We have learned so much less than nothing from Covid. We’re actively stupider.
It’s 2026, and here we are again, lying about the virus because we are worried that people exposed to it or from the wrong place might face stigma otherwise.
Envidreamz: A local news story highlights passengers on the hantavirus stricken MV Hondius cruise ship who are more worried about facing stigma and rejection back home than the virus itself.
The piece quotes passengers fearing social media backlash, portrays their onboard life as calm with masks and birdwatching, and leans heavily on WHO experts downplaying human to human transmission while stressing it’s nothing like Covid.
Is anyone else getting these local news articles trying to sympathize with the passengers on the ship because they fear the stigma that nobody wants to be around them when they go home?
And then further minimizing this virus saying it’s not easily transmissible human to human and basically no big deal?
Wtf? None of this is even true.
Meanwhile, the other side is already also out in force and stupidity, to say in advance that no one is going to trust anything anyone says, or follow any orders, and if they say anything is happening it is fake, that if it does happen it will be because of the Covid vaccine, and I’ve even seen a literal claim that one should take Ivermectin.
The CDC is nowhere to be found. The WHO lacks any authority to get anyone to quarantine, and then there are those who treat this as a valid excuse.
What matters most is that we have now learned that Earth, in 2026, is completely incapable of taking steps to prevent even a highly preventable obvious pandemic.
Caroline Orr Bueno, Ph.D: Omg even at the National Quarantine Unit where the hantavirus contacts are, they’re still only using surgical masks when people in quarantine interact with the medical staff. If one of them gets infected, you’ll know why.
Why not use an N95 out of an abundance of caution? What possible reason is there for not taking that tiny step just to ensure they’re protected?
spor: ultimately this whole thing will probably blow over and we’ll be fine, but man, it almost feels like they are actively trying to cause an outbreak.
An abundance of caution is not always good. But when costs are trivial, for things like ‘wear at least N95s on the literal quarantine unit’? At least do that, you fools.
The good news is I believe, and Peter Wildeford agrees, that the chance of getting an actual Hantavirus pandemic are quite small. If I was trading the prediction markets, I would sell. Hantavirus is almost certainly not infectious enough in its current form to cause a pandemic. Even with our complete lack of reasonable precautions, R0 will probably be less than one.
The bad news is that this is entirely good luck. If hantavirus was capable in its current form of causing a pandemic, you know what we would be facing down? A pandemic.
It cannot be overstated how determined we are not to take any actions that would actually prevent a potential pandemic. That could soon include something engineered via AI, and again I expect us to react maximally stupidly if that happens.
I realize we are incapable of keeping the AI in a box, but you would think we would be able to keep people on a ship. Instead, even after we know what is happening, we send them home on flights, and hope for the best, while pretending it is all going to be fine.
Even if there is little risk of a full pandemic, even the amount of attention this has already gotten, and the amount of distraction and stress it has already caused, has exceeded the costs of handling the situation properly. If even a handful of cases get out, panic could ensue along much larger areas, again even if we are correct that the risk of a pandemic is minimal.
We badly need to repeal the Davis-Bacon Act, which requires ‘prevailing wages’ on federal construction contracts. On top of directly raising costs it creates giant tracking headaches and can lead to ‘retroactive pay.’
A large percentage of costs is compliance, which means this is far worse than a typical minimum wage that at least is easy to understand and comply with. Another simple argument against this is that if it truly was the prevailing wage, you wouldn’t be able to get away with not paying it. If you did, it wasn’t so prevailing, was it?
The blue versus red button experiment discourse is like a social disease.
It’s embarrassing.
It’s recurring – it goes away, it comes back, it goes away, it comes back.
If you don’t engage with it, you can’t be infected.
(With apologies to Marion Barry and also the former Capital Steps, IYKYK.)
If you catch a giant bug or inefficiency worth eight or nine figures, you might get a percent of the profits, or you might get a pat on the back. Incentives do not seem ideal. Seems important to know which kind of employer you are working for.
Harnoor Singh: Engineer prevents $80-90M recall. credited as a “good catch” lol
CFO mentions the release on the earnings call six months later.
The problem isn’t that companies are ungrateful. It’s that there’s no mechanism to reward the person at the start of the value chain.
Senior engineers: how do you make invisible impact visible before review season?
Gaurav: Someone in an adjacent team once told me that because leadership doesn’t acknowledge pre-incident bugs that were fixed, some people resorted to storing these kind of information with them till the incident happened.
Once the incident happened, they would jump in, solve the incident in record time, and then get credited with solving a S1/S2 incident.
Next review cycle, they would either get promoted or get good ratings.
Not saying this is ethical or good for the team/company, but the entire perf review process needs to change if companies don’t want these kinds of things to happen.
☉rthonormalist: I had a friend at Facebook that caught a nine figure infra inefficiency
They gave him a four million dollar over five year bonus
Idk how common this is though
Wendigo: Caught a 7 figure one. Got a raise and a promotion out of it (probably worth proportionally in the same ballpark as what your friend got)
æthernet port: Lol, in my org at amazon there’d always be one or two people at the twice a year hackathon that’d give a presentation on how they optimized compute costs by $X-XX million a year, thus winning the $50 amazon gift card hackathon prize. Cracked me up every time.
Scott Alexander is correct that your solution to debate won’t work, even more so than that your startup probably won’t work, and gives some good explanations of why this problem is unusually hard. That doesn’t mean you shouldn’t try, and sometimes something comes along, like LessWrong, that at least makes things less awful.
I mostly buy the framing here that ‘the airport effect,’ as in worrying about any little thing that goes wrong spiraling into a nightmare, is the cause of a lot of unhappiness via continuous low-level anxiety. If things can go wrong but the loss is bounded, that is being alive, and all in good fun. You can relax. But when you worry that saying the wrong word or at any time being filmed doing the wrong thing, or any other small mistake, could ruin everything, that’s very different.
It’s also worth paying a rather high price to avoid this. I’m blessed to have been able to structure my life such that it would be very difficult to make a mistake that is all that large or spirals out of control, not without it basically being on purpose.
If you are often late or don’t show up to things, yes, this is you.
