I knew the UK was arresting people for online posts, including private text messages, including ones that were very obviously harmless. I didn’t realize it was 1,000 people a month.
In cases like this, I would like to see a representative random selection of the messages. (I can't trust either side to give me that, instead of messages carefully selected to prove their point.) With some context, such as "the author sent 100 such messages during one week", or "this was written as a response to this", etc.
That's because I can imagine a world where most of those messages are credible death threats or something like that, in which case I generally approve of the law, even if it means that once in a while an innocent message gets accidentally marked as death threat (ideally that shouldn't happen, but is that the standard we actually use in other parts of life?). But I can also imagine a world where most of those messages are just expressing an opinion. I would like to make my opinion based on actual data, not on whoever decides to show me their selection.
on bsky you get added to a global block list and get hidden from half the people.
I am not familiar with bsky, does this mean that most of the users are from the same political tribe? More than a half, because not all people from the same tribe would subscribe to the same blocklist.
Is the future of internet bubbles that each political tribe gets their own social network?
when you buy a new computer, by default it will come pre-loaded with a lot of crap, and Windows 11 will waste a bunch of resources, and that Macs come with a lot of objectively terrible software, and that it takes a remarkable amount of deliberate effort to fix all this even if you mostly know what you are doing, during which you are subjected to ads.
Sometimes I am shocked when I have to use a non-tech person's computer or smartphone. I saw a friend starting his work laptop (a non-tech friend working for a non-tech company), the first few minutes was just various applications starting and showing an error message or asking to install an update... then he closed all the windows and started doing the actual work. To me it seemed like hell, for him it was Tuesday.
Later, my relative asked me about some settings on her phone. It was full of various Chinese crap that came preinstalled, the desktop (or whatever it is called on smartphones) was several screens long, each screen containing a few icons at random positions, but you couldn't re-arrange them because the desktop was locked (I didn't have enough patience to figure out how to unlock it). If I had a phone like that, I would throw it away and buy the cheapest non-Chinese phone, but for her, this experience is normal.
What I don’t fully understand is why all of this is true. It’s one of those ‘yes I get it but also I don’t, actually.’ It seems like an obviously easy selling point to say ‘this PC doesn’t come with any ads’ and similar.
I wonder, could it be that you are wrong here, and the non-tech people consider all technology intrinsically confusing to such degree that these additional inconveniences, horrible as they might feel for a tech person such as you or me, are just an extra 1% inconvenience for them, nothing to specifically complain about?
forcing your kid to live like a ‘normal person’ and obsess over small amounts of money so they can be like everyone else or learn to ‘value the dollar’ or what not, is actually terrible training for being wealthy, for the same reason lottery winners mostly blow their winnings. [...] You do want them to ‘earn it’ and more importantly learn how to operate in the world, it’s totally fine to attach some fixed conditions to the unlock of parts of the trust fund, but you don’t do that by giving them Normal People Money Problems.
Yes. If you make your kids spend a decade or two doing lifestyle X, and then switch them to lifestyle Y, they will come unprepared. Even worse, if the moment of switch is your death, you will not be there to help them.
Training to survive the "normal mode" seems like a good idea, but can be done in a more controlled way. As an analogy, the fact that I have a car (or could call a cab anytime I want) does not prevent me from using my feet. I can have a car, and simply choose not to use it today. And take my kids with me for the trip. I could even go further and e.g. spend a few days camping in a forest, leaving all electronic devices as home. You can do the same thing no matter how rich you are. (Okay, maybe at some level you will need to hire bodyguards to protect your family from being kidnapped. But they can follow you at a distance, and be specifically instructed to ignore your kids if they try to get their assistance with something.) Or you could buy an apartment in a different city where no one knows you, and spend one month each year living there, spending only as much money as the average person makes in a month, taking no debts. The kids would help do the dishes, cook, clean up the room; all the things that an army of servants does for you for the rest of the year. No rich-person objects allowed; the kids are not allowed to disclose your identity. (Hmmm, this sounds more pointless than I originally imagined, but hey, it's just the first idea that came to my mind.)
The problem with actually not having money is that it closes some genuinely good options for you: You take a part-time job when you should be fully focused on studying. You are unfamiliar with the latest technologies, because you can't afford the hardware. You don't attend conferences, because you can't afford the travel costs, and the job you have to keep does not give you enough vacation. Your car randomly breaks, and that creates a cascade of problems that increase your stress levels and ruin your plans. -- These are actual problems, unrelated to building willpower and working habits, which are the supposed goals of keeping the money away from you.
And if you know that you are going to get tons of money in future, it can actually reduce your motivation to actually solve these problems, because you know that the greatest improvement to your quality of life will come unrelated to all your personal effort, and all you need is just to survive until the day that happens.
I actually know a guy whose rich parents are probably doing the worst of both options. They refuse to give him any cash. But they won't literally kick him out of the house; there is always free food and bedroom available if he needs them. He just needs to pretend that he is trying. He is now over 40 years old, and he never kept a job for longer than three months, always with at least three months break between the jobs. He has an "internet business" (a few web pages with almost zero content, uses SEO to make people click on them, then gets referral money when the visitors proceed towards the stuff they were actually looking for, but spends most of that money on adsense to get traffic to the web pages) that makes a few hundred euro each month, and didn't grow after a decade. He was married and has a child, but his wife divorced him and lives with the child in a different city. Despite spending most of his time online, his English, both written and spoken, is barely intelligible. One day his parents will die and he will inherit the millions. But he has already wasted most of his life, and will probably waste the money, too. (Not leaving much to his kid.) And then he is completely screwed.
