All of adamzerner's Comments + Replies

Bayes' theorem, plausible deniability, and smiley faces

Huh, that's interesting. I wouldn't have thought of that but now that you mention it, I also feel like that sort of perception can happen. However, I also get the sense that when you're in such a situation, you'd be able to tell, and thus avoid having your smiley face misinterpreted.

Bayes' theorem, plausible deniability, and smiley faces

Thanks! That's what it was intended as. I'm not sure how receptive the LW crowd is to posts like these, so it's good to get your data point here.

Bayes' theorem, plausible deniability, and smiley faces

Yeah, but then there's the same problem, just with exclamations points instead of emojis. (Of course, calling it a "problem" in the first place is a little silly.)

Covid 4/9: Another Vaccine Passport Objection

Vaccinations were not prioritized perfectly, but they were prioritized well, and they protect even better against death than infection.

I'd be very interested in hearing (from Zvi or anyone else) more about where you stand on the extent to which vaccinations protect against death. Death is the main thing I'm trying to avoid. If we were confident enough that it protected against death enough, I would up my risk tolerance.

2ChristianKl1dGiven the numbers COVID-19 vaccines are less effective against asymptomatic infections then they are against symptomatic infections. We also see them to be more effective against hospitalization then against symptomatic infections. Given that death is at the tail, they are likely more protective against death then asymptomatic or symptomatic infections. The effect they have on long Covid is less clear. Long Covid won't kill you immediately but can plausibly cost you a few years of lifespan.

Yeah, agent vs object is the main thought I have as well after reading this post.

Coming from the perspective of affordances as belonging to the object, I have been thinking about the following recently. In programming, one of the big reasons to "choose the right tool for the job" is because of affordances. For example, suppose you have a bunch of functions that you want to wrap in a namespace. You can have a class and then make all of the functions static methods of the class. This is something I see Ruby programmers do. However, a class affords a few impo... (read more)

Rationalism before the Sequences

Great point. A few (related) examples come to mind:

  • Paul Graham's essay The Top Idea in Your Mind. "I realized recently that what one thinks about in the shower in the morning is more important than I'd thought. I knew it was a good time to have ideas. Now I'd go further: now I'd say it's hard to do a really good job on anything you don't think about in the shower."
  • Trying to figure out dinner is the worst when I'm already hungry. I still haven't reached a level of success where I'm satisfied, but I've had some success with 1) planning out meals for the n
... (read more)
Rationalism before the Sequences

Thanks for making that connection to Zen Buddhism. I never thought of it as a central theme of The Sequences before this.

I'm still not sure if I'm convinced that it actually is a central theme. In the preface to Rationality From AI to Zombies, Eliezer writes:

It ties in to the first-largest mistake in my writing, which was that I didn’t realize that the big problem in learning this valuable way of thinking was figuring out how to practice it, not knowing the theory. I didn’t realize that part was the priority; and regarding this I can only say “Oops” and

... (read more)
3Eric Raymond13dI actually wouldn't call Zen a "central theme". More "a recurring rhetorical device". It's not Zen Buddhist content that the Sequences use, it's the emulation of Zen rhetoric as a device to subtly shift the reader's mental stance.
Think like an educator about code quality

It seems to me that your post is missing something: what specifically do you want people to learn?

Hm, I'm not sure I'm following. It sounds like you're asking this from the perspective of "I'm a developer. What specifically do I want to teach the other developers about how this code works?" The answer to that totally depends on what the code is for. I do think it would have been better if I had a running, concrete example to reference throughout the post though.

For code-quality I think you're asking "how do I help colleagues with different expertise w

... (read more)
The best frequently don't rise to the top

Hm, I think we agree about a lot of things. There are a lot of people I'd prefer to watch over Bottura for all of the reasons you said. But I like to supplement the other stuff with Bottura. Which, actually, sounds like something you also agree with.

I agree that all of the examples I gave are things that you could pick up, probably more efficiently, from other sources. But 1) getting the stamp of approval from someone like Bottura allows for a belief update that I think is larger than the update you can perform by hearing the sixth moderately skilled YouTu... (read more)

The best frequently don't rise to the top

Thank you for your input here, it's great to get some more data points!

