Links
3 stars
Against Learning From Dramatic Events | Astral Codex Ten
Perhaps I particularly like this because it addresses one of my intellectual pet peeves:
It’s bad enough when people fail to consider events that obviously could happen but haven’t. But it’s even worse when people fail to consider events that have happened hundreds of times, treating each new instance as if it demands a massive update.
This is how I feel about mass shootings. Every so often, there is a mass shooting. Sometimes it’s by a Muslim terrorist immigrant. Sometimes it’s by a white person (maybe even a far-right racist white person). Sometimes it’s by some other group entirely.
Every time, people get very excited about how maybe this proves their politics correct. “I bet everyone thought this latest shooting was by a Muslim. But actually, it was by a white person! This just goes to show how everyone except me is racist!” Or “I bet everyone thought this shooting was by a white racist! But actually it was a Muslim! This just goes to show how everyone except me has been poisoned by political correctness!” […]
The problem with the US response to 9-11 wasn’t just that we didn’t predict it. It was that, after it happened, we were so surprised that we flung ourselves to the opposite extreme and saw terrorists behind every tree and around every corner. Then we made the opposite kind of failure (believing Saddam was hatching terrorist plots, and invading Iraq).
The solution is not to update much on single events, even if those events are really big deals.
2 stars
GovDocs to the Rescue! Debunking an Immigration Myth | DttP
One question that routinely comes up in genealogy research: why is the family’s surname different from its (presumed) original form? Most people have heard one explanation: those names were “changed at Ellis Island,” altered either maliciously or ignorantly by port officials when the immigrant passed through. The charge against immigration officials, however, is provably false: no names were written down at Ellis Island, and thus no names were changed there. The names of arriving passengers were already written down on manifests required by the federal government, lists which crossed the ocean with the passengers. Records kept by the government demonstrate conclusively that immigrants left Ellis Island with the same surnames they had arrived with. The idea that names were changed at the point of entry is a myth, an urban legend promoted by a popular film. Changes were made later, by the immigrants themselves, usually during the naturalization process.
Techno-optimism for 2024 | Noahpinion
One technology that I’m consistently hyper-optimistic about is batteries. This is because batteries represent something humanity has never really had before — a way to store a large amount of energy and carry it around.
The storage aspect of this, just on its own, is incredibly important. Increasingly, people are using batteries as the solution to smooth out the intermittency of solar and wind power, providing stable electricity supply through night, rain, and even winter. Around the world, battery storage for grid power is growing exponentially. The EIA expects it to nearly double in the U.S. [in 2024].
A new mRNA vaccine for malaria promises to help conquer humanity’s most deadly chronic illness. And a new mRNA vaccine for pancreatic cancer — which you actually take as a therapy after you’re diagnosed, not before — promises to save a lot of lives from one of the deadliest cancers. Other cancers are probably also treatable this way.
Another technology that’s just now coming to fruition is genetic engineering, especially via CRISPR. The FDA just approved a drug that actually goes in and reengineers the cells in a living human body to treat sickle cell anemia. Other such drugs are in the pipeline.
Huge ancient lost city found in the Amazon | BBC
A huge ancient city has been found in the Amazon, hidden for thousands of years by lush vegetation.
The discovery changes what we know about the history of people living in the Amazon.
The houses and plazas in the Upano area in eastern Ecuador were connected by an astounding network of roads and canals.
Xi, Biden and the $10 trillion cost of war over Taiwan | Bloomberg [via Yahoo!]
War over Taiwan would have a cost in blood and treasure so vast that even those unhappiest with the status quo have reason not to risk it. Bloomberg Economics estimate the price tag at around $10 trillion, equal to about 10% of global GDP — dwarfing the blow from the war in Ukraine, Covid pandemic and Global Financial Crisis. […]
Bloomberg Economics has modeled two scenarios: a Chinese invasion drawing the US into a local conflict, and a blockade cutting Taiwan off from trade with the rest of the world. A suite of models is used to estimate the impact on GDP, taking account of the blow to semiconductor supply, disruption to shipping in the region, trade sanctions and tariffs, and the impact on financial markets.
How to be More Agentic | Useful Fictions
Things you learn dating Cate Hall | Sasha’s ‘Newsletter’
Two interesting pieces, best read together; the first is by Cate Hall:
I often hear agency talked about as if it’s an inherent trait: Either someone has it or they don’t -- in which case, too bad, they’re doomed to a life in the minor leagues. This hasn’t been my experience. Over the years, as I’ve gradually grown dumber relative to my peers through a combination of aging and making smarter friends, one of the main ways I’ve compensated has been through dialing up my agency, which I think of as something like “manifest determination to make things happen.”