I am not sure what exactly would happen in a parallel universe where his parents created a trust fund that gives him a certain unconditional perpetual income. At least, he does not need to do all that pretend-work. Either he plays computer games all day long, or he might contribute to the family business in some little way (like maybe doing SEO for his brother's company instead). Maybe his family didn't fall apart because of financial problems.
1Me: What do I have to wear to this thing?
Jessica: You can wear anything you want.
I wonder what sources of information on clothing he believes that she is privy to and he is not.
He's asking her for rules about clothing in a way that sounds like he's asking her to choose a specific outfit.
She's replying as if he asked whether a specific outfit was necessary, and as if she assumes he already knows the rules governing how to select an outfit for the occasion.
I wonder why such a smart guy has had the same unproductive conversation 100 times instead of seeking alternative ways to communicate the underlying question more clearly, such as "would <clothes I usually wear> be acceptable for this?"
I'm not at all privy to the financial planning approaches of the very wealthy, but now that people tend to live into their 80s, how are options #1 and #2 not also forcing the future inheritors to mostly provide for themselves until age 50 or 60? They may not be so worried about saving for retirement, and know there's a level below which their parents won't let them fail, so it's not the same. But, it seems like without a trust fund they're not really going to inherit significant wealth until their own kids have graduated college?
It’s always a nice break to see what else is going on out there.
Bad News
Study finds sleep in male full-time workers falls as income rises, with one cause being other leisure activities substituting for sleep. It makes sense that sleep doesn’t cost money while other things often do, but the marginal cost of much leisure is very low. I don’t buy this as the cause. Perhaps reverse causation, those who need or prefer less sleep earn more money?
The productivity statistics continue to be awful, contra Alex Tabarrok part of this recent -3.88% Q1 print is presumably imports anticipating tariffs driving down measured GDP and thus productivity. The more I wonder what’s wrong with the productivity statistics the more I think they’re just a terrible measure of productivity?
A model of America’s electoral system that primary voters don’t know much about candidate positions, they know even less than general election voters about this, so they mostly depend on endorsements, which are often acquired by adopting crazy positions on the relevant questions for each endorsement, resulting in extreme candidates that the primary voters wouldn’t even want if they understood.
It’s not really news, is it?
Rules of Confidentiality
Alex Thompson says “If you don’t tell the truth, off the record no longer applies,” proceeding to share an off-the-record unequivocal denial of a fact that was later confirmed.
I think anything short of 100% (minus epsilon) confidence that someone indeed intentionally flat out lied to your face in order to fool you in a way that actively hurt you should be insufficient to break default off-the-record. If things did get to that level? If all of that applies, and you need to do it to fix the problem, then okay I get it.
However, you are welcome to make whatever deals you like, so if your off-the-record is conditional on statements being true, or in good faith, or what not, that’s fine so long as your counterparties are aware of this.
Something Is Wrong On The Internet
Scott Alexander asserts ‘If It’s Worth Your Time To Lie, It’s Worth My Time To Correct It’ and I want to strongly claim that no, this is usually not true for outright lies and it definitely usually isn’t true for misleading presentations of facts that one could nitpick, and often doing so only falls into various traps, although it’s not clear Scott ultimately disagrees with that objection, he walks this back a bunch at the end. I do agree with his caveat that the actual important principle is that, if someone does decide to offer the correction, you don’t get to say they’re ‘supporting’ a side by doing so, or call them ‘cringe’ or describe it as ‘well acktually’ or anything like that.
What is the actual solution? Recalibration. As in, you pick up on the patterns, and adjust accordingly based on which sources pull such tricks and which ones are various amounts of careful not to do so. And yes, this does involve some amount of pointing it out.
Scott follows this up with the contrast of “But” versus “Yes, But.” As in, if [A] points out [B] while arguing for [X] was wrong about something that was load bearing to their argument, [B] needs to acknowledge they were wrong (the ‘yes’) before pivoting to other arguments for [X].
I always find these ‘definitely the world will look almost exactly the same’ claims to be hilarious, given how that wouldn’t be true even without a singularity and hasn’t been true historically for a long time, but that’s beside the point here.
I mean, not actually a cool thought either way. These thoughts are absurdly and utterly wrong. I for one want to say that while various sci-fi things would be nice, life right now (at least for me) is righteously awesome with only two real problems: Existential risk and other tail risks for highly capable future artificial intelligences, and our failure so far to cure human aging, because I hate getting old and I hate dying even more. That’s it.
But the ‘yes’ first would at least help, especially if you want continued engagement.
It is of course fine to say ‘I believe this because of [ABCD…Z], and any one of those would be sufficient, so even if you are right that [A] is false that doesn’t matter.’ But you have to actually say that, and also if I tear through [ABCD] in order you should be suspicious that this might correlate with the rest of your list.
Government Working
I am doing my best to avoid commenting on politics. As usual my lack of comment on other fronts should not be taken to mean I lack strong opinions on them. Yet sometimes, things reach a point where I cannot fail to point them out.
If you are looking to avoid such things, I have split out this section, so you can skip it.
The Federal Reserve is cutting its workforce by 10% to be a ‘responsible steward of public resources.’ This is not a place I would be skimping on head count. There are so many ways for a marginally better Fed to make us a lot more money than it costs.
