Monthly Roundup #11: October 2023

Link post

It never stops. I’m increasingly building distinct roundups for various topics, in particular I’m splitting medical and health news out. Let’s get to the rest of it.

Bad News

A simple model of why everything sucks: It is all optimized almost entirely for the marginal user, who the post calls Marl. Marl hates when there are extra buttons on the screen or any bit of complexity is offered, even when he is under zero obligation to use it or care, let alone being asked to think, so everything gets dumbed down.

Could companies really be this stupid, so eager to chase the marginal user a little bit more that they cripple the functionality of their products? Very much so, yes, well past the point where it makes financial sense to do so. The metrics form a tyranny, invisible costs are increasingly paid on the alter of visible DAUs and cost of customer acquisition and 30-day retention, and that’s that.

What is to be done about it? My proposed solution is to build interfaces, filters, recommendation engines and other such goodies on top of existing sucky products, probably involving the use of LLMs and other AI in various ways, to make the sucky products suck less. In many cases this seems super doable. With the rise of AI, the data you would gather along the way would potentially pay for the whole operation. I continue trying to make this happen low-key behind the scenes.

Periodic reminder from Patrick McKenzie that your phone number with any major American carrier can and will be compromised at a time not of your choosing if someone cares enough to do that, as happened recently to Vitalik Buterin. Socially engineering a store employee is a rather trivial task. So if you care about your security, you need to avoid letting anyone use your phone for two-factor authentication or otherwise plan to be fine when this happens.

Hasan Minhaj admits that he made up a lot of the key details he uses in his stand-up, in ways that greatly alter the serious impact of the story, not merely modifying for comedic effect. Eliezer says this is sad, as he knew journalists did such things but expected better of a comedian. Robin Hanson confirms that it matters via a poll.

Robin Hanson: “Does it matter that much of it never happened to him?” Apparently yes, it does matter.

Hasan Minhaj has talent. His joke construction and delivery is spot on, despite a constant struggle with the axes he is constantly grinding. Now we know that he was cheating with the axes, which makes it much worse. Indeed, despite claiming he only lies in his stand-up, in a real sense his comedy was genuine the whole time, but he felt the need to mix it with deeply dishonest journalism.

The concept of lightgassing, as proposed by Spencer Greenberg: Affirming someone’s known-to-be false beliefs or statements in order to be supportive (or, I would add, to avoid making them angry or incur favor, which is also common). As Spencer notes, the key is often to validate someone’s feelings, without validating their false beliefs. Having a name for this might be useful, so people can request others avoid it, or explain why they are not doing it.

Disunity

Unity is a highly useful game development tool. If you program in Unity, the result will work across a wide variety of platforms. Emergents TCG was programmed in Unity, which solved some of our problems without creating any new ones.

Then Unity decided to retroactively change its pricing to make itself prohibitively expensive to small developers. Including removing the GitHub repo that tracks license changes, and updating their license to remove the clause that lets you use the TOS from the version you shipped with, then insisting already shipped games pay the new fees. Whoops.

Here’s Darkfrost on Reddit, first indent is me, second is them:

Darkfrost: After their previous controversy with license changes, in 2019, after disagreements with Improbable, unity updated their Terms of Service, with the following statement:

When you obtain a version of Unity, and don’t upgrade your project, we think you should be able to stick to that version of the TOS.

As part of their “commitment to being an open platform”, they made a Github repository, that tracks changes to the unity terms to “give developers full transparency about what changes are happening, and when”

Well, sometime around June last year, they silently deleted that Github repo.

April 3rd this year (slightly before the release of 2022 LTS in June), they updated their terms of service to remove the clause that was added after the 2019 controversy. That clause was as follows:

Unity may update these Unity Software Additional Terms at any time for any reason and without notice (the “Updated Terms”) and those Updated Terms will apply to the most recent current-year version of the Unity Software, provided that, if the Updated Terms adversely impact your rights, you may elect to continue to use any current-year versions of the Unity Software (e.g., 2018.x and 2018.y and any Long Term Supported (LTS) versions for that current-year release) according to the terms that applied just prior to the Updated Terms (the “Prior Terms”). The Updated Terms will then not apply to your use of those current-year versions unless and until you update to a subsequent year version of the Unity Software (e.g. from 2019.4 to 2020.1). If material modifications are made to these Terms, Unity will endeavor to notify you of the modification.

This clause is completely missing in the new terms of service.

This, along with Unity’s claim that “the fee applies to eligible games currently in market that continue to distribute the runtime.” Flies in the face of their previous announcement of “full transparency”. They’re now expecting people to trust their questionable metrics on user installs, that are rife for abuse, but how can users trust them after going this far to burn all goodwill?

They’ve purposefully removed the repo that shows license changes, removed the clause that means you could avoid future license changes, then changed the license to add additional fees retroactively, with no way to opt-out. After this behavior, are we meant to trust they won’t increase these fees, or add new fees in the future?

I for one, do not.

fish993: I recently learned about this concept of the Trust Thermocline. It’s basically where a company is (perhaps inadvertently) relying on consumer goodwill and inertia to maintain their revenue/​sales while making the consumer experience gradually worse over time, until they eventually push it too far, lose all their goodwill at once, and the annoyance with all the negative changes made up to that point is finally enough for most of their consumers to move to a different product. The key part is that once the company has reached that point, they can’t just row back the last change to reverse it – the goodwill and trust is already gone and it’s extremely hard for them to ever regain that.

It was brought up with regard to CA’s handling of Total War: Warhammer and it sounds like exactly what’s happening with Unity as well. A lot of the comments are specifically mentioning that while they themselves are not directly affected, they now don’t trust that Unity won’t pull some other shit in the future and it’s not worth the risk of developing future projects on their engine.

The new fee is not only large, it is on installs, which do not correspond so well to revenue, so the fee can easily bankrupt a game developer. The bigger issue is that trust is completely destroyed. How can you develop using a game engine that can impose arbitrary fees that apply retroactively to games already developed or released, and has shown a willingness to do so?

As Tom Francis says, the size of this particular rug-pulling is not the point.