Should you beware your social media and its ability to sink your job prospects? That it will make you look unserious if you share sexy pictures or say the wrong thing? Sometimes yes, if you are looking to work at or apply to an institution with a stick up its ass. Some of those institutions pay very good money or offer a lot of prestige or influence. So you might want to curate your social media if that is your path. Then there is the opportunity for social media to make your career and give you opportunities, connections and reputation. It can go both ways, and how you act should depend on your particular situation.
A good rule of thumb is that in prediction markets, if they’re asking whether or not specific thing will happen, in a single name market, and you lack insider information, you can only bet no within the range of ~20%-70%. You can bet yes on things that are essentially done deals if the market is being obviously stubborn. That doesn’t mean automatically vote no, since there is adverse selection, but the bias here is very large. If you can simply avoid informed order flow, you can make quite quite a lot betting no.
A true point I follow but not as much as I should:
Alan Cole: In modern life you have to be truly ruthless about stripping companies of notification/email privileges. There is simply no other way to survive.
Even one unwanted email or notification should prompt ‘do I want to kill your access?’
Do not too strongly guard your recipes. At minimum, there should always be two people that know anything worth passing on, or that people love, and also it should be written down somewhere in case you pass away.
Oliver Habryka: In short: If you want a space to feel natural, buy lightbulbs with at least 95 CRI, ideally 98.
… If you are lighting a room with plenty of natural light, just use 2000K-3000K lights.
… The world got ugly when we invented LEDs.
Lighthaven is doing several other important things beyond this, but yes the light also matters. Said the person who spends all day looking at screens in a usually otherwise dark room.
The wisdom of crowds only works when people don’t put too much faith in the wisdom of crowds, or have too much modesty about experts.
(I don’t drink coffee and thus have no opinion about the underlying fact question.)
Joe Weisenthal: Is “caffè americano” (replicating drip coffee by diluting espresso with water) meant to be something of a subtle dig? It is in fact how I prefer to drink coffee, but am I supposed to be feel a tad embarrassed when I order it? (It’s fine, I can handle it, but I’m just curious)
Joe Weisenthal: If you look at the replies [to the above], you’ll see a lot of people confidently giving both possible answers. This is one reason why “hallucinations” are very low on my list of reasons to be worried about AI. Bullshit is just endemic to the production of language.
Alex Imas: I think the worry is that yes there’s a lot of noise in opinions/replies. But wisdom of crowds works by aggregating across many idiosyncratic answers. The thing that breaks wisdom of the crowds is a public source that everyone relies on instead of their private signal. so if hallucinations are “common” for everyone, it’s a very different type of bullshit in the aggregate.
In other ways, yes, you need everyone to agree to things and have a common knowledge base for coordination and shared reference purposes, but this is very different thing when done correctly.
Joe Weisenthal: This makes sense. But you could make the flip argument, that society only works if it has some shared myths (hallucinations) about the past and present, which can’t be maintained amid too much noise.
Alex Imas: Yes absolutely but these serve different purposes. Shared myths are for coordination of individuals within societies (you need public coordination devices, à la Schelling for that). But you also need independent signals for information aggregation.
As in, any given person has a level of intelligence and conscientiousness and motivation, and can have basic skill in any set of these four categories.
If you have any skill within the cluster, the theory goes, you can cross over to any other skill in the cluster within six months. But if you don’t yet have any skill in the cluster, then buckle up, it’s probably going to be a struggle and take a year or more.
Oliver Habryka:
And my current, schizo galaxy-brained theory is that there are exactly 4 skills:
Design skills: The ability to make good frontend design decisions, writing and explaining yourself well, designing a room, writing a good legal defense, knowing how to architect a complicated software system
Technical skills: Follow and perform mathematical proofs, know how to program, make Fermi estimates, make solid analytic arguments, read and understand a paper in STEM, follow economic arguments, make a business plan, perform structural calculations for your architectural plans
Management skills: Know how to hire people, know how to give employees feedback, generally manage people, navigate difficult organizational politics
Physical skills: Be expert level at any sport, have the physical dexterity to renovate a room by yourself, know how to dance
If you are good at any task in any of those categories, you can become expert-level within 6 months at any other task in the same category.
I definitely know what it feels like to transition from not having management skills to having management skills, and yeah, that is pretty brutal, but it can be done, as can learning to be up to some level of physically skilled. It is not obvious that you can force your way into design or technical as easily.
There’s definitely some things importantly missing here, but one could argue that the missing things do not belong to the category of ‘skills’ as it is being imagined here. Or perhaps there is a fifth category, a kind of ‘make things happen’ that goes well beyond knowing how to manage people and feels distinct to me, among other things missing.
But yeah, it’s a cool fake framework.
While I Cannot Condone This
The general case of this is remarkably common, where good news is bad news:
Paul Graham: A watchmaker told me that he prefers it when he opens a watch that needs service and the movement is dirty. If the movement is clean, it’s more likely the problem is structural.
Matt Levine says that obviously Tesla should have given Elon Musk supervoting stock, so he could keep control of the company without having to constantly demand more stock, but now that they are public it is too late. I would say, nay, it is much better to not give Elon Musk such stock, because not having it allows him to constantly demand being paid additional huge amounts of stock, while still having effective control of the company. That’s much better. For Elon Musk.
Basically Hungary was being governed by corrupt bandits, Orban was a traditional tyrant in the ancient Greek sense where people empower an individual to overthrow the system when the system is sufficiently stacked against them and are willing to accept the tradeoffs involved in that, and Obran went about consolidating power, including control of the courts and media, and tried to improve some of the things the people hated.
But the costs to Orban of maintaining the centralized patronage system required to maintain his control caught up with him.
When he tried to flood the system with money and tempt the economic gods, fighting back with price controls, he predictably got smote.
The EU whined about Orban’s actions and failures to follow EU procedures, but its only lever was money, which it only used when Orban started vetoing sanctions against Russia, and the EU dropped the matter after he stopped.
Finally in February 2024 there was a scandal and claims of corruption that broke through in language the people understood and cared about, energy prices spiked and the combination turned people against Orban. He needed the EU’s money and he needed to also symbolically be fighting the EU at the same time, and he couldn’t do both.
The media was controlled but the elections were real. The government fell.
His summary is here, very nicely compacted:
@ben_r_hoffman: The decentralized system before him had been corrupt and unresponsive; Orbán’s was corrupt and responsive. But centralized patronage accelerates: yesterday’s bribe is today’s entitlement, so each cycle requires more resources to maintain the coalition.