A fun question, who has the better business climate, California or North Carolina? It depends what kind of business. If you’re trying to build a car or open a sandwich shop, your job is way easier in North Carolina. But for some purposes, in particular tech, California’s refusal to enforce non-competes plausibly trumps everything else.
I knew the UK was arresting people for online posts, including private text messages, including ones that were very obviously harmless. I didn’t realize it was 1,000 people a month. The UK has a little over 40 million adults, so your risk every year is about 3bps (0.03%), over 1% chance that at some point this happens to you personally. That’s completely insane.
Field notes on Trump’s executive orders on nuclear power. It all seems neutral or better, but it’s not clear how much of it is new and actually meaningful. My guess is this on its own doesn’t do that much, and is less important than retaining or expanding subsidies.
Is US immigration still, for those who know the way, open for business?
Jones Act Watch
“Trump’s shipbuilding agenda is sinking.” The Navy is described as 20 years behind in its goals. We need to face facts, we don’t have meaningful shipbuilding, and perhaps we want to pay massive amounts to change that while using proper tactics like export discipline, but the Jones Act and similar laws haven’t ‘protected’ American shipbuilding, they’ve destroyed it, and they need to go.
Signatures as Solemnization
The point of a signature is usually not to prove that it was you who signed. Here’s a fun thread pointing out what’s really going on, that it’s mostly a tripwire that says ‘we are no longer Just Talking’ except in situations where you are for-real signing a for-real contract.
Antisocial Media
Twitter introduces Polymarket as an official prediction market partner. Even the mid version is pretty cool, and executed properly this could be amazing, an absolute game changer. For now this only looks like Polymarket incorporating Twitter. I love that for Polymarket but the real value is the other way around. We need to incorporate Polymarket into Twitter, at minimum letting Tweets embed markets, and ideally complete community notes with markets attached, spin a market out of any tweet, trade right on Twitter, and so on.
Removing political ads from Facebook and Instagram for six weeks before the 2020 American elections for a given user had no detectable effect on their political knowledge, polarization, election legitimacy, campaign contributions, candidate favorability or turnout. I basically buy it in this context. Very little changed in that six week period. It makes sense that in the end stages of a campaign, all the information flow here is already oversaturated. You’ve already seen the ads, and those ads are still flowing to everyone else you know and back to you, if you’re using Facebook and Instagram, and the rest of your world is also unchanged.
The most interesting aspect is that the most common type of ad was a fundraiser, they kept running them, and yet they didn’t have a noticeable impact on fundraising. This is strongly suggesting that the ads on the margin shift distribution, avenue or timing of contributions but don’t impact overall giving, at least once you get to this stage of the campaign. Alternatively, such fundraising had reached such over-the-top obnoxious levels that they were driving people away as much as they raised money. That could also explain there being no net impacts in other ways, that saturation especially of fundraising but also other ads served to piss voters off about as often as they helped.
Also I think this was a fantastic experiment and we should do More Like This, although perhaps Meta is not so excited to let them do it again for obvious reasons.
BlueSky appeals to creators over Twitter by not downranking links, drawing many of them into joining. It really is this simple, and it’s such Obvious Nonsense to think Elon Musk knows what he is doing trying to ‘keep people on the website.’
New Twitter bot plot twist, bots that reply to you then block the author, which gets the author deboosted, en masse, and the impact adds up. Neat trick, it’s new and the network of bots doing this is reported to be massive, in the thousands or more. Or perhaps the deboosting is incidental to the standard goal?
This seems like a clear Skill Issue by Twitter. The algorithm should be able to figure this one out, also in this case they keep tagging the same account which gives the game away but there should already be strong statistical evidence that the activity isn’t real.
A special 0.5% of Facebook users were kept ad-free since 2013, and their ‘give-up-Facebook’ price is the same as ad-exposed users (~$32/month).
This suggests the ads are efficient, or even that more ads would be better, since they have marginal value to Facebook and the users don’t actually mind. Alternatively, it means people’s posts aren’t better than ads.
The thing about TikTok is indeed that it’s… not meant to be interesting? That’s actually a category error?
If you have a Twitter variation with too sharp a cultural focus, it is extremely hard to break out of that focus and become a general solution.
This usage pattern doesn’t have to doom you. It makes sense that you’d get traffic largely driven by major events and moments where new users show up who might not be a fit so many leave, and that without them you’d be static or in decline, and this is compatible with long term growth. Nothing wrong with losing small most days and winning big on occasion if the wins are big enough. In this case however it looks pretty doomed, as it’s hard to think of additional similar events.
A wonderful illustration of toxoplasma of rage as an active social media strategy. It’s great when people admit this is the plan.
If you’re dealing with Twitter bots, what’s the play?
This depends on what you care about. If you’re only thinking of your own experience, there is not that much to do beyond blocking the bot, short of getting block lists.
If you’re trying to promote the general welfare and fight on Team Humans, you can do more. Adam Cochran suggests first find the account the bot is shilling (if any), mute and then block it, to inflict maximum algorithmic pain. Then block the commentator. That sounds like more work than I’d be willing to do, but I notice that if I had one a quick way to tell an AI to do it, I’d be down.
Technology Advances
The correlations here make sense, but a lot of potential explanations are not causal even before one examines the research methodologies involved:
It’s definitely great if the early versions of digital tech are associated with reduced risk of cognitive decline, although I would not presume this result carries over to long term exposure to modern smartphones and the associated typical use cases. And even if it does still apply, this is one of many aspects of digital technology, and it very much does not allow us to ‘think on the margin’ about this, as the most likely causal explanations involve enabling people to meet minimal thresholds of cognitive activity that serve to slow down cognitive decline.