For Eric Normand I think it can be a bit hit or miss. Some of them I find to be meh, but others I really love. Two that I've watched recently that I've loved were How is Haskell faster than C? and The Early History of Smalltalk. The former opened my eyes to a perspective on how to think about how "fast" a language is. The latter I really appreciated because he distilled a paper that I am interested in but otherwise would never have been able to parse. Again, it makes me very sad and con... (read more)

The best frequently don't rise to the top

Definitely closer to the first than the second. I'm not trying to touch on the second at all, but I also don't think the first is entirely accurate either. In all honesty, I'm having trouble articulating what it is that I am trying to articulate. Let me give it another stab:

  • Imagine a metric that encapsulated everything related to quality. Call it "encapsulated quality". (I'm not just trying to point at "raw quality" alone in this post. I am totally onboard with the idea that various other things matter.)
  • Imagine a world where "encapsulated quality" leads
... (read more)
The best frequently don't rise to the top

Funny enough, I've recently been contemplating asking a question here on LW like "what actually-quite-good youtube channels are out there?" precisely because I suspected there were a lot of hidden gems like you mention! I didn't get around to it, though, perhaps because watching youtube feels low status so I felt like it would be embarassing to visibly put effort into optimizing my youtube-watching.


Also I think you might really like Eric Normand. In particular the videos/podcasts where he spends 60-120 minutes exploring, distilling and discussing ... (read more)

The best frequently don't rise to the top

Why, concretely, would I watch one of his videos?

Good idea about getting more concrete. Here are some examples that come to mind:

  • A few weeks ago I had some leftover pasta and meatballs. The pasta wasn't enough to feed myself and my girlfriend, so I opened up a box of pasta that was shaped differently, eg. rigatoni when the leftovers were shells, and mixed it with the sauce and meatballs. My girlfriend was upset and thought that was really weird. I left that experience feeling confused about my culinary intuitions, because while she had strong feelings
... (read more)
6gjm18dThanks for the concrete examples. Of course the drawback of getting concrete is that since my reasons for finding them uncompelling are also rather concrete and specific and don't fit into any particular pattern, it's not clear what conclusions to draw :-). But: * I already make pasta-bake dishes with different kinds of pasta, and I didn't need to watch a multi-Michelin chef to tell me that was OK. * My bechamel sauces are already pretty simple. * I already try to use high-quality ingredients. (This is a thing that's emphasized, e.g., in a lot of cookery books.) It may be that the likes of Bottura are better than I am at seeing which ingredients it matters more for, but I think I have reasonable intuitions for this already. * I already make pasta-bake dishes with different kinds of cheese, and again I didn't need a multi-Michelin chef to tell me this. (If there's a specific thing I learned this from, it's probably the experience of eating a "quatro formaggi" pizza many many years ago. No multi-Michelin chefs involved there either.) * When I make pasta-bake dishes I already leave the cheese in chunks. Again, no multi-Michelin chefs involved. * I agree that the fact that he has his ingredients prepped before the start of each video is some evidence that mise en place matters even in non-restaurant situations. But (1) there are already plenty of cookery books aimed at home cooks urging you to bother with mise en place, (2) there are plenty of non-world-class cooking YouTubers who also get their ingredients prepped before they start / at the start, so you could learn the same lesson elsewhere, and (3) it shouldn't be very strong evidence because another explanation is just that watching someone finding things in their cupboards doesn't make very compelling video. Maybe there is a pattern to the above: the things you say you learned from watching a master at work, I (and I'm fairly sure many others) managed to le
The best frequently don't rise to the top

The meritocratic part is the best are significantly more likely to rise to the top, real world is best thought of as a stochastic place, full of imperfect information and surprises.

That makes sense, I agree.

The best frequently don't rise to the top

Yeah, maybe. My thinking is that some people care about it more than others, but in my experience watching YouTube videos (and I happen to watch a ton of cooking videos on YouTube), I recall seeing videos that seem basically even to Massimo in terms of production quality, personality, etc., yet have hundreds of thousands of views.

The best frequently don't rise to the top

Agreed. I think the caveat you make about some people preferring one and others preferring another use case is an important one, but in general I agree that my logic was wrong and it's an important mistake to avoid.

