What public figures publish and believe, in their own words.
About this feed
Highlights: posts that did unusually well for the person who wrote them, everything
they published at length, each release and new project, and every belief — at most
two a day from anyone. Day by day, newest
day first; within a day, the people with the most beliefs on this site come first. Nothing
else orders it. Show everything instead.
The quoted blocks are what people actually said; a beneath one is
the belief those words support, in korrents' wording. Nobody here wrote their own page.
Top people are the people in this feed with the most beliefs on this site, then the
most here. Choose an area and the row leads with the people whose beliefs are about it;
tap a face for their feed.
Similarly, AI translation will prevent you from learning to speak another language, AI summaries will block you from really understanding the arguments laid out in a book, and AI analyses of your personal problems will make it harder for you to think through solutions to your own problems.
Point of view is essential for effective twists. Contrast Frozen with Fight Club, where the narrator's unreliability works because the entire story flows through his perspective and he maintains a direct relationship with the audience through fourth-wall breaks.
Cormac McCarthy's Child of God is the perfect starter book to see if you enjoy McCarthy's maximalist prose and macabre images before tackling the longer Blood Meridian.
A reminder that my novel Metallic Realms is available at a bookstore near you. Reviews call it "brilliant" (Esquire), "a total blast" (Chicago Tribune), "unrelentingly smart and inventive" (Locus), and "just plain wonderful" (Booklist). Maybe the perfect holiday gift for the booklover in your life?
I just realized the only 2 countries left with actual substantial startup activity now are literally only the US and China The rest of the world can't really do startups, doesn't have the funding, can't grow them and it's more like performative hobby projects for their governments Which might tell us where the future wealth will be concentrated in the world
AI has been compared to various historical precedents: electricity, industrial revolution, etc., I think the strongest analogy is that of AI as a new computing paradigm (Software 2.0) because both are fundamentally about the automation of digital information processing.
It significantly lowered the bar to production. It is how we got the whole
society to run on software. If you make it harder for hobbyists maintainers, you
are going to crash society.
Making licenses suddenly commercial for big entities? You are forcing people to
go back through the Procurement process. They will shift over time back
to a FOSS solution, even if it is less good.
FOSS solve that problem far more than it solves “we don’t want to pay”. Most
corporations would be surprised by how cheap it would cost them to pay
for the FOSS software they use.
But even more important than efficiency, designed experiments can inform about causality, which is very difficult to determine from collected observed data.
this fundamentally this category of coding and AI is probably going to be one of the biggest categories right it is a software factory category in fact it may be bigger than knowledge work
So this AI thing will be that right. So if you take coding um what we built with GitHub and VS code in over whatever decades uh suddenly the coding assistant is that big in one year and so that I think is what's going to happen as well which is the market expands massively.
even if the tech is diffusing fast uh this time around for true economic growth to appear it has to sort of diffuse to a point where the work the work artifact and the workflow has to change and so that's kind of one place where I think uh the change management required for a corporation to truly change I think is something we shouldn't discount
Theodora Goss's latest book of short stories is 'Letters From an Imaginary Country,' and it manages to be one of the most voraciously, delightfully readable books I've ever read and it's one of the most writerly books, too. What a treat!
If a bed is too deep, one has to use a very coarse grind. The result in the cup may or may not be good enough, but the extraction level will be low, which is inefficient, in the sense of wasting coffee material.
That's great, but I don't think it can replace a human poking at virtual machines; my experience with EC2 is that images break in all sorts of wild ways, and nothing beats a human for saying "huh, I have no idea what's going on here but something seems weird".
In Science Comics Computers: How Digital Hardware Works, legendary cypherpunk Perry Metzger teams up with Penelope Spector and illustrator Jerel Dye for a tour-de-force young adult comic book that uses hilarious steampunk dinosaurs to demystify the most foundational building-blocks of computers. It's astounding
I am far from certain that NASA can regain anything like the vigor necessary to respond decisively to the Chinese threat, but I am determined that the free people of Earth not live under a Red Moon
Even for $5.2b per launch, Orion is so overweight that there does not exist a booster nor an upper stage that can launch it to Low Lunar Orbit and return it to Earth.
I think the power and money corrupted him in the end as well, but I also think that there’s something fundamentally anti-human about people who want to build utopias or paradises or heavens, because what they’re saying is, “I like humans apart from the bad bits.”