As per this righteous rant thread from Conrad Bastable, it is very true that when you buy a new computer, by default it will come pre-loaded with a lot of crap, and Windows 11 will waste a bunch of resources, and that Macs come with a lot of objectively terrible software, and that it takes a remarkable amount of deliberate effort to fix all this even if you mostly know what you are doing, during which you are subjected to ads.
This isn’t a big deal for a knowledgeable user. No, the startup button in Windows spiking the CPU isn’t a practical problem, and it is a one-time not that high cost to get a lot of the stupid bloat out of the system, and letting apps send you notifications is good, actually, given there’s an easy way to turn it off for each app the moment they try to abuse it. I have made peace with the setup process.
What I don’t fully understand is why all of this is true. It’s one of those ‘yes I get it but also I don’t, actually.’ It seems like an obviously easy selling point to say ‘this PC doesn’t come with any ads’ and similar. People would definitely pay a bit more for that. You’d get customer loyalty and word of mouth, as your product would perform better. An expensive purchase is exactly a place where you shouldn’t need to do ads. And yet, yes, they consistently do this anyway, even when it is obviously net destructive to long term profits. Why is my smart TV so terrible, even if it’s ultimately fine and way better than an old dumb one? Yes, I tried using an XBox or Playstation instead, ultimately I found it not better enough to bother.
But even if you hate all that, what are people going to do, use Linux? Seriously?
I mean, okay, one out of six with the right link, I bet it’s more if you knew where to look, and you’re going to suggest this to people who can’t debloat their Windows box? Every now and then someone pitches me on switching to Linux, and every time it seems like an endless nightmare to get up to speed and deal with compatibility and adoption issues, even if I didn’t care about games.
Ah, the circle of online life:
As your account grows in reach, the value of blocking any one account that is interacting with you goes down, so you bother doing it less, and eventually have to rely on mass blocks, block lists and automated tools. That makes sense. I’m still in full block people mode, although ‘a little unkind’ is not enough to trigger it for me. It used to mostly be mute but it’s now block since that doesn’t stop you from seeing my posts either way.
For Science!
Steve Hsu reports lots of optimism for both artifical wombs and gene editing of embryos, and that a meeting at Lighthaven convinced him they’re coming sooner than you would think. I don’t have the optimism he has about this as a hedge against existential risk given the timelines involved, but there is wide uncertainty about that, it’s very much a ‘free’ action in that regard, it’s all upside.
Variously Effective Altruism
A reasonable perspective on both Rationality and also The Culture, except that with this perspective redemption is possible only through grace, and that kind of thinking has a long history of really messing people up if they don’t feel the grace enough:
There definitely is an important pattern that bright abstract thinkers do many things, including Effective Altruism, that seem often motivated by striving to be and think of oneself as Good rather than Bad, with more discussion at the thread. We have a culture telling people (especially men but also everyone) that they and key parts of themselves and our entire civilization (and often even humans period) are Bad – we de facto now have the Christian idea of original sin without the Christian standard path to redemption, oh no. The historically prototypical ‘good deeds’ and ways to make yourself Good are, to bright abstract thinkers, either obviously fake or transparently inefficient and silly, or at least not sufficient to do the trick.
Checkmate.
You shouldn’t not use common sense, but the word ‘just’ is not okay here.
There was a recent exchange of posts between Tyler Cowen and Scott Alexander about USAID. I wrote an extended analysis of that exchange and learned by doing so, but am invoking Virtue of Silence and not posting it. I will simply note that everyone involved agrees that it is completely false to say that anything remotely like only 12% of USAID money ultimately went to helping recipients, and that anyone using that claim in a debate (such as Marco Rubio or JD Vance) is doing a no-good very bad thing.
Lab Grown Meat Ban Redux
Last year Florida banned lab-grown meat, and I went through three rounds of trying to explain why, even though I wouldn’t ban it, a reasonable person who wanted to continue eating regular meat might want such a ban, given how many people are itching to ban or ostracize or otherwise destroy all other meat consumption.
I am the first to admit I am not always perfect about adhering to the principle of ‘if you care about an issue you need to understand where all sides and especially the opposition are coming from’ but I do try and I think it is important to do so.
I am quoting people on this again because there was another round of this argument recently, and because it is a microcosm of a lot of other things going on in politics, even more so than when this came up last year, culminating in Matt Yglesias presenting this as a fully one sided issue and saying ‘Banning Lab Grown Meat Is Stupid.’
It’s also worth noting that a number of the largest multinational meat packers are acquiring or investing in lab meat start-ups, to hedge their bets.
Again, I don’t want to ban lab grown meat, because I am willing to stick to my guns that you don’t go around banning things, even if you don’t like where all of this is going. But I understand, including for those who don’t have any financial skin in the game. I disagree, but it isn’t dumb.
If you don’t understand why, think about it until you do.
It is flat out gaslighting to call this, as Derek Thompson does, ‘a make-believe fear’ or ‘psychological issue’ rather than a very real concern, or to deny that this is the stated intention of the entire lab grown meat industry and a large number of others as well, and it is gaslighting to deny the clear pattern of similar other bans and requirements that have made people’s lived experiences directly worse – whether or not you think those other bans and requirements were justified.
There is less than zero credibility for the claim that no one will be coming for your meat down the line, no one is even pretending that they’re not coming for it, at best they are pretending to pretend.