Tom Francis: Don’t often weigh in on the clusterfuck of the day, but in case it needs saying: @unity rug-pulling a new fee structure on devs is an astonishing scumbag move.

Not because of how much it shakes out to, but because a partner who can and will change how much of your revenue you owe them *after* you’ve made and released your game needs to be avoided like the plague.

I hadn’t realised they even legally could. I gather Epic’s Unreal license is perpetual per-version, so whatever the deal is when you commit, you’re guaranteed that forever if you’re happy to stick with that version.

Their 5% is more than Unity’s new fees in most scenarios, but compared to a company that can help themselves to any amount of your money at any time, suddenly sounding more reasonable.

[links to his post on this]

Unity is saying that 90% of customers won’t be impacted, they won’t count re-installs, and clarifies the fee won’t apply to installs that happen before 2024. The fee seems to start off at $0.15 per install beyond the minimum thresholds, with a discount for emerging markets.

Wow it is hard to overreach this badly.

SSFF: “We have never made a public statement before. That is how badly you fucked up.”, might be the greatest thing ever written in a press release.

Image

Jakob Wahlberg asks his lawyer, his lawyer says no, companies cannot retroactively change terms, this is not how any of this works, however the practical legal costs involved in fighting this could be prohibitive.

Freya hears a theory: I saw someone theorize that whoever made this decision at Unity is so mobile pilled that to them “an install” is unambiguously just a number from google play or the apple app store I hate how it’s so painfully believable.

Emmett Shear: Oh my god this makes so much sense.

This is not the first such incident with Unity, they mention the 2019 incident. There are also many other incidents in gaming along similar lines. Some company makes a change to allow it to make more money, or control its brand, or what not, thinking it is harmless. The gaming community instead informs them as one voice that this is a massive breach of trust they will never forget, and that is if it is reversed. The most recent other examples involve Wizards of the Coast, which has done this multiple times.

This comes from failing to think about how things will be seen by outsiders and your customers. In every such case, simply talking to a modest sample of such people would have revealed the issue. Sometimes it is not a change I would have personally minded or objected to, but in every case it is a change I can imagine someone else minding quite a bit. The trick is to ask. You do not have to f*** around this decisively in order to find out.

Or in this case, Tom Francis says they knew exactly what was going to happen, their staff warned them, they did it anyway, presumably underestimating the sheer magnitude of the response. Wow.

Tom Francis speculates that they could still fix this, if and only if they not only fully reverse the fee but they also put into the new EULA that you are allowed to keep current payment terms forever if you keep using the current version of Unity, in a way they can’t back out of later. That does sound both necessary and sufficient here.

Otherwise, yeah, I don’t see how you could risk using Unity going forward.

Ban Gain of Function Research

There seems to be a breed of dog that was bread to be more dangerous. As in ‘Half of all XL Bully dogs in Britain descend from ‘Killer Kimbo.’’ Five people have been killed by them in the last two years in the UK. Sounds healthy and wise. People are suggesting we ban XL Bully dogs. The British have now done so over the long term, with various other restrictions in the meantime.

Seems right. We should at minimum be able to agree to ban owning as a pet that which we are intentionally breeding for unpredictably killing those around you. I would also be so bold as to suggest banning the breeding process itself, as well.

But others disagree.

Daniel Eth: Virologist logic be like:

RandomSpirit: Breeding dogs to be more dangerous is good. In fact, it should be federally funded. This helps us understand how dangerous dogs can be and develop mitigation strategies if dogs like this were created naturally.

The AI metaphor can be whatever you would like.

Tyler Cowen supports the ban for the UK, where the dogs have become a symbol of aggression, but says the case is weaker in the USA, where we have lower population density and more tolerance of risk and violence. I agree it is weaker in the sense that we are stupid about such things with talk about ‘our freedoms,’ which Tyler rightfully rejects as not a meaningful objection here. As he notes, a tax would not be sufficiently de facto enforced, or I would note be high enough to cover the negative externalities. We already ban a lot of far less dangerous pets.

Dog lovers should support such a ban most of all. Allowing actively dangerous dogs ruins it for everyone. It is not reasonable to expect me to differentiate whether your dog in particular is harmless, or set a policy that relies on that distinction. And if someone has even one bad experience with any dog, they’re likely going to be poorly disposed on all dogs for a very long time.

Meanwhile, here in New York, we cannot even do full genetic screening and embryo selection for traits that are actually good, because ‘ethicists’ say it would be ‘eugenics.’

The Right Price to Be Willing to Pay is Not $0

Elon Musk floats charging a tiny amount to use Twitter, massive numbers of people say on Twitter they would leave in response, defend this partly by saying massive numbers of people would leave in response, more commonly simply saying millions of hours for arguing online but not one cent for tribute. Some select quotes:

Dave: For a penny a year, I’d leave my account active but I wouldn’t do anything with it. Can’t be worth less than zero, I wouldn’t pay to delete it, either.

Anton: If I would not have joined if an account had cost X, I will not stay if an account costs X. I don’t like how network effects encourage monopolism.

Sam Black: Worth more than 1, but I’d be willing to pay exactly 0.

Also many comparisons to other services or products, looking at relative rather than absolute value. People seem deeply confused about what Twitter is worth to them.

My humble suggestion remains that if you would not pay a nominal amount to use a service where you spend hours of your life each week, your time is valuable so you should – for your own good, mind – delete your account now.

That is not only about Twitter, or even only about online services. Think about anything you sink your time into on a regular basis. If it cost a nominal amount, would you stop spending your time there? If so, why are you spending time now?

Opportunity Knocks

The Jane Street interns, with link to application. Becoming a Jane Street intern is highly recommended to anyone who would plausibly be one, but not beyond that.

Government Working

New Mexico Governor Michelle Lujan Grisham violates her oath of office, enacts blatantly unconstitutional declaration of emergency to say that gun violence is a public health emergency and so she can alter gun laws however she likes. That is not how any of this, technically speaking, works. [EDIT: A commentor wants me to note that officials quickly indicated this illegal order would not be enforced.]