And then how do they speak? You know, because fundamentally, it doesn’t really matter what’s going on in their head; they haven’t actually got one, but what they say is what’s going to make you realize who they are.
Because I always thought that the games were kind of a mathematical equation. They were the personality of the world, you know, multiplied or divided by the personality of the protagonist.
You know, you were in, you were in these worlds, and you went there as a visitor, and they existed almost independent of you. It felt like when you turned up, the world was running. It didn’t feel like you’d started it.
I do not think this book could really be enjoyed without the previous book, but with it, it's lots of skillfully-written fun, and leaves me eager for more.
I think a great book could be written on all this. (Among other things, it should be catnip to anyone who wants to call themselves a historical materialist.) This book, however, is merely OK.
Raymond Biesinger's new book 9 Times My Work Has Been Ripped Off is a masterclass in how creative workers can transform the endless, low-grade seething about the endless ripoffs of the industry into something productive and even profound
But politics aside, what’s notable about that 2001 report, put together by men who regarded themselves as hard-headed realists, is that it envisaged an energy future reliant almost entirely on fossil fuels plus a bit of nuclear energy.
there are many people who still refuse to believe that renewables can be practical even as renewables keep growing around the world, in fact accounting for the bulk of growth in electricity generation
If I took my daughter to China for a year, and we just lived there with no language instruction, I'd come back knowing approximately six Mandarin words and she'd be fluent.
An idea I'm trying to hold tighter is that the only way to be useful here is to risk offending people. And perhaps if you're not taking any risk offending people then it's unlikely you're saying something useful.
I had typed "Taylor Swift" in a browser, and the response had literally zero links to Taylor Swift's actual website. If you stayed within what Atlas generated, you would have no way of knowing that Taylor Swift has a website at all.
Read this in the context of Christopher et al., and the reasonable conclusion is that many of the problems associated with homelessness are the result of unsheltered homelessness.
It turns out that simply getting people into temporary shelter—often derided by the homelessness-industrial complex as a band-aid—significantly reduces both social and individual harms of unsheltered homelessness.
My sense is that this hybrid is fairly common in practice; solvers aren't magical and if you can deduce additional structure using domain-specific analysis, it will often give the solver an important boost.
If human beings can dismiss academic arguments merely because they lead to politically unwelcome conclusions, what’s the point of having academic arguments about politically-charged subjects?
It's screamingly funny, devilishly inventive, and deeply, profoundly satisfying. With Fever Beach, Hiaasen makes a compelling case for Florida as the perfect microcosm of the terrifying state of America, and an even more compelling case for his position as its supreme storyteller.
An average email or line of code is fine. Average art isn’t. To make something alive with AI, we have to resist its pull towards average by working beside it, shaping what it gives, and listening for what’s missing.
So while vibe coding may be useful for short-term work, it’s not a suitable approach for anything intended to last longer than a tub of yogurt. Time saved is not strength gained, so I went looking for other examples about how to work with the machine.
The fractures fell neatly along disciplines: engineers using AI to wish away designers, designers wishing away engineers, product managers wishing away both. In this climate, AI becomes frenemy identification technology, another way to avoid working together.
You use a tool, but you play an instrument. It’s a more expansive way of doing, and the doing of it all is important, because that’s where you develop the instincts for excellence. There is no purpose to better machines if they do not also produce better humans.
To get to principal, you need to put yourself on the critical path. To be effective as a principal and go beyond it, you need to actively remove yourself from it.
With great freedom comes great responsibility. You have the autonomy to choose what to work on, but there's the expectation of accountability and impact.
Being right is less than half the battle. You also have to convince others that you're right, and more importantly, convince them to care enough to act on it.
One possible reading of the last century or so of political history in China, therefore, is that governments who provide a meritocratic channel for social advancement have legitimacy and popular support, and governments who do not provide one, do not.
And it just so turns out that um this was extremely early, way too early. so early that we shouldn't have been working on that, you know, uh because um if you're just stumbling your way around and keyboard mashing and mouse clicking and trying to get rewards in these environments, um your reward is too sparse and you just won't learn and you're going to burn a forest uh computing and you're never actually going to get something off the ground.
So like they don't have enough intelligence. They're not multimodal enough. They can't do computer use and all this kind of stuff. And they don't do a lot of things. You know, they don't have continual learning. You can't just tell them something and they'll remember it. And they're just cognitively lacking. And it's just not working. And I just think that it will take about a decade to work through all of those issues.