A claim I have heard is that 50% or more of self-identified ‘vegetarians’ ate meat in the past week. When you ask my followers both questions at once, they don’t agree and it is only 17%. But obviously asking both at once and asking my Twitter followers are both going to lower the percentage here.
Also 22% of respondents were vegetarian or vegan.
Opportunity Knocks
Arc was (as of May 22) hiring a Chief Scientific Officer. Seems like a great opportunity.
Good Books
Ryan Peterson asks what is the best history book you’ve read?
Lots of other people replied as well, seems overall like great picks but there are so many so I haven’t rad most of them. If I had to pick one, I’d pick Thucydides, History of the Peloponnesian War.
Commitment
Cate Hall warns against the ‘my emotions really mean it’ approach to commitment, because it doesn’t work. Your emotions will change. What you have to do is engineer things, especially using forcing functions, such that success is easier than failure. For her the most prominent of these was diving into her relationship and marrying quickly. The whole system requires real accountability and real consequences. It has to actually hurt if you don’t do it, but not hurt so much you’ll back out entirely. You also need to watch out before doing too much to cementing a life or lifestyle you don’t want.
The method Cate Hall describes here is highly useful, but far from the only component, and by its very nature this strategy involves real risk of costs big enough to hurt and without corresponding upside. As Cate would no doubt agree, a bet (where you can win or lose) is better expected value than a Beeminder (where there is only downside, except for the motivation).
Ideally you want other commitment mechanisms that involve upside. My central move is, quite literally, to realize that when one commits one is betting not only one’s external reputation but also one’s inner reputation, one’s thinking of oneself as someone who keeps similar commitments and thus the power of similar commitments to bind your actions.
Being able to use commitments effectively is super valuable, so endangering that by breaking commitments is a big cost, which enforces the commitment on its own. And every time you hold strong, the power grows. And this dynamic forces you to calibrate, and only use commitments when you mean them.
How To Pass On Generational Wealth
If your children are set to inherent vast amounts of wealth, there are three schools of thought on how to handle this:
I am with Sebastian and Mason that plan #2, forcing your kid to live like a ‘normal person’ and obsess over small amounts of money so they can be like everyone else or learn to ‘value the dollar’ or what not, is actually terrible training for being wealthy, for the same reason lottery winners mostly blow their winnings. Choose #1 or #3.
You do want them to ‘earn it’ and more importantly learn how to operate in the world, it’s totally fine to attach some fixed conditions to the unlock of parts of the trust fund, but you don’t do that by giving them Normal People Money Problems.
I do think this is different from putting them in charge of your own business straight away. It makes sense to make them train via other work, but that doesn’t mean you want to squeeze them for money while they do that. You want to design an actual curriculum that makes sense and gets them the right experience, in the real world.
I see a big difference between ‘you have to hold down a real job’ versus ‘you have to focus on personal cash flow.’
Rain On Your Wedding Day
Wedding costs seem completely nuts to me. Also, buffet is better? But I love that the objection here to that people have to get up, not to the RSVP-ordering rule.
In general, yes, a good restaurant is better than a buffet. But a buffet will be better than most catering that you’ll get at a wedding or other similar venue.
The deal on the catering is universally awful compared to what you get, even without considering that many people do not want it, and often quality is extremely bad. And by going buffet you get the advantages of a buffet, everyone gets what they actually want, or at least don’t mind. So yes, this means you have to get up to get your food, but so totally worth it, and I don’t even see that as a disadvantage. You can use an excuse to flee the table sometimes.
Social Versus Transactional Exchanges
A series of branching exchanges on the ways to draw distinctions between interactions that are ‘transactional’ versus ‘social.’
I would go a step beyond Rob here, there are a lot of related but distinct modes here, and asking for symbolic compensation does not make something transactional, even if that symbol costs some money. The point of buying dinner is not that it’s a fair price, it’s that it shows your appreciation. Indeed, this can serve to avoid having to write an entry, or a much larger entry, into that social ledger.
Then there’s the question of smaller things like dinner parties.
I consider it fine to show up empty handed, but better to ask if you can bring anything. We have one friend who explicitly bars anyone from bringing anything, because he wants to curate the whole experience. I think that’s great.
My family try to host Shabbat dinners on Friday evenings for friends (if I know you and you’d like to come some time, hit me up), and we explicitly say that we don’t expect you to bring anything, although you are welcome to do so, especially dessert, or wine if you want to be drinking but our crowd usually doesn’t drink. It is nice when people do this. But also I think at this level, a thank you works fine.
While I Cannot Condone This
Ah, the joys of someone with authority demanding documentation to prove that you do not need documentation. And the dilemma of what to do in response, do you produce it (if you have it) or do you become the joker? Here the example is, woman who is 24 weeks pregnant being told to prove she isn’t 28 weeks pregnant, because if she was she would need a doctor’s note to fly.
A fun proposal is to reduce sweets consumption by having sweet food with probability 2/3rds, determined by random draw. My problem with this proposal is that, when something isn’t always available, it makes you much more likely to take that opportunity when it is available, you don’t want to miss out on your opportunity. Thus, you need a variant: If you get a yes, you keep that yes until you use it, so you get rewarded rather than punished if you skip your chance.
Benjamin Hoffman offers a model of the kind of personas, performances and deceptions we expect from politicians and the managerial class.
I think those people mostly don’t even see it as deception or lying anymore. It’s simply the move you make in that situation. And I can actually model that perspective pretty well, because it’s exactly how I feel when I’m playing poker or Diplomacy.