As a reminder:

Now that you have once again enjoyed my Ted Talk, what are we going to do about it?

Always remember that government can always be worse.

Pradyumna: You a reasonable person: the city should encourage carpooling to reduce congestion

Bengaluru’s Transport Department (a very stable genius): Taxi drivers complained and so we will ban carpooling.

Some Canada school libraries de facto excluding all books published before 2008 in the name of an inclusivity review, resulting in mostly empty shelves. Seems like there are increasingly robust efforts to destroy school libraries coming from all sides, usually with a ‘oh we did not intend them to respond that way, that’s not our fault, we only wanted to make sure all the books conformed to our preferences.’

Google employees told not to say things that could be taken out of context to make them sound bad in an anti-trust suit, says government taking things out of context to make them sound bad in an anti-trust suit.

TSA once again caught stealing from luggage.

Flying taxi executive Michael Whitaker nominated to head the FAA. Promising.

Worthwhile Canadian Immigration

Tyler Cowen proposes that American immigration policy is not as bad as it looks because Canadian immigration policy is much better, and we are then much more generous with immigration from Canada, including the children of immigrants to Canada.

Tyler Cowen: Keep in mind that Canadian immigration policy also is U.S. immigration policy. Currently Canada is taking in about 400,000 people a year, with plans to raise that to 500,000. At the same time, the U.S. is doing little to boost high-skilled immigration. But Canada is serving as a kind of farm system for the U.S. The very best Canadian arrivals — or their children — have the best chance of getting into the United States, if only through O-1 visas.

So if the quantity of Canadian immigration is going up, the quality of U.S. immigration is going up too, through Canada in this case. Call it cherry-picking if you like.

Not surprisingly, neither side in the immigration debate wants to scream this loudly from the rooftops. The pro-immigration side wants to present the status quo as dire. The immigration-skeptical side does not want to stress that there are perfectly good ways of screening for immigrant quality, some of which already are in place.

David Dennison (in the comments): Unless the data here are incorrect, roughly 100 to 500 O1s are approved from Canada each year. So unless you also wanted to consider other visa’s like the TN or the H1B, these numbers are even less than “slightly” https://​​visagrader.com/​​visa-approvals-and-refusals/​​O1/​​canada-can

Wilbur: Most highly qualified Canadians work in the US on TN (later converted to H-1B for ease of applying for a Green Card).

O-1’s are the domain of the very few. That said, many of my Canadian colleagues working at august institutions did come on an O-1, but O-1s are offered to citizens of every country not just Canada.

TNs however are only for Canadians and Mexicans (TN-1 is Canadian, and TN-2 Mexican).

Are we using Canada partly as a filter here? Surely yes, and the O-1 is dramatically underutilized, but also the O-1 is dramatically underutilized and the percentage of even the best who end up ultimately making it to America is and will remain quite low. This is a drop in the bucket.

The number of immigrants Canada is accepting is impressive. If properly filtered, supported and assimilated, ensuring the infrastructure is in place to support the growing population and that norms and culture remain protected, it is potentially a great move. It is not quite the full blown ‘Canada stops collecting taxes because it will grow so much faster than its debts’ gambit.

The problem is that Canada is not a YIMBY paradise that is carefully planning how to incorporate its new citizens. There are many dimensions where trouble looms.

For contrast, here’s the chart of United States permanent immigration requirements to see if you are maybe eligible to immigrate. Actually less complex than I expected.

Image

One dynamic reason why we can’t have nice things, think general case.

Matt Bruenig: Whenever a non-lefty Dem office contacts me for a policy idea, I usually give them the idea of eliminating the income and asset test for SSI for disabled children. Small policy. Obviously sensible. Reception is usually good but, I’m told, the disability orgs say no. Very sad!

I guess the reason the other disability orgs say no is because they want to prioritize increasing the SSI asset test to $10k or something like that. That they’ve been trying to do this for a long time and never succeed doesn’t seem to matter.

People Don’t Do Things

Americans continue to increasingly not do things.

Image

The political spectrum distribution is interestingly asymmetrical. Conservative Republicans and Moderate Democrats do a relatively good job of socializing and doing activities together. Liberal Democrats, Independent and Moderate Republicans do worse. I can come up with an obvious just-so story, but I would not have expected Moderate Democrats to be a peak here, so it would be post-hoc.

The bigger issue is that this is not much socializing or doing activities, even in the categories doing relatively well. We need to figure out what to do about this. Worshiping together is an example of something that there are good reasons to want to move on from, but if you do not substitute other secular activities in its place then you lose a lot of important other benefits, in addition to the impacts inherent in that particular activity for better and for worse.

Here are some people far more in favor of not doing things:

Stanley Pignal: 41% of French population is in favour of a proposal to limit everyone to 4 flights in their entire life.

59% of 18-24 year-olds agree.

2 return flights. Not per year: per life.

Thinking is Hard

People often really are hopeless. Alex Tabarrok notes, from a new paper in Cognition from Meyer and Fredrick, the story of the ball and the bat from the cognitive reflection test. As they say, most people get this wrong:

  • A bat and a ball cost $1.10 in total.

  • The bat costs $1.00 more than the ball.

  • How much does the ball cost? ____ cents.

Only 77% of people get this right when explicitly told what the answer is and to write it down. If you merely ask them to consider the right answer, it’s only 31%.

This causes explicit cognitive dissonance:

We then asked all participants whether the pair of prices (which they had either been provided with or generated) differed by $1.00. Among those provided with a $1.00 bat and $0.10 ball, only 6% said “Yes,” whereas 76% of those who had generated those same two prices did so

As in, 76% of those who made the mistake – and most people make the mistake – then went on to say that $1.00 and $0.10 differed by a dollar. Ut oh.

Also, yeah, people just do not read:

A bat and a ball cost $110 in total.

The bat costs $100.

What is the difference in price between the bat and the ball? ______

Here, there is no opportunity to misinterpret the second number as the price of the bat, because it is the price of the bat. However, the second number can still be subtracted from the first, and many do that: among 1032 GCS respondents, 56% answered $10.

Le sigh. Seems hard to work with any of this.