This is the final boss of centralized patronage, which is the only known way to sustain authoritarianism without fully legitimated moral authority. You might mostly mean well, and even do a bunch of good things, but costs rise and eventually they catch up with you. The issue is, what do you do if the existing legitimated authorities are hopelessly unresponsive and in what is ultimately a death spiral?
As usual, you can be somebody or you can do something.
JMDavis: I can’t believe I need to say this, but if, as a journalist, you use my time, research & expertise without quoting me, don’t be surprised if I never return your call.
I guess I could add here: I always ask if the interview is on background. And I don’t agree to do it if that’s the case.
Seth Burn: If you use my research and expertise, I’ll answer your call any time I see it.
The point is to get that research and expertise out there. Mostly. You talk to the reporter to help them get the story right, not to get the credit.
I typically ask if the interview is on background, and ideally for most topics it is, because then I can speak more freely, and we can discuss any potential quotes later. When it’s on the record, I need to choose my words carefully.
Davis clarifies that he means cases where he gives hours of his time to a story, and then fails to get any acknowledgment. I do think that, at some point, this is fair. I’ve never spent many hours of my time helping with someone else’s story, and if I did then I would expect to be at least acknowledged in some way, and at some point I’d want to be paid for my time.
shako: substack should have a button on paid articles, enabled as an option by authors, that lets a benefactor “buy it for everyone.” e.g. a rich benefactor may think it to be worth $500 or some such amount to make an article behind a paywall free for everyone.
I think the GameStop bid to buy EBay actually had a lot more real merit than people are giving it credit for. Yes, the financial part of it is a train wreck, but I’m talking about the part where EBay uses GameStop’s outlets as physical stores for grading, and presumably also marketing, pickup, drop-off and inventory management, and highlighting offerings, even if they hadn’t figured those parts out yet.
From what I can tell, GameStop is sitting on a severely underutilized asset, which is a large number of physical stores with a bunch of monopoly businesses and loyal customers and goodwill, but which mostly are sitting idle all day. You could use that resource for any number of related things. EBay wouldn’t have been my first pick, but it makes a lot of sense. My actual first pick is community gaming centers and hosters of tournaments, as a proper third space.
Are gamers who remember old 80s or 90s games fondly using rose colored glasses full of nostalgia? Partly, of course, but also the great games were legitimately great, and the limited complexity and discrete graphics bred creativity and made the player have all the fun, and those making games just made whatever they wanted. There’s a reason my kids love their mini-SNES and mini-NES, and I largely play games that could have been made back then even if they weren’t.
The old games also benefited from patience, and a willingness to be frustrated and persevere and make up front investments. You didn’t have unlimited other options, you didn’t have constant demands on your attention, and it showed.
I notice the parallel to movies. In movies I think the older movies, up until the 90s, sometimes had their charms but overall were basically just worse and now we demand more, whereas in games where the old games in many important ways were way worse I have the nostalgia. I still think I’m right about both of them.
The Spire Sleeps And So Shall I
Slay the Spire II is more Slay the Spire. This is high praise. If you enjoyed many hours of the first game, play the second game. I would describe StS II as a heavy mod more than a full game, but that is a good thing, and good for at least dozens of hours. I recommend playing a bunch of runs blind first, and only looking up info or watching streams once you start losing a lot of runs.
Ascension now condenses 20 levels down into 10. Up through Level 7 it’s all fun and games, Level 8 is touch and go, and then Level 9 is a huge jump that flipped my experience from ‘I should usually be able to win’ to ‘wow this is brutal.’ Level 10 isn’t that much harder than 9, although the change did cost me one of the four a10 wins I would have otherwise gotten.
The game has suffered from some review bombs, that I hear are mostly coming from China, because players feel like the designers are taking away their fun.
I interpret the criticisms largely as, basically, ‘I want to do my thing and then win the run. When my thing does not win the run, I get mad.’ People don’t want to lose runs that they feel they ‘deserve to win.’ They especially don’t want to hit a whammy boss, where their deck suddenly doesn’t work.
I think the game’s actual main problem is exactly the opposite. The difficulty is far too concentrated in the Act 1, whereas Act 3 usually ends up as a victory lap or a quest to find that extra scaling for the end boss. I want the opposite of that. I want to take risks early exactly because I need to be stronger later, and I want to be sweating the final fights and taking cards and making plans in particular for that ending. Jorbs explains all of this quite well in his videos.
In the first game, a brutal second act forces you to take risks to get strong quickly, and then the boss gauntlet poses a bunch of hard questions you need to plan for the whole game. Now the biggest challenge is getting past the Act 1 boss while having anything going on at all, and after that it usually snowballs. The end bosses ask you to scale your damage, but don’t have the same ‘you need to plan for this in particular’ that we got with Time Eater, Awakened One or The Heart.
There was an experimental redesign of The Doormaker that eats every tenth card you draw. So if you go into the final battle depending on one card, you might lose, and if your plan is repeatedly drawing a card or cards and going infinite, you almost certainly lose. You need backup plans. Jorbs called it the best boss design he’d ever seen, but people hated it so much they had to revert it. Sad.
I think the plan of ‘make Ascension 0 easier, and Ascension 10 harder’ makes sense, especially making Act 3 harder on higher difficulties, and hopefully introducing a new ending and better final fights. Maybe True Doormaker shows up only at Ascension 10, and other similar changes, in addition to the double boss, or we go to Ascension 11. By contrast, I would make it easier to get out of Act 1 on lower difficulties.
The other big controversy was that going infinite started off rather easy. That’s what I am told, anyway. I was playing in ‘infinite is easy’ mode, and I had a number of decks that could have gone infinite in theory, but every time I was setting up to do it, the enemy died first. Then they changed a bunch of cards to make infinites harder. I think they did this in ways that ‘killed the fun’ and were too paranoid about the infinites, and instead of making fun cards worse we should focus on specific anti-infinite measures.
Across the Obelisk did this with madness levels, which are similar to Ascension levels. In the base game infinite is rather straightforward if you build to it. Then you get a modifier that every time you shuffle, everything you draw after that on the same turn costs 1 more, so you can’t do it. That particular rule is too harsh for Slay the Spire, because the first shuffle being penalized is crazy, but a modified version could be an ascension restriction, such as ‘when you replay a card on the same turn, bind it’ or ‘you can only reshuffle your deck once each turn.’