I think that for those who can create and then articulate a framework around what is taste for us that is really a important skill and uh then I think people can a lot of people can basically match a framework not many people can create the framework.
It's just this loop of okay, I'm having an experience of any sense. Maybe I'm looking at art, maybe I'm hearing music, maybe I'm literally eating food and tasting something, you know, but like do I like it? Do I not like it? Why?
we're looking for high judgment individuals, people that are going to roll up their sleeves and do a lot, whether they're IC's or managers, and people that are going to get the details and perfect their craft, because we know that's how we're going to win is by having the best craft, the best design.
I think in this world where software can be created more easily, design matters so much and designers matter so much. I think designers are going to be the leaders of the future and I think that more designers need to speak step into that leadership role
if you want to win in the game of software, you need to differentiate through design. Like that's again how you win or lose. Craft matters. And so we're no longer in this era of good enough is fine. It's like good enough is not enough.
Cursor… it’s tab completion model is industry leading, if you still write code yourself. I use VS Code mostly, I do like them pushing things like browser automation and plan mode tho.
I keep an eye on China’s open models, and it’s impressive how quickly they catch up. GLM 4.6 and Kimi K2.1 are strong contenders that slowly reach Sonnet 3.7 quality
I keep an eye on China’s open models, and it’s impressive how quickly they catch up. GLM 4.6 and Kimi K2.1 are strong contenders that slowly reach Sonnet 3.7 quality
The empirical evidence strongly refutes the idea that rising opportunity costs of childbearing are driving low fertility. Declining marriage due to exogenous technology and economics shocks is tanking fertility irrespective of child direct or opportunity costs.
the shift from 50th to 99th percentile earnings has no negative effect on women’s fertility, and unearned income for women has a neutral-to-positive effect
Although the cross-sectional relationship between national income and fertility is negative, the longitudinal relationship between national income and fertility is positive: economic booms create baby booms, economic busts create baby busts.
I don’t see population collapse or lack of workers as the central policy problem: the central policy problem is human happiness, human flourishing, and the freedom to live life well.
In general your body would rather use glycogen than fat, because glycogen loses almost no energy in the conversion from and to sugar and fat loses a lot.
Over the past few decades, journal editors and peer-reviewers have increasingly insisted that papers must present large datasets that have been treated using complex statistical methods in order to make even the mildest claims about what caused what.
This is the best screenshot utility for Macs. You’d think how much can there be in a screenshot utility, and CleanShot X will still color you surprised.
I approached Echoes of October with trepidation. A graphic novel about violence and grief isn’t easy terrain. But it succeeds in a haunting, urgent way.
Data centers are so much more efficient with their water that they generate 50x as much tax revenue per unit of water used than golf courses in the county:
So AI consumes approximately 0.04% of America’s freshwater if you include onsite and offsite use, and only 0.008% if you include just the water in data centers.
For much more on that side of things, you should read Money Stuff or listen to Odd Lots, which cover “private credit is the new bank lending” all the time.
For much more on that side of things, you should read Money Stuff or listen to Odd Lots, which cover “private credit is the new bank lending” all the time.
A Lawrence Berkeley National Laboratory study found that state-level load growth from 2019 to 2024 — including growth driven by data centers, manufacturing, and electrification — was associated with lower average retail electricity prices, not higher ones.
In some cases, design is what something looks like. In other cases, design is how something works. But the most interesting designs to me are when design changes your behavior. Even the smallest details can change how someone interacts with something.
And I think what's also kind of interesting is that just human nature is that if you have a question that you feel like, you know, you might be a bit embarrassed to ask or you don't want to bother someone, you just don't ask the question, right? And so by the nature of having these tools, you get actually a pretty large explosion of questions being asked.
And if I'm starting to see like more and more experiments that are not statistically significant, that may be a signal to me to say, okay, we might have kind of tried to exploit a little bit too far. Like there might not be as much juice to squeeze.
I mean in my experience the typical win rate and I hate to use that term for experiments is is often something like 30 to 50%. Like usually you're not actually get like you're trying a bunch of things a lot of hypotheses turn out not to be true. you know, consumer products are very unpredictable like that.