Satisfaction Guaranteed To Stay Around 83%

Americans consistently report being satisfied with their personal life.

83% of Americans are satisfied with their personal life, matching the historical average.

– Half of the participants are “very satisfied.”

– Family life, housing, and education rank highest in satisfaction.

Image

Consistent contentment since 1979 Americans have maintained a steady satisfaction rate with their personal lives, ranging from 73 to 90 percent over the past few decades, according to Gallup’s measures.

What’s making Americans happy? It’s not just the general satisfaction that’s notable, but the specifics:

– Family and home: Leading the pack, 66 percent of Americans are very satisfied with their family life, and 63 percent are happy with their current housing.

– Education and community: Over half are content with their education (53%) and their community as a dwelling place (51%).

– Jobs: The employed populace showed a majority (54%) being very satisfied with their work. On the flip side, areas like leisure time (43%), personal health (41%), standard of living (37%), and household income (30%) didn’t score as high. But, adding those who are “somewhat satisfied”, no area dropped below 71 percent satisfaction.

#Hopescrolling

What about 2008? We had the great financial crisis. Life for many people got objectively far worse. Many people were ruined. Barely a blip.

What about 2020? Covid seems like it was a pretty big deal.

Seems like the life satisfaction rate is uncorrelated with how satisfying are the conditions of life. So my hypothesis is that this 83% satisfaction rate is a relative measure. People are evaluating their lives by comparing them to those around them. When things get worse, people say they are satisfied with less. So this tells us very little about anything that matters. Did you know that only 1% are in the top 1%?

Good News, Everyone

Intervention to reduce lead in turmeric in Bangladesh estimated to show cost-benefit of about USD$1 per daily adjusted life-year (DALY). People were using lead for coloring, and the spice gets ingested, so it is very high leverage to make them stop. There are always caveats. The ones that suggest discounting efficiency are not big enough to make this stop being fantastic, and I pretty much believe this intervention happened.

The questions that I suspect matter are, is this taxing highly valuable state capacity in ways that we aren’t accounting for? And can this intervention be repeated at scale without losing effectiveness, or was this the ideal conditions and it will rapidly get less worthwhile if we branch out?

I do not know. I do know that the use of lead seems pervasive, the impacts of poisoning are cumulative and these are true of many places. Someone should certainly be on this. But also, once it is demonstrated this works, shouldn’t plenty of people be happy to get on this? Lead is a classic villain at this point.

Community Notes on Twitter applies to advertisements as well. As Emmett Shear points out, this is in equilibrium good for advertisers, letting viewers trust that if something was out of line there would likely be a note attached, increasing value on all sides and letting honest ads outcompete dishonest ads.

Cultural Tutor had 100k Twitter followers and was training for the British Military after working at McDonalds, David Perell gave him a stipend to get things off the ground and writing a thread per day, now 114 months later there’s 1.5 million followers and growing fast.

This is one of the most underappreciated moves out there – find someone who is creating something you love and want to have a bigger impact on the world, and pay them what they need to be able to focus on it.

Potentially largest Lithium deposit in the world discovered in a volcanic crater in the United States, 20 to 40 million tons.

In the meantime, Li-Ion batteries have already reversed their 2022 price increases and are scheduled to hit new lows. Turns out that yes, markets solve this and keep solving it, why do people keep not expecting this.

Jeff Bezos makes the case (1:18) for meetings based on memos you read at the start, rather than PowerPoint, which Amazon banned. I mostly buy this for interactive meetings. For presentations, especially to be rebroadcast, you may not have that choice.

Contrast this with the Tobi Lutke meeting strategy of identifying a problem and proactively taking the status quo off the table, and insisting the problem be solved. Something must be done. Therefore we must do it. It must proclaim to solve the problem. This seems like an excellent way to railroad your something through, whether or not you are indeed open to a superior alternative something. Or more generously, when there are clearly good plans available, and you want to avoid the enemy of a perfect plan tomorrow. Or when what matters is making some damn choice at all. So there’s a time and place for this, but you still have to worry about the obvious epic failure modes.

Guzey’s 100 lifehacks, many of them quite good, many others of the form ‘not something you should often actually do but something you should often ponder.’ And near’s top 20 from that list.

Guzey’s lifehacks, edited by @nearcyan to show only my favorite 20

1. what are you being a coward about? be specific

2. seek ground truth and poke reality. don’t settle for proxies or for winning arguments

3. send cold emails. assume that everyone is a friend

4. lots of alpha in low status

5. we become the people whose opinion we care about

6. we become the people we spend the most time (physically and mentally) with

7. never give up

8. do something courageous

9. feeling stupid now is better than feeling stupid in 10 years

10. “should” sus

11. everything is power-law distributed

12. you haven’t failed until you’ve given up

13. when a prisoner at auschwitz was trying to escape the gas chamber, other prisoners alerted the guards 14. never make decisions on the basis of tax rates

14. never make decisions on the basis of tax rates

15. if you have more than a few months of runway, start spending more

16. spend 10-20% of your time reorienting

17. ask the girl you like out

18. if you feel bad, low energy, depressed, demotivated, there’s probably a reason & there’s probably a solution

19. the smarter someone is the more they can afford to have terrible epistemics and still be successful

20. risks are less risky than you think. take as many as you can.

Mostly this is great. My main disagreement would be #7/​#12, the concept that one should never give up and you only fail by giving up. Not so. It is very important to know when to quit. My biggest mistake as a founder was not knowing when to quit, and I very definitely had failed well before I did so.

Also #14, which is true on many margins but very, very not true in general.

Otherwise, endorsed. Full list not as reliable, but quite good.

Incidentally, when I asked Bing to transcribe the list from an image, it instead generated a list of 20 completely different generic pieces of life advice.

The Ancient Art of Getting Through a Conversation

Don’t know what to say? Can’t handle five minutes on astrology or the Roman Empire? Impress with this one weird trick.

Jakeup: Guys, do you know enough about astrology to survive a 5 minute conversation? Girls, same for the Roman Empire?

It’s not that I do not know enough about astrology to survive a five minute conversation. It’s that I know too much about astrology to survive a five minute conversation. So what to do?