The important thing is, I want my cards to be fun. I want to have the ‘good stuff’ feeling, like every card does something cool or that could be cool, even if it’s not ultimately that useful or impactful, unless you actively want to put whammies in so that players can transform into them.
I’d also note the balance is a bit out of whack between characters. Silent is way more powerful than the other four. In my experience Ironclad is hardest on higher difficulty by a decent margin, when its ‘spend my health’ strategies run out of health. Defect is unique in that it seems like it actually fails to scale reasonably often, also it has too many powers that kind of suck. I get pulling back on focus but it feels like we went a bit too far in focusing the power in value cards.
My plan is to play occasionally, and at some point pick up the other two Ascension 10 wins, and come back when there’s a major upgrade, like another character or an ending.
Tom Steyer: AI shouldn’t put California truckers out of work to pad Big Tech’s profits. As governor, I’ll reverse the DMV’s autonomous trucking rules and keep human drivers on the road.
Needless to say, this is quite bad, but also yes you would rather pay a human to do literal nothing than let that human drive the truck.
I consider this utterly disqualifying for the job of Governor of California.
Organizermemes: Dc not allowing Waymo for safety reasons is insane when there’s no ban on Maryland drivers.
Self-driving buses are a great idea, potentially dramatically lowering the cost of providing buses and enabling much better service on multiple levels. Alas, there are those worried about ‘jobs’ trying to prevent this, as one would expect.
Joakim: For the first time in Norwegian history, a bus will carry passengers in regular traffic without any human behind the wheel. The first pilot without a safety driver was tested Friday, and if all goes as planned, anyone can ride driverless buses starting in May.
Alexander: For those that don’t buy this, these buses have operated driverless in Stavanger for years now, but have had a driver just in case, so they already know how this bus operates, that’s how they feel comfortable actually going all the way now!
I do think there are real concerns with fully unattended buses. You totally 100% do not need a human to drive the bus, but the driver is also keeping order and collecting fares, including acting as a deterrent, and answering questions. This feels like it is being under considered on all sides. It’s not obvious you need that enough to justify a human, but it also isn’t obvious you don’t.
One important downside here is that drug dealers and other sellers of vice benefit from the synergies of being universal suppliers, and there is more reason to become corrupt. If your local drug dealer is where you have to go for cigarettes, rule of law suffers, and that is going to boost availability of all the other drugs, along with things like gambling and prostitution and so on. So it’s not all bad.
If you are going to ban smoking, grandfathering in existing legal smokers is the right way to do it. You don’t want to strand existing smokers who are addicted, and I could not figure out a clean way to grandfather in only the addicts. It does mean access to cigarettes will remain pretty easy for a long time, and that’s the price. But you really should not be doing it at all.
@levelsio: My gf is banned from reviewing places in Europe on Google Maps after she gave one restaurant in Portugal a 1-star review
When she reviews inside EU it gets auto rejected, outside EU she can review any place
Free speech in Europe has sadly died a long time ago
j.m. kettle: Giving a restaurant a three star review is illegal in Germany.
Karl DXB: At least they show you how many reviews an establishment wanted to have removed. Which says more about the establishment than the reviews in many cases.
Porkchop Express: I once had an argument with a German seller on Amazon and said I’ll have to leave a negative review if they don’t refund me and they threatened me with a legal proceeding. Amazon confirmed I shouldn’t have said that.
Sid Kingsley: I live in Austria this is a known thing in Germany. The restaurants can say any review is defamatory and Google has to remove it. It’s why you can’t trust google reviews at any restaurants in Deutschland.
We have junk mail because the government gives a massive discount to junk mail, which it calls marketing mail. We could simply stop offering that discount.
The Jones Act is pure rent seeking. This is common knowledge.
If you see someone defending the Jones Act, 100% of the time it is either a rent seeker, or someone who is carrying water for rent seekers for whatever reason, usually out of political fear but occasionally out of confusion. That is it.
The recent experiment of waiving part of the Jones Act only makes this even clearer.
Colin Grabow: Zero US tankers are unemployed due to the JA waiver. Thus far, international ships have been supplementing, not replacing, the US fleet (such as it is, with a mere 54 tankers).
Eric Priante Martin: White House spokeswoman Taylor Rogers said data showed that “significantly more supply was able to reach US ports faster” after the initial waiver.
Scott Lincicome: Yep. We’re also told that more/cheaper ships aren’t needed (bc of sufficient JA ships and rail/interstate alternatives), yet even this limited JA waiver has already been used dozens of times for short- and long-haul trips.
Colin Grabow: By country of vessel owner/operator:
Tradewinds covered the impact of the partial waiver so far. They don’t see that many voyages so far, counting 19 at the time, which is now up to at least 33. This would rise over time if the waiver was permanent, and thus allowed for long term planning and investment. One important thing is that no ships have been displaced. These voyages are in addition to, not instead of.
It is hard to know the magnitude of the impact on gas prices, since the changes in global supply and market expectations for supply are going to dwarf everything else. The Trump administration says the new ships have been ‘incredibly helpful’ in stabilizing markets.
I mean, they’re actually trying this line now, which is purely false in addition to how ludicrous it would be even if it technically wasn’t:
So now they continue as pure rent seekers. They only lose when the Jones Act is actually repealed.
The even better news is that we are approaching stage three, where they fight you.
Scott Lincicome: HOW INTERESTING: Within the span of a few hours yesterday, three X influencers (~1M total followers) who’d never posted about the Jones Act suddenly came out in support.
One noted she was paid for the post. I’m sure the other two are a coincidence.
We’re winning, folks.
TO BE CLEAR: This isn’t abt whether these folks violated X disclosure rules. Instead, it’s just more proof that Jones Act support isn’t some grassroots, pro-worker movement of Americans sincerely worried about China or natsec.
It’s just cronies, mercenaries, political opportunists, & a few crusty/misguided cranks. That’s it.
The posts are by Kambree (or should I say ChatGPT?), the second is Arynne Wexler (using China fearmongering as if any ship that ever touched China is irrevocably ‘dependance’ or infiltrated somehow, or something, and therefore we shouldn’t buy Korean or Dutch ships) and the third is Olivia Krolczyk (the one that is disclosed as paid and straight up copies industry text). A fourth followed from Kaya Jones.