When you lose a game now as opposed to surfacing your blunders and your like horrible stuff that you did, we flip it on its head. And so we show you your brilliant moves, your best moves. And we have coach say something encouraging. You know, losing just part of learning like keep it up, that type of thing. That change alone was pretty dramatic for us. It grew game reviews by 25%, subscriptions by 20%
If you do too much exploration, you can have your team feel a little bit too scattershot, just trying a 100 different random ideas. What's the through line? What's the strategy? How do you pattern match, you know, successes across them? And if you do too much in exploitation, which is often the MMO of growth teams, it can lead to this like saturation and stagnation where you're just locally maximizing a thing.
The overall workflow is intuitive, especially for those new to formal evaluation processes. The UI guides you through creating datasets, running experiments, and annotating results.
The panel had a generally positive view of Braintrust, highlighting its clean UI and structured approach to evaluations. The tool’s emphasis on human-in-the-loop workflows was a significant strength.
The panel had a generally positive view of Phoenix, with one panelist calling it one of his “favorite open source eval tools.” The tool is positioned as a developer-first, notebook-centric platform.
As for me personally, I tend to use these tools as a backend data store and use Jupyter notebooks as well as my own custom built annotation interfaces for most of my needs.
And you're supposed to be really connected, but the conclusion you reach very early is that the more connected and accessible you are, the less productive you are.
I don't think a phone is a necessary device. I remember growing up, I didn't have a mobile phone. When I was a student at the university, I didn't have a mobile phone. When I finally got to use a mobile phone, I never used phone calls.
So it's extremely important to curate the information sources that you have, so that you wouldn't be somebody who is left to the will of AI-based algorithmic feed telling you what's important so that you end up consuming the same information, the same stuff, the same memes, the same news as everybody else.
So think about this. If your brain is this most valuable tool you have in your journey to success and happiness, why would you destroy this tool for short-term pleasure? This sounds ridiculous.
Well, the biggest enemies of freedom are fear and greed, so you make sure that they don't stand in your way. If you imagine the worst thing that can happen to you and then make yourself be comfortable with it, there is nothing more left to be afraid of.
it seems like struggle is a part of life, either you do it earlier or you do it later. And it’s true, that’s such a good point that competition does seem to be a really powerful driver of skill development
I think this situation, this historic wrong that’s been done is, put simply, is just a gigantic PR mistake for France. There’s no entrepreneur that sees, that aspires to be the next Pavel Durov to create the next Telegram, sees this and wants to operate in France after seeing this.
One of the ways to achieve freedom of thought is to remove your mind from the influences, the forces that manipulate you. That’s really important to realize the content you consume, especially on the internet, when a large percentage of it is designed to manipulate your mind. You have to disconnect yourself.
If you’re stuck in a depressive cycle, even when you really, really, really, really don’t want to do anything, to do something. Try to make progress because the good feeling comes on the end of that. The whole point is to do first and then feel, not feel and then do.
This is a short (~150 pp), brisk first book on the topic for physicists. Most of the book --- the first ten chapters --- are good introduction to common models of spin glasses.
The scalable method is you learn from experience. Um you uh you you try things, you see what you see what works. No one no one has to tell you. First of all, you have a goal. So without a goal, uh there's no sense of right or wrong or better or worse. So large language models are trying to get by without having a goal or a sense of better or worse. That's just, you know, it's exactly starting in the wrong place.
So there's no ground truth. You can't have prior knowledge if you don't have ground truth because the prior knowledge is supposed to be a hint or an initial belief about what the truth is.
reinforcement learning is about understanding your world whereas large language models are about mimicking people doing what people say you should do. They're not about figuring out what to do.
the perspective I outline is very bullish on AI capabilities and very concerned about the risks which may arise by default, but also hopeful that, through foresight and vigorous action, we can mitigate the worst of these risks.
Companies and countries are tempted to cut corners by default and would benefit from some rules of the road, at least if they're sure others will follow those rules.