The best frequently don't rise to the top

You are confusing expertise in different domains: just because one is exceptional in something, it does not follow they are good at teaching it or making videos of it.

I agree about there being different domains, that each is important, and that expertise in one does not imply expertise or even competency in another.

You might be able to make points about how being brilliant doesn't necessarily make you good on camera, or about how marketing and promotion matters a lot, or about releasing videos on a consistent schedule, or about how maybe they're not p

... (read more)
8Viliam18dMaybe the one-number metric that matters on YouTube has a very strong component of how technically well the video was made. Like, even if someone who "deserves it" makes technically great videos and gets tons of followers, most of them follow the channel for "wrong reasons", that is, they would not have subscribed if the same content had lower technical quality. So if someone who "deserves it" makes technically crappy videos, they only get a few followers. In other words, most people watch YouTube because they want some short funny visual experience. Not because they want to learn from world-class experts. Okay, maybe they watch something short funny visually attractive that allows them to pretend to themselves that they are doing it to learn. That would kinda reduce your thesis to "the best at X frequently don't rise to the top when Y matters". The difficult part is figuring out what exactly it is that matters on YouTube (or on the market).
The best frequently don't rise to the top

Good point, exponential does seem like it could sense. I'm not sure.

The best frequently don't rise to the top

I agree with the general point but disagree that that's happening with Massimo, and more generally with the other examples of lack of success. I do think that's part of what happened with my startups though.

The best frequently don't rise to the top

My perspective is that things like the camerawork and background noise in his videos a) aren't particularly bad, and b) given (a), I'd think it'd only knock him down a few points. Maybe from, let's say 800k views to 700k or 600k, not all the way down to 10-20k (very ballpark-y of course).

As far as Michelin chef magic, personally I feel like I have picked up some things, like the amount of ingredients he uses. I see it as valuable to be able to watch someone with that degree of mastery. A lot of times such people don't actually know how to explain or articulate what they're doing correctly, so observing them can be hit or miss, but still seems pretty valuable to me.

It's not just the camerawork and background noise. The information density is low. The bad camerawork, as well as being annoying in itself, makes it difficult to see any details of what he's up to, so you can't e.g. improve your knife skills by observing exactly what he does when chopping things. He explains rather little of what he's doing. At least in the video I watched all of, things like quantities and times are left unspecified. He wasn't doing anything particularly imaginative or highly polished. I can readily believe that I was watching a master at... (read more)

7dmnissley18dI think you're underweighting production quality (false consensus bias?). Just as an example, my girlfriend is big into youtube, but she only watches high production channels, and will quickly skip past anything that looks like it's been shot on a phone. She'll also basically refuse to watch any movie or show that was shot before about the year 2000 for similar reasons. Another big factor is personality of the creators. She likes to watch channels where she feels some connection to the creators. They're usually a couple of people or more where there is entertaining and comedic banter that's always going on besides whatever subject the actual video is about. It's chiefly entertainment for her, with any educational content just being a bonus. I would posit that most people who watch youtube have preferences closer to hers than to yours, although my own are probably closer to yours. Here's an example of the kind of video I'm talking about:

This seems to assume roughly linear relationship between quality and reception. As I mentioned in my other comment, this seems far from necessary. We can have something like an exponential relationship between the two, in which case a small difference in quality can create a massive difference in reception.

I just don't get why you think moderately bad camera work should a priori best be thought of as some percentage of views lost (taking something from 800k to 700k or 600k) rather than in orders of magnitude (taking something from 800k to 80k or 8k).

The best things are often free or cheap

Fantastic point! That didn't hit me until you pointed it out. We live in an amazing time :)

The best things are often free or cheap

But I don't think anyone would seriously contest the idea that difficulty of reproduction is an important factor which drives price. The question is more whether there are "many" top-in-class experiences which you can only have via difficult-to-reproduce things.


Cars, phones, laptops, doctors, shoes, clothing, watches, jewelry, haircuts, makeup (both the physical makeup you might buy, and the attention of a professional to apply it really well), pillows, mattresses, chairs, yards/parks, gold, silver, bitcoin, ...