There is no mistaking the rhetoric involved. It is not people thinking or having actual beliefs. It’s pure and unmistakable talking points and marching orders.
Ira Joseph: The Jones Act was once so powerful, it did not bother with defending itself from detractors. The fact that someone felt the need to bot up some resistance suggests vulnerabilities are finally starting to appear now that exemptions have emerged.
Technology Advances
It remains possible to fill up all your Google storage and thus become unable to receive emails, in conflict with Google’s long term promises that you’ll have de facto infinite storage even if you never throw anything away. That’s true for a default user, but yes if you abuse the system with tons of huge attachments eventually you’re going to run into space issues. The system could do a better job helping you not actually be unable to take in new emails then, especially ones without large attachments.
Yes, we should 100% bring back the iPod in its full glory, including the clickwheel, except with a modern processor and storage capacity, USB-C charger, bluetooth and access to Apple Music. There is a lot of value in the old school, and in being able to not be tempted by a full smartphone. Doing this for the 25th anniversary as a 1-off would be great, and if it’s a big enough hit, then who knows. I mean, yeah, I get that the total market might be small, but I suspect it isn’t.
I Said Woo Hoo
Kaj Sotala asks, how should we think about woo, meaning various practices that obviously don’t work the literal way they are presented, but that seem to do real work? Think tarot, energy healing, chakras, transcendental meditation and so on. They are sold and presented in ways that make no scientific sense, yet something is happening.
I agree with his interpretation of Tarot. Tarot is a way to generate random associative prompts, both vague and specific at once, that allow you to explore, and pay to know what you really think in important senses, or use to help explore together. It’s actually good technology, and at times I’ve gotten use out of it.
The trick is, it’s an advanced concept to be in the mindset where you know what the various woo things are and don’t suspend disbelief, and yet you can still let the active ingredients work. I find that relatively easy for tarot (or a later example, IFS), but for this second example, chakras and placing things within the body, it’s harder, and my guess is for energy healing (the third example) it would be harder still.
Then there are other at least somewhat ‘woo’ things where I’ve gotten a bunch of value, far more than with tarot, that he doesn’t namecheck, but that’s cases of IYKYK. I do think ‘try things’ is a good principle if you’re not susceptible to falling into traps.
Variously Effective Altruism
Ulisse Mini is on point here: It’s grifting and should feel like grifting if you ask someone for money inauthentically or in a way that the person giving you money would regret if they had the full story. If you genuinely only want them to say yes if they would benefit too, then is fine and it should feel fine.
Aaron Rupar: RFK Jr: “A Democratic senator claimed it’s mathematically impossible to have a drug drop by 600%. I said, ‘Well, if the drug was $100 and it raises to $600, that would be a 600% rise. If it drops from $600 to $100, that’s a 600% savings.’”
At least we probably won’t have another pandemic. And we still have a partial Jones Act waiver. For now.
Small victories.
Table of Contents
Hanta Hanta I Don’t Wanta
We have learned so much less than nothing from Covid. We’re actively stupider.
It’s 2026, and here we are again, lying about the virus because we are worried that people exposed to it or from the wrong place might face stigma otherwise.
Meanwhile, the other side is already also out in force and stupidity, to say in advance that no one is going to trust anything anyone says, or follow any orders, and if they say anything is happening it is fake, that if it does happen it will be because of the Covid vaccine, and I’ve even seen a literal claim that one should take Ivermectin.
The CDC is nowhere to be found. The WHO lacks any authority to get anyone to quarantine, and then there are those who treat this as a valid excuse.
What matters most is that we have now learned that Earth, in 2026, is completely incapable of taking steps to prevent even a highly preventable obvious pandemic.
An abundance of caution is not always good. But when costs are trivial, for things like ‘wear at least N95s on the literal quarantine unit’? At least do that, you fools.
The good news is I believe, and Peter Wildeford agrees, that the chance of getting an actual Hantavirus pandemic are quite small. If I was trading the prediction markets, I would sell. Hantavirus is almost certainly not infectious enough in its current form to cause a pandemic. Even with our complete lack of reasonable precautions, R0 will probably be less than one.
The bad news is that this is entirely good luck. If hantavirus was capable in its current form of causing a pandemic, you know what we would be facing down? A pandemic.
It cannot be overstated how determined we are not to take any actions that would actually prevent a potential pandemic. That could soon include something engineered via AI, and again I expect us to react maximally stupidly if that happens.
I realize we are incapable of keeping the AI in a box, but you would think we would be able to keep people on a ship. Instead, even after we know what is happening, we send them home on flights, and hope for the best, while pretending it is all going to be fine.
Even if there is little risk of a full pandemic, even the amount of attention this has already gotten, and the amount of distraction and stress it has already caused, has exceeded the costs of handling the situation properly. If even a handful of cases get out, panic could ensue along much larger areas, again even if we are correct that the risk of a pandemic is minimal.
Bad News
Patrick McKenzie covers the indictment of SPLC for bank fraud, concluding that they definitely did a bunch of bank fraud, and also documents how they led a coalition to create de facto financial infrastructure and work to deplatform political opposition. I consider Patrick a supremely credible source here. Alex Tabarrok summarizes.
We badly need to repeal the Davis-Bacon Act, which requires ‘prevailing wages’ on federal construction contracts. On top of directly raising costs it creates giant tracking headaches and can lead to ‘retroactive pay.’
A large percentage of costs is compliance, which means this is far worse than a typical minimum wage that at least is easy to understand and comply with. Another simple argument against this is that if it truly was the prevailing wage, you wouldn’t be able to get away with not paying it. If you did, it wasn’t so prevailing, was it?
The blue versus red button experiment discourse is like a social disease.
(With apologies to Marion Barry and also the former Capital Steps, IYKYK.)
If you catch a giant bug or inefficiency worth eight or nine figures, you might get a percent of the profits, or you might get a pat on the back. Incentives do not seem ideal. Seems important to know which kind of employer you are working for.
Scott Alexander is correct that your solution to debate won’t work, even more so than that your startup probably won’t work, and gives some good explanations of why this problem is unusually hard. That doesn’t mean you shouldn’t try, and sometimes something comes along, like LessWrong, that at least makes things less awful.
