscious, it will take the stick from you and start beating itself up willy nilly.

  • Increased the "price" of rewards like an hour of fanfiction, dipping into my chocolate bowl, and going out for drinks. I'm way overloaded on GoldPoints in HabitRPG, it doesn't mean anything to hit the reward buttons if I can't see an effect.

emotional responses

  • Spent a few days in a haze of fanfiction. Curse you, Alicorn, for making effulgence so entrancing. Curse you too Eliezer, for recommending Toasterverse. I finally broke out with an offhand realization that I hadn't done a happiness meditation in a while. It had all been mindfulness lately, as opposed to the style of sitting and thinking about whether I could be okay with just being me. A lot of minor things were irritating me, and I was starting to internalize feeling uncomfortable as an inherent property of being me. Fanfiction is an okay escape from being me, but it's difficult to stumble upon better solutions to a problem I'm not allowing myself to acknowledge.

Interesting data I am still digesting: I ran across this story, which summarizes the results of a study someone did to see if doctors were able to correctly interpret test results. original source. The important thing to note here is that these people are generally trained on this notion that tests have false positives and negatives and how to deal with that. The poor success of doctors to do this kind of analysis, despite being trained on it, and potentially using that skill every day, and caring about the outcome, and being in the presence of the outcome fairly often - this has raised my perception of the difficulty required to train people to learn and apply even small amounts of probability theory. Specifically, I think this provides some hint that moral or utility arguments are not going to convert the masses. I've been wondering if there's a way to trick people into developing and applying skills.

Fun project: I started working on a dinner game. I'm not entirely sure what the end form of this game will look like, but right now, it involves sticking an object in a closed box, passing the box around, and having people try to determine what is in the box without looking inside it. I tried this out with my family last night; they seemed to enjoy the idea of trying to experimentally determine properties of an unseen object and then trying to figure out what it might be. They were able to pretty precisely determine the shape of the object and two were able to guess the object correctly (though one changed position upon hearing an observation from another player). Several identified unreal properties of the object, but didn't fixate on them. Based on the success of that run, I am going to try more exotic shapes and materials in future games to see how that changes things. Ultimately, I'd like people to weigh in on propositions like, "The object is red" despite being uncertain what the object is.

I've started taking an online course on naturalism, and have been quite surprised by the quality of the discussion questions, which have illustrated to me that my thinking is not particularly precise about various unlikely metaphysical propositions. And speaking of metaphysical propositions and imprecise thinking, I am still looking for a good argument somewhere about how to set sensible priors for metaphysical propositions.

I'm not sure if you describe Kim's touch game (the Wikipedia entry doesn't mention this only in German). I have played touch Kim with my sons and they can easily recognize most items when 5 years old so I wonder what the limitations of your box were - or what unusual items you chose. Many items can be recognzed by feeling them with the feet only. Remembering many of them is more difficult though.

I'm not familiar with Kim's touch game, but I did run across a different game that someone came up with to practice applying Bayes theorem, which involved touching people in the shoulder with one of two objects (in the example, it was a coat-hanger, if memory serves), and then testing their ability to update their predictions based on more information about the object they were touched with. I wish I could find that page again, but I haven't been able to find it again. It might have even been linked off of a LessWrong meetup group.

I'm less concerned with getting people to apply Bayes theorem (which would be GREAT, mind) than I am with getting people to be more comfortable with collecting information, sharing observations, and not getting fixated on personal theories. I'd especially like them to get comfortable with making the jump to making reasoned predictions about hidden properties of objects, given their theories about what an object is, but I'd like to find a way to make that process at least as fun as shaking a box to determine its contents.

The first game was just a deck of cards and a cardboard box large enough to allow the object to be flipped (though one type of flip was not possible in certain orientations). The players were all adults and I consider them to be quite astute; I expect that children could also play, and it could be a useful lesson about how to not get fixated on your own ideas, how to incorporate observations from others, and how to share observations constructively.

The idea came up in a discussion with a friend about how terrible our science classes had been as children, and how learning individual facts was not particularly useful.