Most of the examples you mention seem ... (read more)

The best things are often free or cheap

Hm, when I wrote the post and decided on the term "the best", I didn't actually think too hard about whether that is literally true. What I had in mind were things that are very, very close to the best. I'm not sure exactly what that means. Where is the bar that you have to pass to meet my operational definition of "the best"? I can imagine situations where things pass the bar and they aren't even in the top tier of things. But these things are still very close to being "the best".

Thinking about it now, I still believe that "the best things" is a more accu... (read more)

What are fun little puzzles / games / exercises to learn interesting concepts?

The Credence Calibration Game

The idea is to answer a bunch of questions like "How confident are you that Baltimore is larger than Dallas?". The goal is to get better at confidence calibration. Eg. when you say you're 70% sure you end up being correct about 70% of the time.

1davidgasquez1moNice resource! It reminds me about the Chapter 3 of Adventures in Cognitive Biases []. Is there anything similar to The Credence Calibration Game but works online and has some recent facts? It would be cool to be able to replay some closed Metaculus questions and see how well would you do against that!
The best things are often free or cheap

True. I'm having trouble thinking of good examples of things that have larger replications costs. Food is a good example. Companionship isn't the sort of ~consumer type of thing I was going for here. Living accommodations is interesting. At first I didn't include it because I was focused more on single experiences, but now that you mention it, I think that it belongs. Eg. experiencing what it is like to live nicest, most comfortable, luxurious atmosphere. That is something available for purchase via expensive hotels or Airbnbs, and they are in fact very expensive, consistent with your point about replication cost. Do any other examples come to mind for you?

3abramdemski1moCars, phones, laptops, doctors, shoes, clothing, watches, jewelry, haircuts, makeup (both the physical makeup you might buy, and the attention of a professional to apply it really well), pillows, mattresses, chairs, yards/parks, gold, silver, bitcoin, ... OK, out of that list, the only thing that's free/cheap is parks. (Of course my list is very biased, since I was primed with the idea of expensive things. But I don't think anyone would seriously contest the idea that difficulty of reproduction is an important factor which drives price. The question is more whether there are "many" top-in-class experiences which you can only have via difficult-to-reproduce things.)
The best things are often free or cheap

Interesting take. I don't know enough about markets to really say. To feel confident in your argument I'd want to feel confident that 1) your list of A and B is exhaustive and 2) that both A and B are true. To feel confident in 1 and 2, I'd want to see if I could come up with counterexamples. That seems like a difficult thing for a person to do though, because the space of best things is large.

Five examples

It is like having two people called John, but you point your cursor at one of them, and the IDE understands you mean that one... instead of simply doing textual "Search/Replace" on your source code.

Thanks for explaining this. I had been planning on investigating how it compares to search and replace, but I think this clarified a lot for me.

And thank you for the rest of your thoughts too. I think my lack of experience with static typing is making it hard for me to fully grok them, but I am groking them to some extent, and they do give me the vibe of being correct.

Five examples

Great points. I see how making your types more specific/complicated helps you catch bugs, and the example with your compiler really helped me to see that. However, making types more complex also has a larger upfront cost. It requires more thought from you. I don't have a good intuition for which side of the tradeoff is stronger though.

The feeling of breaking an Overton window

Share observations (not theories) of what it’s like to be you right now trying to look at this stuff.

Thinking about this situation, I feel that same sense of awkwardness and discomfort at the idea of saying something like, "Yes, I'm stocking up with a ton of food because I'm worried about the virus." I also get a sense of that same alien process driving me away from such an answer.

Five examples

None come to mind right now. But the thing is, that probably just means I can't think of them as opposed to them not existing.

Five examples

Good to know. Thanks!

Five examples

Yeah those caveats make a lot of sense. However, I strongly suspect that they're not exhaustive. Not that you're implying they are, but it's important to note because when you acknowledge that they're not exhaustive, you rightly treat this sort of advice as more of a heuristic than a rule.

1crl8262moSure. I'd love to hear what other caveats you think are important.
Five examples

Thank you for those thoughts, they're helpful.