I mostly buy the framing here that ‘the airport effect,’ as in worrying about any little thing that goes wrong spiraling into a nightmare, is the cause of a lot of unhappiness via continuous low-level anxiety. If things can go wrong but the loss is bounded, that is being alive, and all in good fun. You can relax. But when you worry that saying the wrong word or at any time being filmed doing the wrong thing, or any other small mistake, could ruin everything, that’s very different.
It’s also worth paying a rather high price to avoid this. I’m blessed to have been able to structure my life such that it would be very difficult to make a mistake that is all that large or spirals out of control, not without it basically being on purpose.
Predictions Can Be Easy Even About The Future
For example, if you predict and bet on yourself to run for Senate, and then run.
Good Advice
If you are often late or don’t show up to things, yes, this is you.
Should you beware your social media and its ability to sink your job prospects? That it will make you look unserious if you share sexy pictures or say the wrong thing? Sometimes yes, if you are looking to work at or apply to an institution with a stick up its ass. Some of those institutions pay very good money or offer a lot of prestige or influence. So you might want to curate your social media if that is your path. Then there is the opportunity for social media to make your career and give you opportunities, connections and reputation. It can go both ways, and how you act should depend on your particular situation.
A good rule of thumb is that in prediction markets, if they’re asking whether or not specific thing will happen, in a single name market, and you lack insider information, you can only bet no within the range of ~20%-70%. You can bet yes on things that are essentially done deals if the market is being obviously stubborn. That doesn’t mean automatically vote no, since there is adverse selection, but the bias here is very large. If you can simply avoid informed order flow, you can make quite quite a lot betting no.
A true point I follow but not as much as I should:
Even one unwanted email or notification should prompt ‘do I want to kill your access?’
Do not too strongly guard your recipes. At minimum, there should always be two people that know anything worth passing on, or that people love, and also it should be written down somewhere in case you pass away.
A plausible life hack, maintain your to-do list on your iPhone via screenshots, since it’s an easy button press to take one or look at them? Seems plausible, and a lot more interesting now that you can have AI comb through it all and turn it into a better format.
Oliver Habryka advises that if people don’t like your space, usually it is because of low quality lighting.
Lighthaven is doing several other important things beyond this, but yes the light also matters. Said the person who spends all day looking at screens in a usually otherwise dark room.
If you are on a trip abroad and you are asked in what currency you wish to pay, it is almost always cheaper to pay with your card in local currency than in dollars.
The Efficient Market Hypothesis Is False
The wisdom of crowds only works when people don’t put too much faith in the wisdom of crowds, or have too much modesty about experts.
(I don’t drink coffee and thus have no opinion about the underlying fact question.)
In other ways, yes, you need everyone to agree to things and have a common knowledge base for coordination and shared reference purposes, but this is very different thing when done correctly.
There Are Four Skills
We love a good schizo galaxy-brained theory here, so may I present Oliver Habryka’s thesis that there are only four skills: Design, technical, management and physical.
As in, any given person has a level of intelligence and conscientiousness and motivation, and can have basic skill in any set of these four categories.
If you have any skill within the cluster, the theory goes, you can cross over to any other skill in the cluster within six months. But if you don’t yet have any skill in the cluster, then buckle up, it’s probably going to be a struggle and take a year or more.
I definitely know what it feels like to transition from not having management skills to having management skills, and yeah, that is pretty brutal, but it can be done, as can learning to be up to some level of physically skilled. It is not obvious that you can force your way into design or technical as easily.
There’s definitely some things importantly missing here, but one could argue that the missing things do not belong to the category of ‘skills’ as it is being imagined here. Or perhaps there is a fifth category, a kind of ‘make things happen’ that goes well beyond knowing how to manage people and feels distinct to me, among other things missing.
But yeah, it’s a cool fake framework.
While I Cannot Condone This
The general case of this is remarkably common, where good news is bad news:
Matt Levine says that obviously Tesla should have given Elon Musk supervoting stock, so he could keep control of the company without having to constantly demand more stock, but now that they are public it is too late. I would say, nay, it is much better to not give Elon Musk such stock, because not having it allows him to constantly demand being paid additional huge amounts of stock, while still having effective control of the company. That’s much better. For Elon Musk.
Do not confused patent filings with finding ideas.
Benjamin Hoffman gives us a factual overview of what actually happened with Orban.
My summary of his history here:
His summary is here, very nicely compacted:
This is the final boss of centralized patronage, which is the only known way to sustain authoritarianism without fully legitimated moral authority. You might mostly mean well, and even do a bunch of good things, but costs rise and eventually they catch up with you. The issue is, what do you do if the existing legitimated authorities are hopelessly unresponsive and in what is ultimately a death spiral?
As usual, you can be somebody or you can do something.
The point is to get that research and expertise out there. Mostly. You talk to the reporter to help them get the story right, not to get the credit.
I typically ask if the interview is on background, and ideally for most topics it is, because then I can speak more freely, and we can discuss any potential quotes later. When it’s on the record, I need to choose my words carefully.
Davis clarifies that he means cases where he gives hours of his time to a story, and then fails to get any acknowledgment. I do think that, at some point, this is fair. I’ve never spent many hours of my time helping with someone else’s story, and if I did then I would expect to be at least acknowledged in some way, and at some point I’d want to be paid for my time.
Good News, Everyone
Reminder that you can buy 50lbs of rice at Costco for ~1hr of average hourly wages. You know, maybe we’re not that poor.
This is a great idea:
Cate Hall recommends the MyHalos Sleep Mask, says it gets it done for only $10. Tenobrus by contrast had to shell out $90 for the Manta Pro.
For Your Entertainment
Movie theaters are recovering, with number of tickets sold going up despite, let’s face it, a rather terrible first four months of the year in terms of movies.
The most desired movie tickets for new releases are getting more expensive, up to $50, although not yet expensive enough given the market did not clear.
Geeks and Gamers tries to incept the idea that Disney is going to retcon away the entire sequel trilogy from Star Wars. We thank you for the noble attempt.
Gamers Gonna Game Game Game Game Game
I think the GameStop bid to buy EBay actually had a lot more real merit than people are giving it credit for. Yes, the financial part of it is a train wreck, but I’m talking about the part where EBay uses GameStop’s outlets as physical stores for grading, and presumably also marketing, pickup, drop-off and inventory management, and highlighting offerings, even if they hadn’t figured those parts out yet.