I see Jackall as describing the attractor, and ‘how high you bid’ in moving towards that attractor, which is mostly about signaling how high you will continue to bid later, is a large determinant of ‘success’ within context. Willingness to engage a facade is sufficient to do okay, but you’ll probably lose to those who never drop the facade, and especially those who become the mask.
There is now sometimes a huge discount for buying airline tickets in bulk? In one extreme example it was only $1 more to buy two tickets rather than one. This is happening on American, United and Delta, so far confined to a few one way flights.
The growing Chinese trend of paying a fake office so you can pretend to work. Isn’t this at its base level just renting coworking space? The prices seem highly reasonable. For 30-50 yuan ($4-$7) a day, or 400 yean you get a desk, Wi-Fi, coffee and lunch. For more money you can get fake reviews, or fake employees, or fake tasks, and so on. Fun.
The base reported use case is to do this while looking for work, but also you could use it to, you know, actually work? There’s nothing actually stopping you, you have Wi-Fi.
This month I learned the Air Force had an extensive ritual where they told high ranking officers they were secretly reverse engineering alien aircraft purely as a way to mess with their heads, and this likely had a role in the whole UFO mythology thing. Of course, some who have bought in will respond like this:
You see, the evidence that this was faked only shows how deep the conspiracy goes.
Costly Signaling Is a Weird Name
It’s tricky to get this right and avoid anyone being misled, because you’re combining a number of related strategies and concepts into the group that is contrasted with, essentially, ‘cheap talk,’ and this risks conflating multiple concepts. But I think ‘costly signal’ is still our best option here most but not all of the time, as a baseline term, because simplicity matters a lot.
I get why one would think that ‘costly in utility to fake’ is the core concept that matters, but a bunch of other differences seem important too.
This matters because the goal is to have signals that maximize bang-for-the-buck in all these senses. I worry that calling it ‘costly-to-fake’ while accurate would (in addition to being longer) centrally point people towards asking the wrong follow-up questions.
Travel
“Travel” doesn’t have to be fake. Tyler Cowen travels. Eigenrobot goes a little too far here, on many fronts. But most of us at best Travel
, which can involve seeing some cool sights but mostly is something one does to have done it, except when we are meeting up with particular people or attending an event, which is a valid third thing.
Again, are some real advantages to Travel
, especially if your home base is in a place without all the things, you shouldn’t quite do zero of it. But most Travel
is, I think, is either a skill issue due to an inability to relax without it, or a parasocial or future memory and anticipation play.
Plus you have to plan it. Some people enjoy this, and they are space aliens.
Good News, Everyone
Lighthaven now has a podcast studio. I had the chance to tape in there with Patrick McKenzie, it is a quite nice podcast studio.
The IRS tax filing software that everyone except TurboTax loves, and that they got the Trump administration to attempt to kill, has instead gone open source, with its creators leaving government to continue working on it. Something tells me plenty of people will be happy to fund this. So maybe this was a win after all. I am sad that my taxes involve too many quirky details to use anything like this.
Kalshi sports betting is by all reports going well.
This is very much The Big Game, so these numbers are not as impressive as they look. But it’s still respectable, and most importantly the pricing is good and it is fully legal. I consider products like DraftKings and FanDuel effectively abominations at this point, but if you don’t want to use overseas books this seems fine:
Here’s another ‘why America’s implementation of sports betting is terrible’ essay. We not only allow but essentially mandate and mainstreamed the most predatory version possible, and the states are not realizing their promised revenues.
There exist contact lenses that only need to be changed every few weeks, and the report is they feel like normal contacts.
Cate Hall explains why agency is both a trainable skill and also Definitely a Thing distinct from ‘success,’ and that if you don’t want your future successes to be in air quotes, you should go build that skill.
Cate Hall also points out that while you may be high agency in some ways, that does not mean you are automatically high agency in other ways. In many ways you likey are not actually trying. In particular, you’re likely stuck on whatever level of resourcefulness you had when you first encountered a given problem, which I’d raise to you likely being stuck with the particular strategies you found. You don’t step back and treat problems in other spheres the way you do in your areas of focus, even if you are exerting a lot of effort it is often not well-aimed or considered.
Denmark repeals its ban on nuclear energy.
The source tweet from George here plays it up too much but there is as one would expect a substantial (r~0.4) correlation between honesty and humility. From what I can tell all the ‘good’ personality traits correlate and so do all the bad ones. That’s very helpful in many ways, including getting a read on people.
Chinese attempt to recruit and blackmail fed official John Rogers, largely through the Chinese wife he found on a matchmaking service, likely thinking Rogers had far more access to actionable information than he did. As Tyler Cowen says, there probably wasn’t much to learn from him. It seems the Chinese pushed the threats too hard, and he reported their attempts to the Fed.
Lie detection is something we’re actually pretty good at if we pay attention.
Put On a Happy Face
This is true if and only if the cheerfulness doesn’t interfere with something load bearing, which can be true in unexpected ways sometimes, hence missing mood:
There are two levels to missing moods.
The first level, where everyone here is in violent agreement, is not to celebrate the need to pay costs, and not to confuse costs with benefits.
The second level, which can play back into confusions on the first level, is that you need the missing moods because that is how humans track such things, and how we generate reward signals to fine-tune our brains. You should generate such signals deliberately and ensure that they train your brain in ways that you prefer.