Visakan Veerasamy: one of my best ‘party tricks’ is that I can have an interesting conversation with any subject matter expert about anything for an hour despite me having 0 prior knowledge of the thing. core thing is to ask earnest “I’d love to learn more” questions

5 minutes? just ask the person how they got into the thing that they’re into. when did you first get into astrology /​ neuroscience /​ norse mythology /​ mongolian throat singing? How did that happen? Wow, no way!! And then? You’re kidding. What?? Yea I see it. Damn! How else?

What do newbies like me typically get wrong about X? Why do think that is? What’s your favorite thing about X? Why? What’s the biggest source of drama in the X community right now? are you on either side or do you see both merits? etc etc. 3 hours have gone by.

Do you want three hours to go by here? Which way, modern man?

The good news is it leaves a good impression.

Patrick McKenzie: A weird thing about this is that people often remember the curious party as being much sharper on the subject than the text of the conversation actually reflects.

Not to spread dark arts buuuuut that’s a useful thing to know if e.g. you are doing a job interview.

(A lot of good candidates don’t have a question at the ready for interviewers and if that is you then just ask the interviewer to monologue a bit about recent projects. Bonus: will be most honest window into culture/​org you get during interview.)

Do you have to mean it? Or can you do a good enough job of faking that?

Money Stuff

(Reminder: If you are not signed up for Matt Levine’s Money Stuff, the One True Newsletter, you are missing out.)

Patrick McKenzie reviews Number Go Up, a book about the massive fraud that is Tether. Recommended if relevant to your interests. He calls it the best book about crypto, but is frustrated that the book did not go farther and uncover more of the most important crimes. Patrick warns that we were on the verge of having our civilization’s core institutions subverted by fraudsters. The main place we differ is that we are on opposite sides of the ‘will Tether de-peg?’ market. No matter what they did in the past, I now see a 5% safe return as creating such a good business that no one will doubt ability to pay. Sometimes they really do get away with it, ya know?

Patrick McKenzie points out a key problem with containing fraud. If you notice you’ve lost money in a fraud, you can face punishment and loss of face now, or you can lose orders of magnitude more money to postpone that reckoning, perhaps indefinitely. So it is key that we give people who realize they’ve been had a way out that saves face, or at least minimizes face loss. We can start by doing this locally – if someone bails and admits they’ve been had, be as kind and non-punishing as possible.

Oh no.

Image

Songpinganq: Xi Jinping is hiring 87,000 agriculture police officers, and they’re coming for Chinese farmers. Farmers can only plant the crops the government approves. Today officers arrest this family and destroy all their Mango trees——the government wants this family to grow sugar cane (has video of event at link).

Oh no again.

Joe Weisenthal: Chinese police detain Evergrande employees after financial arm of indebted company fails to pay investors.

SCMP: Evergrande Wealth Management staff detained in Shenzhen, but it is unclear what charges they face or if general manager Du Liang is among them.

It would lead to some highly interesting effects if China has decided that employees can be held criminally liable when their company goes bankrupt. Patrick McKenzie is going to have to reload the international popcorn reserve.

EU may ban payments for breast milk, sperm and blood. So, very little of that for those in the EU, then. Does seem very on EU’s brand.

Strikes also destroy value, and are thus a failure mode where negotiations have broken down. Unions and management should be able to look ahead to the strike, see the result, and agree to a compromise without a strike. Which mostly they have done for many years now in America, the UAW strike is an unfortunate blip.

Image

It is kind of crazy that we used to lose over a day a year to this failure. In many other countries, it seems people still do, as it is seen as virtuous to strike and as likely to lead to superior outcomes to show them periodically that you mean business. Or perhaps general strikes are more about using unhappiness as a justification for time off. There’s that, too.

Dan Luu asks why private equity doesn’t take over tech companies that refuse to actually do the things that maximize profits and have lots of worthless unnecessary staff, when they could often make huge profits by firing a bunch of worthless unnecessary staff and have the remaining engineers do things that make you money, or otherwise run a sensible business? Instead PE firms try to milk companies for value but get so greedy and short term and stupid about it that we seem to only get things like VerticalScope, a PE firm that seems to buy up internet forums and then quickly destroy them, which turns out not to be a good business model?

A potential hint is Dan’s observation that everyone expects PE takeovers to result in butchery and death, and they react badly to them. So if you are PE and take over, your customers and other key players will flee in anticipation of that, which destroys too much value, so perhaps you don’t try. Or perhaps it is one of those cases where the people who can pull off the takeover part are not also capable of understanding how to run the company. So you can see transformations that look like what you’d see in a PE takeover, but they only happen without PE involvement? It’s weird.

Canadian Prime Minister Trudeau threatens that if supermarkets can’t find a way to stabilize food prices, ‘we will take further action and we are not ruling anything out, including tax measures.’ I mean, yes, providing tax incentives to the supermarkets would work I suppose. Threatening them with taxes if they don’t lower their prices would be… less effective. An obvious possibility would be to break up the big chains into their various banners, to provide more competition, or to offer subsidies for new upstarts.

Claim that US manufacturing sector needs to up its game, and this is an opportunity.

Zack Kanter: I tried to make products in the US when I had my auto parts business. US manufacturers only wanted to do 1-2 pieces of the process – forging but not machining, machining but not heat treat, heat treat but not packaging. Taiwanese manufacturers offered to print my business cards.

The general attitude I experienced with US manufacturers was that I was lucky that they were considering taking me as a customer. There is a tremendous opportunity for new US manufacturers who are hungry and well-run.

As Tim Cook points out in the clip at the link, China is not the low wage destination anymore. Taiwan certainly is not. America’s issues go far beyond high wages.

He was always a businessman who loved the art of the deal. We could have made a deal. What would it have cost to get Trump to not run for President? It is claimed, admittedly not by the most reliable sources, that the floated price – to Sam Bankman-Fried, of course, because who else would have had what it takes to actually do it – was $5 billion. That sounds spot on to me. What a deal! Enough to change Trump’s fortunes quite a lot such that he couldn’t turn it down, still far smaller than the value him running would plausibly destroy. Presumably this far down the line, the price will have gone up, but also so has the value. Given how much various people are willing to pay to have a very very small change of impacting whether Trump becomes president, I claim this is a much more efficient approach. Our price cheap. It’s not too late.