  • I actually was aware of the autocomplete benefit before. I've only spent about three months using a staticly typed language (TypeScript). In that time I found myself not using autocomplete too much for whatever reason, but I suspect that this is more the exception than the rule, that autocomplete is usually something that people find useful.
  • I wasn't aware of those benefits for refactoring! That's so awesome! If it's actually as straightforward as you're saying it is, then I see that as a huge benefit of static
... (read more)
3IronLordByron2moNo. I compulsively use the refactor/rename operation (cntrl-shift-r in my own Visual Studio setup) probably 4 or 5 times in a given coding session on my personal Unity project, and trust that all the call sites got fixed automatically. I think this has the downstream effect of having things become a lot more intelligible as my code grows and I start forgetting how particular methods that I wrote work under-the-hood. Find-all-usages is also extremely important when I'm at work; just a couple weeks ago I was changing some authentication logic for a database we used, and needed to see which systems were using it so I could verify they all still worked after the change. So I just right-click and find-usages and I can immediately evaluate everywhere I need to fix. As an aside, I suspect a lot of the critiques of statically-typed languages come from people whose static typing experiences come from C++ and Java, where the compiler isn't quite smart enough to infer most things you care about so you have to repeat a whole bunch of information over and over. These issues are greatly mitigated in more modern languages, like C# (for .NET) and Kotlin (for the JVM), both of which I'm very fond of. Also: I haven't programmed in Java for like three years, so it's possible it has improved since I touched it last. Full disclosure: pretty much all my experiences the last few years have been of statically-typed languages, and my knowledge of the current dynamic language landscape is pretty sparse. All I can say is that if the dynamic language camp has found solutions to the refactor/rename and find-all-usages problems I mentioned, I am not aware of them. You can get some of these benefits from optional/gradual typing systems, like with Typescript; the only thing is that if it's not getting used everywhere you get a situation where such refactorings go from 100% to 90% safe, which is still pretty huge for discouraging refactoring in a beware-trivial-inconveniences sense.
Your Cheerful Price

Thanks! This has definitely helped me to think about the concept of cheerful prices. Here's my current position.

I do see the value in avoiding the situation of "I'm paying you to do X, you accept, but are secretly annoyed about it". By paying instead X + cheerfulness bonus you avoid it. However,

  1. I don't have much IRL experience with rationalists, but I would expect that if you buy into the idea of exchanging money in these sorts of scenarios, that you'd also buy into the idea of Tell Culture, at least enough such that you can have some back and forth and av
... (read more)
7Ben Pace2moSolid points. 1. While in my experience the rationalists have some of the best conversational norms for communicating about conflict and costs and disagreements, I would not say that the rationalists I meet have solved these problems, to the extent where there are not costs that are very difficult to do conscious accounting of. So from one perspective, I’ll take all the tools I can get, and this seems like it may help with some such situations. 2. That said, I think you’re right that the cheerfulness bonus is probably too large in some of my examples. The actual cheerful price for the one I have in mind would’ve been... I feel confused, somewhere between $50 and $500. Still, I think it would’ve been a bit high for them. But my fair price would’ve been lower. I guess I’ll look for opportunities to use cheerful prices with the people I know. I’ll see if I can find at least three occasions to use it in the next few weeks.
Five examples

That would be great! And is a great example of how coming up with examples is often difficult.

Your Cheerful Price

Do any other examples come to mind? I'm finding it difficult to think about without, say, 5 concrete examples to latch on to.

  1. What’s your cheerful price for not drumming your fingers on things when you’re in the living room
  2. What’s your cheerful price for being responsible for background music at our party tonight?
  3. What’s your cheerful price for keeping your mask on until you enter the house, rather than take it off near the house?
  4. What’s your cheerful price for cleaning up the kitchen as soon as you’ve been cooking in there, rather than some point later that same day?
  5. What’s your cheerful price for driving me to the airport at 5am on Monday?
adamzerner's Shortform

I posted an update in the OP. When we initially talked about this I was pretty strongly on the side of pro-Awair+Alen. Now I lean moderately against Alen for most people and slightly against Awair, but slightly in favor of Awair for me personally.

Covid cafes

Funnily enough, after commenting, I ended up getting Chinese food takeout at a place that have a makeshift pickup window. Looks to me like it was pretty cheap and simple to set up.