As a result, I’ve done a bunch of cool stuff in different domains: I was a Supreme Court advocate and the number one female poker player in the world; started art and perfume companies; and led operations at Alvea, a pandemic medicine company I co-founded, when it set the record for the fastest startup to take a drug to clinical trials. All of these things I did in my 30s.
In my way of thinking, radical agency is about finding real edges: things you are willing to do that others aren’t, often because they’re annoying or unpleasant. These don’t always surface in awareness to the point one is actually choosing -- often they live in a cloud of aversion that strategically obscures the tradeoff.
The idea of finding real edges, as contrasted with “eking out wins by grinding harder than everyone,” first clicked for me when I started playing poker. Poker in the modern era is an extraordinarily competitive game, and even 8 years ago pros were spending nearly as much time studying as they were playing, using solver models to seek out tiny mathematical advantages. At the same time, a massive edge was available in the form of physical reads, but almost entirely ignored. (I know an example would make this more compelling, but I’m sorry, it’s like explaining a magic trick.)
This year, I’m going to be married to Cate Hall. Early on in our relationship, I told her that she was a really inspiring person, and that I would keep her in my life indefinitely whether or not we ended up together. I was not just flattering her: I have never met anyone like Cate. She is truly unusual. Someday, her biography will be available at fine bookstores everywhere.
Given that it is not yet available, however, I wanted to tell you some things I’ve learned from being around Cate day-to-day.
Really, Nobody Knows What They’re Doing
Cate’s accomplishments are pretty crazy, especially when you consider that she wasn’t born rich. Part of it is that she’s just smart and hard-working. But that’s not the end of the story—she’s managed to accomplish a lot more than some similarly smart and hard-working people I know, in a dizzying number of fields. […]
I think the crucial difference with Cate is that she has really, really internalized the idea that everyone is winging it, which, luckily, means that you can wing it at everything, too. Like, the people writing symphonies? They’re just flailing in symphony-like ways as best they can. People who run biotech companies? They have just as much impostor syndrome as you do. I think Cate doesn’t allow herself to think she’s not allowed to do things. She said to me once: “it’s so corny, but believing that you can do it really is centrally important.”
AlphaGeometry: An Olympiad-level AI system for geometry | Google DeepMind
In a paper published today in Nature, we introduce AlphaGeometry, an AI system that solves complex geometry problems at a level approaching a human Olympiad gold-medalist - a breakthrough in AI performance. In a benchmarking test of 30 Olympiad geometry problems, AlphaGeometry solved 25 within the standard Olympiad time limit. For comparison, the previous state-of-the-art system solved 10 of these geometry problems, and the average human gold medalist solved 25.9 problems.
Is U.S. industrial policy learning from its mistakes? | Noahpinion
When FDR was running for office in 1932, determined to quash the Great Depression, he promised a policy of “bold, persistent experimentation”. He didn’t really know how to fight a depression — no one really did, because in previous times, fighting depressions and recessions hadn’t been something the government did. FDR just promised to keep trying things until he found what worked.
And that’s exactly what he did. Many of the New Deal policies were missteps, or even debacles — the attempt to create government-sanctioned monopolies in order to raise corporate profits, the attempt to create artificial scarcity in the agriculture industry to boost farm incomes, and so on. To his credit, FDR eventually abandoned most of the approaches that didn’t work. What ultimately remained of the New Deal economic approach after Roosevelt’s death contained only a few pieces of his initial approach; instead, it was mostly made up of more successful ideas that had been discovered along the way, like financial regulation, Social Security, unemployment benefits, and large-scale infrastructure projects. […]
In other words, U.S. industrial policy is running into at least three big sets of problems:
Lack of bureaucratic state capacity
Onerous contracting requirements
Permitting, especially NEPA
The optimistic scenario is that we’ll learn from these failures. […]
But there’s no certainty that the mistakes will get fixed; the government has to be bold and persistent, just like FDR and the New Dealers were. Biden is certainly upset that his signature accomplishments won’t bear fruit by election night, but it’s not clear that the sense of urgency has reached a critical point yet. The hard thing about learning from failures is that you have to want to learn. And I’m not yet sure how badly America’s industrialists want to learn.
The Road To Honest AI | Astral Codex Ten
This is a great new paper from Dan Hendrycks, the Center for AI Safety, and a big team of academic co-authors.
Imagine we could find the neuron representing “honesty” in an AI. If we activated that neuron, would that make the AI honest?
We discussed problems with that approach earlier: AIs don’t store concepts in single neurons. There are complicated ways to “unfold” AIs into more comprehensible AIs that are more amenable to this sort of thing, but nobody’s been able to make them work for cutting-edge LLMs.