Musicians like Bob Dylan who sold their catalogues seem to have largely sold out at the top.

Giving fisherman cell phones changed prices from so wildly swingy that throwing fish out was common, to remarkably stable. Communication and price signals work, as each region in turn got the solution. The improvements in efficiency, and on security of outcomes for the workers, are huge. Markets need information flow, then they work.

Image

Tax rates and tax collection over time.

Rohit: I think about this chart at least once every few weeks

Image

Yes, of course you want to read what everyone else is reading.

Lily: The craziest thing you realize as a big FinTwit account is how many actual serious money managers and traders actually do read the deranged stupidity posted on here daily and take it seriously, including making decisions on it. In general the larger the account, the less value it posts.

Colin Fraser: When I worked in this industry the trader I worked for had one of his 8 computer monitors dedicated to displaying the Zerohedge Twitter feed.

That is how trading works. If others will react to the Zerohedge feed’s nonsense, then you want to know about that. It usually being nonsense does not change this.

What does shock me is using an entire monitor. A Twitter feed should only need maybe a third of a monitor, tops. Those screens are valuable. Also, you don’t need to exclusively do one account there, form a list or something. Terrible screen game.

David Brooks, longtime prominent NYT reporter, pays $78 at Newark Airport for a burger, fries and a scotch, says ‘this is why Americans think the economy is terrible.’ No reasonable not-rich person would order scotch at an airport, but the whole meal is an obviously terrible choice. You can do so much better, if you try to imitate the overpriced generic middle designed to satisfy groups so they can talk while dining alone at an airport, you pretty much deserve what you get. If you want a burger at Newark Airport, the airport has a Shake Shack in Terminal A, which is better in every way. On my recent trip through JFK and SFO, the locally available options were not ideal, but wow is paying $78 a choice.

Very High Marginal Tax Rates

Max Ghenis presents: The California Marginal Tax Rate Schedule (his source), without the childcare and housing subsidies.

Image

Note that the marginal tax rate rises again if you extend the graph to the right.

Image

That first huge drop is a pretty dangerous local maximum. It is very easy to see how someone might get stuck there.

This paper suggests that having high productivity states have higher marginal tax rates than lower productivity states is bad, as it causes people to favor low productivity states. Well, yes, but the same argument can be used against progressive taxation in general, this is hardly the central case.

On the other hand, if you don’t have to pay, the marginal tax rate is kind of 0%? I mean, when has San Francisco not been a disaster area?

Evidence that Opportunity Zones (aka ‘Lower Marginal Tax Rate Zones’) are successfully inducing lots of investment. I wonder if this implies anything.

Having trouble with your taxes? Welcome to 2023, everyone.

Image
Image
Image

I’m almost disappointed that it looks like this isn’t a scam.

So Cold, So Alone

Do we have a loneliness epidemic? In the sense that we have always had a lot of loneliness, definitely yes. In the sense of something new going wrong on net, it is unclear. We spend more time alone. We also prefer more alone time as we get richer and older. Our technology means that when we are physically alone, we are less alone than we used to be. Then again, people could easily be super wrong about whether they should be investing in more time with people, it sure seems like many wealthy people mess this up. So how best to think about this?

Tomas Pueyo: I’m tired of the lazy reporting on loneliness: “We have a loneliness epidemic, it’s making ppl miserable & killing them. Probably due to social media, phones, suburban life, or smthg. Let’s go back to meeting in real life!” No! Yes, we do spend more time alone.

Image

As time passes, we live more and more alone. As we get older, we also spend more and more time alone! And look how many people say they’re lonely!

Image
Image

So that seems like all the lonely people, where do they all come from? Tomas suggests it is because they want it that way.

Narrative violation: Ppl *want to spend more time alone* as they become richer.

Narrative violation: ppl say they’re *less lonely* as they age (and spend more time alone).

Narrative violation: Generations born between 1920 and 1970 in the US all feel the same level of loneliness

Narrative violation: in developed countries, the more ppl live alone, the less lonely they say they are.

Image

That last one is weird, although less weird on a country level. Presumably on an individual level this does not hold?

Narrative violation: US teenagers felt *less lonely* over time between 1977 and 2012

This one I can explain. It is because their parents never leave them be.

Narrative violation: Ppl are committing suicide more than before… mainly in the US!

So what’s happening? Ppl are mixing *loneliness* with *aloneness* We are spending more time alone But this is *not* making us lonelier We like it! There isn’t a loneliness *epidemic*. It’s always been there. We should fight it, but it’s not new, so it’s not due to social media

While I Cannot Condone This

Online sales no longer increasing as a share of all retail, +0% growth recently for online versus +0.6% overall growth. Curious that this aspect of the future is neither advancing nor becoming more evenly distributed.

Customers who seek out discounts are often highly toxic. They are frequently there to get away with something, to game the system, and will not hesitate to complain or impose massive costs on you. Charging higher prices raises expectations, but it also gets rid of the toxic customers.

Patrick McKenzie: The best reason to charge more is to make more money, but the next best reason is that you will price out “pathological customers.” Customers like the one below *are numerous* and higher price points mean you’ll lose less of your brain cells dealing with their many challenges.

“Is ‘pathological customer’ coextensive with ‘poor’?” The great majority of the ones I dealt with over the years were comfortably middle class or above.

Among many, many other things which surprised me and all of my software buddies, you’d be surprised how charging more “causes” your customers to have hard drive failures far less often.

“But hard drives fail approximately randomly, Patrick. That’s why those of us who care about them professionally rate them as having Mean Time Between Failure.”

Yes, and you have never emailed a random software company when one failed, because you are clueful.

And you had backups and/​or staff who were either in charge of backups or in charge of answering routine inquiries like “My hard drive failed and I lost a thing I need; how do I get it back?” without needing to ask customer support to walk you through that six years after purchase.