8gbear6052moA Chinese restaurant near me built a plastic enclosure at their doorway, with an airlock that they place the food into, before closing the door and letting you open the opposite door. It works great, although it was definitely non-trivial to construct.
Covid cafes

My read on it is that companies don't give a shit about customer wellbeing and are almost solely interested in profit. If the impact on customer wellbeing is very large, sometimes they'll care, but 20 microcovids/customer is way below that threshold.

My guess would have been that there were a variety of people, and that many people were going to great lengths to avoid getting covid, so at least their patronage would be altered by the covid dose one gets with one’s coffee. But that doesn’t seem true.

Yeah, I share similar feelings here. I sorta sense that the... (read more)

The 10,000-Hour Rule is a myth

Basically, the rule claims that anyone can become an expert if they spend ten thousand hours practicing. In other words, ten thousand hours of practice is necessary and sufficient to become an expert.

When you think about it, that idea is incredibly silly.

  • If you spend an hour a day cooking, you'll reach 10k hours by age 50, let's say. Is the average 50 year old an expert?
  • They say you spend 80k hours in your career. Are people 1/8th the way into their careers experts?
  • There are a lot of people I encounter who have been playing poker for decades and never progress past the beginner stage.
The 10,000-Hour Rule is a myth

Classical piano is an extreme case that illustrates the problem. Almost no one makes a living as a classical pianist, because listening to a non-famous classical pianist just isn't the kind of entertainment that people will pay for.

Yup. I think what you're describing is the superstar effect.

The 10,000-Hour Rule is a myth

You also need to use that time effectively, which Ericsson claims is best done via deliberate practice. As someone who half-heartedly practiced piano for two years to appease my mother, I can attest that just putting in hours does not promise results. Deliberate practice tries to maximize learning per effort invested.

There are probably many other examples of this, but for some reason the above just made the following lightbulb go off for me.

It's usually a good idea to start with a concrete example before defining the thing. This is something that seems wei... (read more)

Give it a google

Maybe, but "incorrect" is a spectrum. Sometimes it's a close second. Especially for someone who is googling for "basics of poker strategy".

1pseud2moI think a quick web-search is useful. Having read something is an improvement over having no knowledge, and it's ridiculous that people don't do a quick web-search more often. I'm not disagreeing with your point that Googling is better than doing nothing to learn at all. My first comment just pointed out that what you learn may be quite inaccurate or out-of-date. Now, I'll go further and suggest that what you learn may be purposefully misleading. When it comes to politically or financially sensitive topics (and a searcher won't always realise when a topic is such) those supplying the information you access may be influenced by (or be one and the same as) those with an interest in you receiving incorrect information. To continue the poker example, a good poker player is unlikely to give away information for free. Therefore the information you read after a quick google is unlikely to be particularly good (there are exceptions: some players will give out some minimal information in the hope that readers will then pay money for more information). For a variety of reasons, there is plenty of fine information about poker strategy on the internet, and plenty of the most basic stuff (which is, after all, what the web-search we are discussing is about) is free. For other topics, we won't have so much luck. When searching the web, sometimes we leave empty handed, sometimes we leave worse than we started. But mostly we learn something. I'm not disagreeing with the point of your post, just adding my own thoughts.
Give it a google

Yes, but I think "bet half to full pot" is the 80/20 answer, and the point of "give it a google" is often to get that 80/20 answer.

2pseud2moI assume you mean by "80/20 answer" that betting between half and full pot will be the correct sizing approximately 80% of the time one bets. I think the actual percentage is significantly lower than 80%.
Thoughts on Mustachianism

I see. I find that sort of debugging quite enjoyable.

However, I find that students often are very impatient when it comes to traversing deeper down the dependency tree, and instead impatiently just want to "get it working/get the answer" and move on. There are three separate instances in my life that I can think of where I experienced this recently: 1) a backend dev learning frontend stuff, 2) someone entirely new to programming I was tutoring, 3) a college student taking precalc.

Lessons I've Learned from Self-Teaching

I guess the crux is about Anki helping with retention. If you'd lose a lot of your understanding without Anki, I agree it's worthwhile. If you'd only lose a little, it seems like a better use of time to seek deep understandings elsewhere.

Load More