Jane Street was excellent about this. They emphasized the situations in which you ‘should be sad’ about something, versus happy, exactly enough to remind your brain that something was a negative and it we should evaluate and update accordingly. This is part of deliberate practice at all levels, in all things. You need to pay that experiential price, to some extent, to get the results.
Whereas there are other situations in which doing something cheerfully is the action that is load bearing. It is always a little bit load bearing in that you would prefer to be cheerful rather than not, and those around you usually also prefer this. Some actions only work if you are cheerful while doing them or otherwise have the right attitude, or at least they work far better. Those are almost always either worth doing cheerfully, or not worth doing at all.
The trick then is that this conflicts with a well-calibrated brain that systematizes deliberate practice. You need to be able to shut off those functions, in whole or in part, for the moment, while retaining them for other purposes. That’s tricky to do, and a tricky balance to get right even once you have the power to do it.
Egg Prices Keep Going Up
If you’re a young woman at a prestigious university there are those who will pay top dollar, tends of thousands, for your eggs. One comment says they offered $200k. Thanks, people warning about this ‘predatory advertising’ of (for some, it’s an offer, you can turn it down!) the ultimate win-win-win trade. There’s a real health cost, but in every other way I say you’re doing a very obviously good thing, and if you disagree you can just pass on this.
Cultural Stagnation
Noah Smith claims American culture has stagnated.
He starts off citing the standard evidence like movies being dominated by sequels and music being dominated by older songs and the top Broadway shows being revivals and old brands (e.g. DC and Marvel) dominating comics. He also cites the standard counterexample that TV has very clearly experienced a golden age – I strongly agree and observe that with notably rare exceptions older shows now appear remarkably bad even when you get to skip commercials.
I’d also defend Broadway and music consumption mostly being old. Why shouldn’t they be? A time-honored Broadway play is proven to be good, it makes sense to do a lot of exploitation along with the exploration, and we do find new big hits like The Book of Mormon and Hamilton. In music it is even more obviously correct to do mostly exploitation, or exploration of the past where selection has done its job over time, in at-home experiences. In addition, when you play the classics, you get the benefits of a rich tradition, that adds to the experience and builds a common culture and language. I say that’s good, actually. Is it ‘cultural stagnation’ when we read old books? Or is that what leads to culture?
And also there are completely new, rapidly evolving forms of culture, that flat out didn’t exist before, as he notes Katherine Dee arguing. A lot of it is slop, but that’s always been true everywhere, and they are largely what you make of them. Even in a 90%+ (or even 99%+!) slop world, search and filters can work wonders.
Noah considers response to technology, and also has a theory of possibility exhaustion, that the low-hanging fruit has been picked. There are only so many worthy chords, so many good plots. He thinks this is a lot of why movies now are so repetitive, the space of movies is too small.
Noah and I basically agree on what’s going on in another way. Movie people used to largely make movies for each other. The sequels were (I think) always what audiences actually wanted, and the creatives basically refused to deliver and now they’ve stopped refusing. But there’s still plenty of room for the avant-garde and making a ‘good’ ‘original’ movie.
Also, we’re seeing a lot of cultural change start to happen because of AI, which will only accelerate. I’m definitely not worried about ‘cultural stagnation’ going forward.
Tyler Cowen responds that he is inclined to blame what stagnation we do see on lack of audience taste today. I would reframe most of that complaint as saying that before we used to get to overrule or dictate audience taste far more than we do now, with a side of modern audiences having very little patience, which I mostly think they are right about.
For Your Entertainment
Scott Sumner rattles off some overlooked films, predictably I have seen very few of them. Those that I did see seem like at least good picks but not exceptional picks, and of course my choice to see those in particular wasn’t random. There’s a Letterboxd version.
What type of music you like predicts your big 5 personality traits, some details surprised me but they all made sense after I thought about them for a few seconds.
Are the music recommendation engines the problem, or is this an unreasonable ask? I think the answer is both, we are trying to solve the wrong problems using the wrong methods using a wrong model of the world and all our mistakes are fail to cancel out.
There’s a lot of room to do better for power users, but Spotify has to work with users who, like users everywhere, are mostly maximally lazy and who don’t want to invest. I’m actually not so sure there’s much room to improve there, although I use Amazon Music rather than Spotify so I haven’t put it to the test.
Benjamin Hoffman offers a thread on the 1990s and why people are so nostalgic for them. Here’s part of it.
I notice that those are the ones that I loved and remember and seem important, but only 2 of those 4 are in the top 10 grossing movies of the year.
The listed category also would include Magnolia, and arguably also in a way the highest grossing movies of the year if you take them properly seriously, which were Star Wars Episode I about the fall of a republic that had lost its virtue in peacetime and how we ultimately turn to the dark side, The Sixth Sense about (spoiler but come on) and Toy Story 2 about being trapped by a collector and preserved in a box. Then #4 is The Matrix, and #5 is Tarzan, which glamorizes being outside of society, and #7 is Notting Hill which is basically about rejecting modernity for an old bookshop. Huh.
I still think of 1999 as a high water mark in movies. Those were good times.
(As a control I randomly chose 2005 and didn’t see the same pattern, also most of the top films were now some sort of remake. Then I went to 2012 and we’re in a wasteland of franchises that have nothing to say.)
A comment on my call for the Meta-Subscription, saying it exists and it’s YouTube:
I too am a YouTube Premium subscriber, and no I did not realize that my viewing was subsidizing creators more than I would have without the subscription. I only knew not having ads was worth a lot. And yes, YouTube has quite a lot in it, but no it is not the Mega-Subscription even for video. Too many others aren’t playing ball.