From what I can tell, GameStop is sitting on a severely underutilized asset, which is a large number of physical stores with a bunch of monopoly businesses and loyal customers and goodwill, but which mostly are sitting idle all day. You could use that resource for any number of related things. EBay wouldn’t have been my first pick, but it makes a lot of sense. My actual first pick is community gaming centers and hosters of tournaments, as a proper third space.
Are gamers who remember old 80s or 90s games fondly using rose colored glasses full of nostalgia? Partly, of course, but also the great games were legitimately great, and the limited complexity and discrete graphics bred creativity and made the player have all the fun, and those making games just made whatever they wanted. There’s a reason my kids love their mini-SNES and mini-NES, and I largely play games that could have been made back then even if they weren’t.
The old games also benefited from patience, and a willingness to be frustrated and persevere and make up front investments. You didn’t have unlimited other options, you didn’t have constant demands on your attention, and it showed.
I notice the parallel to movies. In movies I think the older movies, up until the 90s, sometimes had their charms but overall were basically just worse and now we demand more, whereas in games where the old games in many important ways were way worse I have the nostalgia. I still think I’m right about both of them.
The Spire Sleeps And So Shall I
Slay the Spire II is more Slay the Spire. This is high praise. If you enjoyed many hours of the first game, play the second game. I would describe StS II as a heavy mod more than a full game, but that is a good thing, and good for at least dozens of hours. I recommend playing a bunch of runs blind first, and only looking up info or watching streams once you start losing a lot of runs.
Ascension now condenses 20 levels down into 10. Up through Level 7 it’s all fun and games, Level 8 is touch and go, and then Level 9 is a huge jump that flipped my experience from ‘I should usually be able to win’ to ‘wow this is brutal.’ Level 10 isn’t that much harder than 9, although the change did cost me one of the four a10 wins I would have otherwise gotten.
The game has suffered from some review bombs, that I hear are mostly coming from China, because players feel like the designers are taking away their fun.
I interpret the criticisms largely as, basically, ‘I want to do my thing and then win the run. When my thing does not win the run, I get mad.’ People don’t want to lose runs that they feel they ‘deserve to win.’ They especially don’t want to hit a whammy boss, where their deck suddenly doesn’t work.
I think the game’s actual main problem is exactly the opposite. The difficulty is far too concentrated in the Act 1, whereas Act 3 usually ends up as a victory lap or a quest to find that extra scaling for the end boss. I want the opposite of that. I want to take risks early exactly because I need to be stronger later, and I want to be sweating the final fights and taking cards and making plans in particular for that ending. Jorbs explains all of this quite well in his videos.
In the first game, a brutal second act forces you to take risks to get strong quickly, and then the boss gauntlet poses a bunch of hard questions you need to plan for the whole game. Now the biggest challenge is getting past the Act 1 boss while having anything going on at all, and after that it usually snowballs. The end bosses ask you to scale your damage, but don’t have the same ‘you need to plan for this in particular’ that we got with Time Eater, Awakened One or The Heart.
There was an experimental redesign of The Doormaker that eats every tenth card you draw. So if you go into the final battle depending on one card, you might lose, and if your plan is repeatedly drawing a card or cards and going infinite, you almost certainly lose. You need backup plans. Jorbs called it the best boss design he’d ever seen, but people hated it so much they had to revert it. Sad.
I think the plan of ‘make Ascension 0 easier, and Ascension 10 harder’ makes sense, especially making Act 3 harder on higher difficulties, and hopefully introducing a new ending and better final fights. Maybe True Doormaker shows up only at Ascension 10, and other similar changes, in addition to the double boss, or we go to Ascension 11. By contrast, I would make it easier to get out of Act 1 on lower difficulties.
The other big controversy was that going infinite started off rather easy. That’s what I am told, anyway. I was playing in ‘infinite is easy’ mode, and I had a number of decks that could have gone infinite in theory, but every time I was setting up to do it, the enemy died first. Then they changed a bunch of cards to make infinites harder. I think they did this in ways that ‘killed the fun’ and were too paranoid about the infinites, and instead of making fun cards worse we should focus on specific anti-infinite measures.
Across the Obelisk did this with madness levels, which are similar to Ascension levels. In the base game infinite is rather straightforward if you build to it. Then you get a modifier that every time you shuffle, everything you draw after that on the same turn costs 1 more, so you can’t do it. That particular rule is too harsh for Slay the Spire, because the first shuffle being penalized is crazy, but a modified version could be an ascension restriction, such as ‘when you replay a card on the same turn, bind it’ or ‘you can only reshuffle your deck once each turn.’
The important thing is, I want my cards to be fun. I want to have the ‘good stuff’ feeling, like every card does something cool or that could be cool, even if it’s not ultimately that useful or impactful, unless you actively want to put whammies in so that players can transform into them.
I’d also note the balance is a bit out of whack between characters. Silent is way more powerful than the other four. In my experience Ironclad is hardest on higher difficulty by a decent margin, when its ‘spend my health’ strategies run out of health. Defect is unique in that it seems like it actually fails to scale reasonably often, also it has too many powers that kind of suck. I get pulling back on focus but it feels like we went a bit too far in focusing the power in value cards.
My plan is to play occasionally, and at some point pick up the other two Ascension 10 wins, and come back when there’s a major upgrade, like another character or an ending.
I Was Promised Flying Self-Driving Cars
There was an article recently claiming self-driving cars are ‘less able to detect people of color.’ Kelsey Piper dutifully looked into this and found the claim entirely false. The problem is that the article cited a source that did make the claim, it’s just that the source was making it up.
Tom Steyer, a leading candidate for governor of California says if elected governor he would mandate human safety drivers in autonomous delivery vehicles, and says the worst version of this.
Needless to say, this is quite bad, but also yes you would rather pay a human to do literal nothing than let that human drive the truck.
I consider this utterly disqualifying for the job of Governor of California.
Alas, the other leading Democratic candidate, Xavier Becerra, has the same position. That leaves Matt Mahan, perhaps?
Yes, Washington DC is claiming to disallow Waymo due to ‘safety’ concerns.
Self-driving buses are a great idea, potentially dramatically lowering the cost of providing buses and enabling much better service on multiple levels. Alas, there are those worried about ‘jobs’ trying to prevent this, as one would expect.