Kendric Tonn: Relatedly, I learned very quickly running my Airbnb to never, ever offer discounts. Guests shopping for discounts were, next-to-universally, disruptive, destructive, and generally the scum of the earth. I will never hit that “run a promotion!” button again.

This is, incidentally, independent of the actual price point. I might manually drop my price 10-20% if it looks like the BnB is likely to go empty, and that’s fine: just as long as I don’t show up in searches for discounted rooms.

John Champaign: When we were running our AirBNB, any mention of a discount would lead to us blocking that guest. We weren’t offended by the request, but just found guests asking for discounts were bad guests overall. It was a strong signal that we didn’t want to host them.

This matches multiple other experiences of myself and those I know.

As Tyler Cowen notes here, the interesting thing about the fake aliens – and no they are still almost certainly not aliens – is that the media is so wisely almost completely ignoring the fake aliens, not even running ‘Mexican government goes crazy’ style headlines.

Also this:

Robin Hanson: Wow. I look forward to hearing more technical details.

Having now seen more details, I’m now a lot less interested in this.

From June: Plant-based meat sales are declining, with refrigerated options down 20% year over year by volume and frozen options down 8% by volume, tanking the value of Impossible Foods. Beyond Meat Inc. declined more than 90%. Will the underlying technology get there eventually? That could go either way. What this does show is that, after the novelty wears off, the tech is not yet ready for prime time.

The case for creating a Greater San Francisco. I agree this would be good for the entire area. Something about a snowball’s chance in hell.

Words of wisdom.

Patrick McKenzie: There is an exchange rate between competence and power, in both directions, and this discomfits many people.

One specific way: many people who do not desire power and keep piling on the competence will have to deal with the moral challenges of power whether they want them or not.

Minimally, it happens when power takes notice of competence and directs competence to apply itself to power’s wishes.

This is a through line of a LOT of discourse about the tech industry, which built an incredibly concentrated well of competence and was genuinely surprised when it woke up one day perceived by power as an acute threat.

No subtweet rule: this thread was inspired by the dilemma Starlink finds itself in, where they were just really good at moving bits around until someone apparently told them explicitly that bits move bombs and so now they would move bombs around, too.

But there is literally no genre of competence where you will not eventually hit the moral challenges of power.

Yep. Display enough competence and you have a form of power, which means people start ordering you to use it, start being worried about how you might use it, blaming you for not using it the way they would like, trying to take it away from you, and also putting you in charge whether you like it or not.

Twitter poll (among a largely EA crowd, presumably) on willingness to go vegan to impact factory farming. What theory explains all the results at the same time? The implied theory of ethics is strange indeed, especially given that this is the result despite anchoring on previous answers. Also lots of people are lying, of course.

Arnold Kling speculates, in essence, that race relations are worse because we believe race relations are worse. If you are constantly worried about perceptions of horrible race relations, or actual horrible race relations, you will have difficulty relating well to other races.

The following story definitely checks out.

Ben Reinhardt: There’s a fuzzy, unnamed, but obscenely important skill that keeps coming up when you dig into successful technology managers across worlds/​disciplines:

Roughly, it’s the ability to know how hard something is – Whether someone is slow because they’re slacking or it’s impossible.

This skill also translates into what you can promise, and what to ask for or wait for versus what to do without. And it especially feeds into design, where you don’t need to know how to implement the design but you do need to know exactly how your design decisions impact degree of difficulty of implementation.

When working on Emergents, it become very clear how valuable this skill was, and how expensive it was to not have it, depending on how it shifted who was effectively making various decisions.

Queue Technology

How can we modify queues to waste less of our time, and let those most willing to pay go to the front? Mostly you can’t. The queue that people respect is an astonishingly powerful and valuable cultural innovation, and it comes with everyone involved thinking it is fair. Any solution has to deal with complexity and perceptions of fairness. No one wants to deal with complicated bids.

The best known alternation is a variation on the Disney Fast Pass, where those who pay a premium get to stand on a second line. Similar systems work for boarding airplanes.

One simple proposal is letting places on line be sold, you pay to swap positions so no one else loses out. There is, alas, a big problem.

In a static situation, that is a tradeoff between preserving the sacredness of the queue plus transaction costs and imposing thinking and math and monetary questions on people in exchange for efficiency of results.

In a dynamic situation, once you can sell your spot on a line, that spot becomes profitable. There is some hourly rate that you can expect for standing in that line. So more people do stand in that line. Those unwilling to pay are indeed worse off, perhaps much worse off. In the extreme, normal people would have no choice but to pay, while lots of time was wasted.

This is not theoretical. When the award from the line can be resold, such as in-demand concert tickets or new video game systems in short supply, what happens? Everyone queues up at large cost, often overnight or more. Then they resell. Some fanatics still get in line for themselves, because transaction costs, taxes and sacredness and all that, but the lines are vastly longer than similar situations when the products cannot be resold.

Gamers Gonna Game Game Game Game Game

YC held a 1.5k premodern Magic tournament. I couldn’t play the whole thing, I did get to check it out for a few rounds. Deck diversity wasn’t quite where I’d like but still very good for a format this stable and old. People remain quite bad at Magic, alas myself included. Elves in particular seems like it offers large opportunity for much better play.

Here are some decklist pics.

Tom: Top 4 of the NYC 1.5k hosted by @BifrostGamesNYC w/​ Goblins.

Decklist by the master @fpawluszmtg

Image

David Kaplan: T8’d Premodern event at @BifrostGamesNYC. Started 4-0 vs. Sligh x 2, GW Slide and another deck, drew with @Llano_world to 4-0-2, 1st seed. Great T8 match against @i_b_TRUE on Goblins. Thanks to @mtgbanding, @LannynynyMTG, Jeff and Jordan for hosting and @fpawluszmtg for the deck!

Image

As I understand it, Dreadnaught is the best deck, but it is highly beatable so it is fine. What I do not understand is leaving home without Wrath of God in the board. I get that you kind of think you do not ‘need’ it but so what? Seems so much better than Last Breath. I do love Armageddon.