Hendrycks et al cut the Gordian knot. They generate simple pairs of situations: in one half of the pair, the AI is doing a certain task honestly, and in the other half, the AI is doing the same task dishonestly. At each point, they read the innards of the AI as it answers the question.
Discovery Alert: Watch the Synchronized Dance of a 6-Planet System | NASA
Rather beautiful:
This animation shows six “sub-Neptune” exoplanets in rhythmic orbits around their star – with a musical tone as each planet passes a line drawn through the system. The line is where the planets cross in front of (“transit”) their star from Earth’s perspective. In these rhythms, known as “resonance,” the innermost planet makes three orbits for every two of the next planet out. Among the outermost planets, a pattern of four orbits for every three of the next planet out is repeated twice.
Stasis is Illiberal | Overcoming Bias
A stereotypical “illiberal” society is culturally and socially static and conservative, tightly bound, and suspicious of outsiders. Each is run by a coalition of elites who maintain strong discretionary control over key social areas: government, law, commerce, religion, culture, and public talk. Such societies can react in swift, decisive, and unified ways to external military threats and opportunities. […]
That is, not only did liberality promote innovation, but innovation also promoted liberality. Less liberal societies were seen to fall behind and look “backwards” in international competitions related to commerce, war tech, population, and culture, all because they less supported innovation. So the world learned this rough lesson: winning societies must support sufficiently neutral competition to allow internal innovation. […]
I’ve said that world population looks to peak soon and then fall far, and that econ theory predicts the world economy will then also fall, with innovation rates falling in rough proportion to economic activity. Thus as population falls, innovation will “grind to a halt.” Many have asked why that is such a bad thing. […]
The above is my response. When innovation grinds to a halt, so also will many of the social pressures that have driven and supported liberalism.
1 star
“Show Me Your Favorite Dance Move” | Kottke
These compilation videos of Ed People asking folks from around the world to teach him how to do their favorite dance moves has been going around social media for awhile. I finally sat down to watch them and they are as wonderful, charming, and happy-making as everyone says they are.
Why woodpeckers don’t get concussions | Physics Today
Contrary to popular belief, the birds don’t have shock absorbers in their heads.
Why Can't Everything Be Free? | Bet On It
“Why can’t everything be free?” I’m always delighted whenever a child asks me, because I have an intellectually solid answer even a child can understand. Namely: If everyone had to produce for free, there would be virtually nothing to buy. If everything had a price of zero, consumers would strive to fill their shopping carts with anything they could get their hands on. Producers, however, would basically stop working. […]
When children ask, “Why can’t everything be free?,” they might be visualizing a world where charging money for goods is illegal. Most of the time, however, they are probably visualizing something quite different: A world where government directly gives everyone what they want, free of charge. Gratis.
What’s wrong with that? The easy answer is to say, “Well, someone has to pay for stuff, so government freebies ultimately mean enormous taxes.” That’s true, but misses the deeper economic insight. Namely: Giving people whatever they want, free of charge, is extremely wasteful, because it leads people to consume products they barely appreciate.
Massive 4,000-Year-Old "Walled Oasis" Discovered In Saudi Arabia | IFL Science
A huge, Bronze Age fortification enclosing the Khaybar Oasis in the North Arabian Desert has been discovered by archaeologists. The walls would have once stretched over 14.5 kilometers (9 miles), making this one of the two largest walled oases ever unearthed in Saudi Arabia.
What remains of the vast rampart has been dated to between 2250 and 1950 BCE, during which time oases were common in the region and were inhabited by sedentary populations.
I’m an Ultrarunner. Taylor Swift’s Treadmill Workout Wrecked Me. | Outside
After three-plus hours on the treadmill belting out every song on the Eras tour, I can tell you why Swift’s concert training regimen works
Numberphile: “A Sudoku Secret to Blow Your Mind” | Kottke
I am not a sudoku player but I do appreciate the logical nature of the game, so Numberphile’s explanation of a simple pattern hidden in every single sudoku puzzle was pretty satisfying.
Hyundai's S-A2 Air Taxi Wants to Be the Uber of the Skies | Daily Beast
Or at least that’s what Hyundai is hoping with the announcement of an air taxi concept revealed at the Consumer Electronics Show (CES) called the S-A2. The electric vertical takeoff and landing (eVTOL) aircraft seats four people and can travel at 120 mph at an altitude of up to 1,500 feet.
The battery allows the craft to operate in 25- to 40-mile trips in urban environments. That’s just enough for a flight from most airports to a downtown area—which is what Hyundai and Supernal, the company’s eVTOL division, is likely aiming for. Hyundai also claimed that the eVTOL will “operate as quietly as a dishwasher” when in use.
AI Robot Bests Marble Maze Game | Kottke
It’s a trip watching how fast CyberRunner can run a marble through this wooden labyrinth maze.