I do think there are real concerns with fully unattended buses. You totally 100% do not need a human to drive the bus, but the driver is also keeping order and collecting fares, including acting as a deterrent, and answering questions. This feels like it is being under considered on all sides. It’s not obvious you need that enough to justify a human, but it also isn’t obvious you don’t.
Government Working
If this is how a federal prison treats a sitting US Senator, imagine how they treat everyone else.
Executive Branch now flat out extrajudiciously banning all wind power projects via denying FAA approvals regardless of merit.
The UK bans smoking permanently for everyone born after 2008, as in everyone who is not currently 18. Prohibition does not work. If you want less of something like smoking, you can tax it, but do not ban.
One important downside here is that drug dealers and other sellers of vice benefit from the synergies of being universal suppliers, and there is more reason to become corrupt. If your local drug dealer is where you have to go for cigarettes, rule of law suffers, and that is going to boost availability of all the other drugs, along with things like gambling and prostitution and so on. So it’s not all bad.
If you are going to ban smoking, grandfathering in existing legal smokers is the right way to do it. You don’t want to strand existing smokers who are addicted, and I could not figure out a clean way to grandfather in only the addicts. It does mean access to cigarettes will remain pretty easy for a long time, and that’s the price. But you really should not be doing it at all.
The contrary view is here from Liz Boeree, who thinks the new smoking ban is ‘insane in every way’ and actively way worse than a full ban.
Criticism via reviews in public is largely de facto impossible in at least some parts of Europe, even when there’s no protected class or other particular sins involved. I love the German example because it’s so clean: Three stars, ‘It was fine.’
We have junk mail because the government gives a massive discount to junk mail, which it calls marketing mail. We could simply stop offering that discount.
I presume there is some explanation for why this is not Trump literally just stealing $1.7 billion dollars from the United States, but for the life of me I can’t find one.
Jones Act Watch
The Jones Act is pure rent seeking. This is common knowledge.
If you see someone defending the Jones Act, 100% of the time it is either a rent seeker, or someone who is carrying water for rent seekers for whatever reason, usually out of political fear but occasionally out of confusion. That is it.
The recent experiment of waiving part of the Jones Act only makes this even clearer.
None of the ships taking advantage of the waiver are Chinese flagged. All the ships are operated by our allies. Yes, some of them are Chinese built, but there is no reason to care about that.
Tradewinds covered the impact of the partial waiver so far. They don’t see that many voyages so far, counting 19 at the time, which is now up to at least 33. This would rise over time if the waiver was permanent, and thus allowed for long term planning and investment. One important thing is that no ships have been displaced. These voyages are in addition to, not instead of.
It is hard to know the magnitude of the impact on gas prices, since the changes in global supply and market expectations for supply are going to dwarf everything else. The Trump administration says the new ships have been ‘incredibly helpful’ in stabilizing markets.
Scott Lincicome has a graph of what routes the ships are taking, and asserts that defenders of the Jones Act have lost. And yes, obviously they have no rhetorical leg to stand on and all their arguments are exposed as bogus. Which was already true, and even more true now.
I mean, they’re actually trying this line now, which is purely false in addition to how ludicrous it would be even if it technically wasn’t:
So now they continue as pure rent seekers. They only lose when the Jones Act is actually repealed.
The even better news is that we are approaching stage three, where they fight you.
The posts are by Kambree (or should I say ChatGPT?), the second is Arynne Wexler (using China fearmongering as if any ship that ever touched China is irrevocably ‘dependance’ or infiltrated somehow, or something, and therefore we shouldn’t buy Korean or Dutch ships) and the third is Olivia Krolczyk (the one that is disclosed as paid and straight up copies industry text). A fourth followed from Kaya Jones.
There is no mistaking the rhetoric involved. It is not people thinking or having actual beliefs. It’s pure and unmistakable talking points and marching orders.
Technology Advances
It remains possible to fill up all your Google storage and thus become unable to receive emails, in conflict with Google’s long term promises that you’ll have de facto infinite storage even if you never throw anything away. That’s true for a default user, but yes if you abuse the system with tons of huge attachments eventually you’re going to run into space issues. The system could do a better job helping you not actually be unable to take in new emails then, especially ones without large attachments.
Yes, we should 100% bring back the iPod in its full glory, including the clickwheel, except with a modern processor and storage capacity, USB-C charger, bluetooth and access to Apple Music. There is a lot of value in the old school, and in being able to not be tempted by a full smartphone. Doing this for the 25th anniversary as a 1-off would be great, and if it’s a big enough hit, then who knows. I mean, yeah, I get that the total market might be small, but I suspect it isn’t.
I Said Woo Hoo
Kaj Sotala asks, how should we think about woo, meaning various practices that obviously don’t work the literal way they are presented, but that seem to do real work? Think tarot, energy healing, chakras, transcendental meditation and so on. They are sold and presented in ways that make no scientific sense, yet something is happening.
I agree with his interpretation of Tarot. Tarot is a way to generate random associative prompts, both vague and specific at once, that allow you to explore, and pay to know what you really think in important senses, or use to help explore together. It’s actually good technology, and at times I’ve gotten use out of it.
The trick is, it’s an advanced concept to be in the mindset where you know what the various woo things are and don’t suspend disbelief, and yet you can still let the active ingredients work. I find that relatively easy for tarot (or a later example, IFS), but for this second example, chakras and placing things within the body, it’s harder, and my guess is for energy healing (the third example) it would be harder still.
Then there are other at least somewhat ‘woo’ things where I’ve gotten a bunch of value, far more than with tarot, that he doesn’t namecheck, but that’s cases of IYKYK. I do think ‘try things’ is a good principle if you’re not susceptible to falling into traps.
Variously Effective Altruism
Ulisse Mini is on point here: It’s grifting and should feel like grifting if you ask someone for money inauthentically or in a way that the person giving you money would regret if they had the full story. If you genuinely only want them to say yes if they would benefit too, then is fine and it should feel fine.
William MacAskill reinvents the idea that there is value in diversity of experience, calls it the best idea he’s ever had. Total Amanda Askell victory.
The Lighter Side
Experience life like Ete Oaks (46 second video).
Is our children savings?
Yeah, basically.
Neither am I.