A rare time when explaining the joke actually is pretty funny.
ESPN finally launches a streaming service. Ben Thompson correctly says ‘finally’ but also notes that everyone is worse off now that sports along with everything else can only be watched intentionally through a weird mix of streaming services. The problem with ESPN as a streaming service is that it won’t actually give you what you want. If you offered me the Sports Streaming Service (SSS), and it had All The Sports, I’d be down for paying a decently large amount. Give me a clean way to subscribe even just to the teams I want, and maybe that works until the playoffs. But if I’m going to be in I want to be in. ESPN alone doesn’t get me in and I don’t want to spend the time figuring out where everything is. So it’s still going to be YouTubeTV for football season and that’s it, I suppose.
Gamers Gonna Game Game Game Game Game
Alas, there’s been no time for me to game this month. I hope to get back to it soon.
I am on the fence about getting back to Blue Price. I do think it’s pretty great in many ways, but not sure it’s my speed and I worry about being puzzle-locked or waiting for a lucky day?
Next up: Clair Obscur: Expedition 33, which everyone loves but I haven’t played yet past the first campfire, and Monster Train 2 which is a ticket I can cash at any time.
And of course Slay the Spire 2, once it arrives.
So instead we have to live vicariously through Magic: The Gathering news.
We now have Magic: the Gathering Standard set built around Final Fantasy? Sign me up! Except actually, no, don’t, and not only because look at the time.
The same way that the Dungeons & Dragons set ‘got’ D&D, the Final Fantasy set mostly doesn’t ‘get’ Final Fantasy, or its particular concepts and characters. There are some clear hits for me, especially Cecil, Dark Knight. But there’s also a ton of ‘look at what they did to my boy.’ Or my girl, Aerith, the vibes are almost reversed. You think that’s Tifa? You think that’s what happens when Kefka transforms? Cloud triggers equipment twice rather than being able to lift his sword at all? The Crystal’s Chosen happens on turn seven? What is even going on with Sin?
I could go on.
Mechanically the set feels schitzo. Which is kind of fine in principle, since Final Fantasy is kind of schitzo and I love it to pieces (in order especially 6, 4 and 7, but I’ve also played through and endorse 12, 1, 10, 8, 3 and 2) you’re doing the entire franchise, but then the resonance mostly feels backwards. We object because we care.
(I don’t know why I bounced off 9, whereas 5 felt like an uber grind and my first attempt got into a nightmare spot in the endgame, must have been doing it wrong. X-2’s opening was insane in the best way but it kind of petered out. I endorse that 13 and 15 weren’t doing the thing anymore, sad, and haven’t tried 16. The MMORPGs don’t count, I tried 14 for a few hours and it felt highly mid.)
I don’t follow Magic right now so I have no direct data on if Prowess is too good in Standard, but Sam Black seems clearly right that if you hold a bunch of major Arena tournaments and no one can beat the best deck there, then that means the best deck is too good, what else were you even hoping for? Arena regular play, including at high Mythic, systematically is both much easier than big tournaments for other reasons and also includes a lot more people experimenting or trying to beat the best deck, and less people playing the best deck.
With distance from the game, I think Magic in general has been way way too reluctant to drop ban hammers. Yes, it’s annoying to strike down people’s cards and decks, but it’s far worse for your game to suck for a while, and for everyone to feel forced to fall in line. In the Arena era, things go faster, no one is holding big secrets back for long, and you find out fast if you have a problem.
The Commander Cube seems to have some issues with game length and timing out, due to the whole point being to create long games.
Commander keeps eating Magic, Sam Black is going to stop doing Drafting Archetypes to focus on cEDH. It makes sense for him to focus on what he enjoys, but this confirms to me that the Magic I loved is essentially dead, and unless and until my kids get interested in playing I need to fully move on.
I Was Promised Flying Self-Driving Cars
Waymo gets regulators to approve roughly a doubling of the area they can operate in around San Francisco, from the gray to now the orange, although for now they are limited by the number of cars so they are expanding gradually.
East Bay when? Alas, it seems they are still not allowed on highways so it wouldn’t link up, which means it wouldn’t make sense to expand there.
Waymos remain severely undersupplied even in the service area of San Francisco.
Arvind Narayanan points out via Matthew Yglesias that we lose a lot of the upside of self-driving cars if we don’t adjust our parking rules, because we’d miss out on much more efficient use of space. And yes, we should try to get that value too, but mostly we just need to actually allow the self-driving cars and that’s where most of the value lies? Arvind then generalizes this to diffusion begin the bottleneck to progress rather than innovation. To me this is a great example of why that is obviously false, the main barrier to self-driving cars was expected to be regulatory but it has decisively proven to mostly be the ability of the AI to drive the car.
If all the AI can do is drive the car, we will quickly replace the taxis and the cars with versions that self-drive (or at least versions capable of it), and then of course we will fix the parking rules.
It’s probably going to be a common sight going forward that protestors attack Waymos, and this was a good time for this reup:
How far ‘behind’ are some self-driving companies? How is Tesla doing?
This brings up something we often deal with in AI. Fast following is substantially easier. In some sense, Tesla is 10 years behind Waymo by this metric. In others, it’s less, because Tesla should be able to get through those 10 years faster given what Waymo already did. So how ‘far behind’ are they is a question that depends on which sense you care about, and whether you care about Tesla’s ability to take a lead.
The Lighter Side
I know the context (and lore) here but it’s arguably better without it.
I have no questions at this time.