Llanoworld Elves: Drew into 2nd after Swiss with 4-0-2 on NO Elves in the NYC Order of the Sacred Torch 1.5K. Thnx to @fivewithflores,

@fpawluszmtg, and Demian Vernieri for the list. Thnx to @BifrostGamesNYC, @mtgbanding, @LannynynyMTG and Jeff for running everything. Fell in QFs to Enchantress.

Decklist:

Main Deck
4 Llanowar Elves
4 Fyndhorn Elves
4 Quirion Ranger
4 Wirewood Symbiote
4 Multani’s Acolyte
4 Priest of Titania
1 Yavimaya Granger
2 Deranged Hermit
1 Druid Lyrist
2 Wall of Roots
1 Squee, Goblin Nabob
1 Anger
1 Masticore
1 Nantuko Vigilante
1 Kamahl, Fist of Krosa
4 Survival of the Fittest
2 Natural Order
1 Verdant Force
4 Wooded Foothills
4 Gaea’s Cradle
1 Mountain
9 Forest

Sideboard
3 Naturalize
1 Reverent Silence
1 Seeds of Innocense
2 Tormod’s Crypt
1 Stabilizer
2 Natural Order
1 Phantom Nishoba
1 Uktabi Orangutan
1 Goblin Sharpshooter
1 Caller of the Claw
1 Genesis

I am pretty sure I understand all the decisions this list makes. Most of it seems fine.

Ari Lax writes on how he avoids getting unintentional draws. Get your mechanics down, do your thinking in advance, know your stock plays, spot and discourage slow play calling judges as needed, scooping lost games once they are clearly lost if time is a factor, the usual stuff, all extra urgent now with 50 minute rounds and cards with four times as many words. Finishing your matches is a key part of winning tournament play.

Magic attempting to bring Standard back. They’re giving it a $75k open, making it the constructed format of a PT, providing promos for a weekly local game store standard showdown. Like Matej Zatlkaj here, I find this disappointing. Once a format like Standard gets dislodged, it takes a lot to bring it back due to the high costs that players can’t later reclaim. Standard is not cheap to keep up with, or to start, especially without effort being put towards creating competitive budget decks. Even if the format plays well, we’ll need much heavier support, and Wizards has to truly want it. Otherwise, players will keep playing Commander with a side of other more evergreen formats.

Jenkins Gage worked for three years on the game Kolor (Steam link), without knowing the first thing when he started out, with no connections.

Tyler Glaiel: in case you’re wondering why the classic game dev advice is always “make and release a bunch of small projects first before jumping into a big one” this is why

Morgan K: This person went from zero dev skills and experience, to a shipped 3D action roguelike in 3 years. If you reframe your expectations and goals just a little bit, this is a wild success. Don’t let capitalism be the only thing motivating you to make art.

Given that, he did an amazing job. Unfortunately, he had issues that generated two negative reviews out of the gate, and the game had and still has a bunch of severe problems because man this stuff is hard. Despite all that, I enjoyed the brief time I spent playing, and then I talked to him and tried to help him improve it, because that seemed like the thing to do here.

I’m not going to pretend this is a game I would recommend in a typical context, but if you are willing to forgive its failings under the circumstances and want to help the guy out, it’s only $10, and it will improve over time if he keeps at it.

Lan D. Ho: Played this along w/​@fivewithflores@ the Premodern 1.5k @ @BifrostGamesNYC: lost r1 to @m4n4_st4tion, who grinded me out w/​Auramancer • Astral Slide • Seal of Cleansing & r2 to Mike Harris on Hermit FEB, which is really tough, & that was my tourney. Still like this deck a bit

Image

All right, that’s bizarre. I have no idea what to make of this.

High stakes Scrabble at the highest level. A fun read. I do think Scrabble requires too much technical understanding to be a good watch if you aren’t pretty into it.

Larry David suggests getting rid of goalposts (1:30) in classic Larry style.

I, too, am shocked, shocked, well not that shocked at the staggering incompetence of Washington DC’s sports betting rollout. It is not easy to be this terrible.

Patrick McKenzie recommends Spell Disk, an action roguelike currently in early access where you set up various attacks that trigger off different conditions to create an engine of destruction. I checked it out during my trip to Manifest. I found it to be a lot of fun, with a number of mechanics worth exploring. Its core issues are that the balance between items is deeply awful, and that the game’s mechanics and character builds are all about offense but the game is at core about learning patterns and how to not get hit by things while your death program does its thing. Also it needs more variety of enemies especially bosses, and ideally also more items, all of which I do think is coming. They added on a Vampire Survivors-style mode, which can be fun but does not seem well balanced. I expect the final version to be a Tier 3 game, in its current state I’d still have it as Tier 4.

Orb of Creation is a stunningly engrossing idle-style game. Jorbs gave it the Do Not Play label, which is appropriate. Exponential growth is fun, continuous new resources are fun, figuring out the fastest advancement path is fun.

Europa Universalis 4 finally gets its big Byzantium update. Good times, man.

The Lighter Side

I did not realize this was possible but yes, you can create a worse airline than Spirit, and its name is Jetstar.

Dope testers show up at the Delhi State Athletics Meet, most athletes bolt, only 1 competes in 100 meter final.

ONLY ONE participant in the final of the men’s 100 metres; a steeplechase athlete who kept running after crossing the finish line to evade dope testing; several winners skipping the medal ceremony as they were worried about being asked to give samples.

Meanwhile, empty packets of Recombinant Human Erythropoietin (EPO) injection, which is used as a performance enhancing drug, could be seen in the washroom on Tuesday.

Ms. Lincoln reports the play was lovely.

Woman who is not wrong (0:12).

I never thought about how Amazon can deliver packages faster than it can apply your Kindle credit.

Byrne Hobart: Putting different priorities on different details of the business means that Amazon can commit to physically delivering an iPhone charger to your door faster than they can commit to updating a table in a database.

Image

Reading is important.

Image

So is listening.

Image

Besides, is he really thinking about ancient Rome? (0:52).

Peter Wildeford instead thinks about this a lot:

Image

Jessica Livingston: Work/​life balance, 2009.