Here’s a list of great investing books that somehow have fewer than 50 reviews! Being Right or Making Money by Ned Davis The Art of the Market by Bob Tamarkin and Les Krantz Triumph of the Optimists: 101 Years of Global Investment Returns by Elroy Dimson, Paul Marsh, and Mike Staunton 2020 Vision: Investment wisdom for tomorrow by Harry Liem The Silver Bears by Paul Emil Erdman
Here’s a list of great investing books that somehow have fewer than 50 reviews! Being Right or Making Money by Ned Davis The Art of the Market by Bob Tamarkin and Les Krantz Triumph of the Optimists: 101 Years of Global Investment Returns by Elroy Dimson, Paul Marsh, and Mike Staunton 2020 Vision: Investment wisdom for tomorrow by Harry Liem The Silver Bears by Paul Emil Erdman
Here’s a list of great investing books that somehow have fewer than 50 reviews! Being Right or Making Money by Ned Davis The Art of the Market by Bob Tamarkin and Les Krantz Triumph of the Optimists: 101 Years of Global Investment Returns by Elroy Dimson, Paul Marsh, and Mike Staunton 2020 Vision: Investment wisdom for tomorrow by Harry Liem The Silver Bears by Paul Emil Erdman
Here’s a list of great investing books that somehow have fewer than 50 reviews! Being Right or Making Money by Ned Davis The Art of the Market by Bob Tamarkin and Les Krantz Triumph of the Optimists: 101 Years of Global Investment Returns by Elroy Dimson, Paul Marsh, and Mike Staunton 2020 Vision: Investment wisdom for tomorrow by Harry Liem The Silver Bears by Paul Emil Erdman
Here’s a list of great investing books that somehow have fewer than 50 reviews! Being Right or Making Money by Ned Davis The Art of the Market by Bob Tamarkin and Les Krantz Triumph of the Optimists: 101 Years of Global Investment Returns by Elroy Dimson, Paul Marsh, and Mike Staunton 2020 Vision: Investment wisdom for tomorrow by Harry Liem The Silver Bears by Paul Emil Erdman
but if I was, you know, at the zero or 10th percentile, it can certainly get me even today very quickly up to like the 60th 70th in terms of um what the state of the art is.
And so the main thing that I felt so excited about and this is something I tell my team all the time is we need to dissolve the boundaries of these traditional roles.
if you're really unclear about what success looks like the prompt, you're probably not going to get the most amazing work. Um I think that's true for managing teams and it's very much true for managing AIs.
And you have to understand the strengths of of you know used to be people but now it's basically models and different models have different strengths. So it's like they have different personalities
management is just about in my mind having an outcome. So you want to get something done. That's the thing. You have a northstar. You have a vision and you're just trying to figure out how to use the resources that you have to get that thing done.
His officers didn't think so. High command said, “No, we're not going to attack the West; we're going to lose.” And Hitler was fanatic about it, he really wanted to attack it. They were planning a coup against him in November 1939 just to prevent him ordering the attack on the West, because it would have been a catastrophe for Germany.
That was the victorious idea. So it's not the drugs. Actually, that idea to go through the Ardennes Mountains, if you think monocausal, you would say that's the reason. That idea was genius, and Hitler immediately understood it, because before, the plan was to attack in the north of Belgium, which is the same as World War I.
And he said this is actually the missing link that historians did not have, especially to explain Hitler's degeneration as a leader. He made very good decisions, good in meaning militarily effective decisions in the beginning of the war, and very bad decisions for the German war effort towards the end. And you can link that to drugs.
So the Navy hired the shoe runners unit from the SS, paid them money, and then gave them drugs, different kinds of drug combinations, methamphetamine combined with cocaine and chewing gum and all kinds of things. So this is a big thing, you know. And there are documents to it.
But it was used by right-wingers, and Hitler didn't mind because people who are drunk are more susceptible to right-wing populism, I would claim now here, and Hitler would agree. So he did not think it was bad that these people were a bit drunk, or maybe even very drunk, because if you're drunk you also get aggressive against others.
I I think this is golden age of humanity right now. It's the I think it's the end of the beginning. And and so I think we will um I think people in 1,000 years will um study these times uh the way we are studying um Cicero's times of the end of the Roman Republic
Just imagine you use very bad software and um the team that started the project decided early to use Xamarin and then they never had a chance of making this as good as all the other things you really appreciate on your phone due to this decision no matter how good they are.
what you are try like trying to do as a tool maker, as an infrastructure builder, um as a product maker is you want to make a uh tool that brings the floor up significantly, but doesn't constrain the ceiling. And that is extremely hard.
You don't even need to predict the future that well. If you just live in everyone else's relative future, which really is just a couple of um high conviction high courage uh update to beta clicks. It's not that hard to live in the future.
Well, when the institutions are over and over spectacularly wrong, whether just incompetent, politicized, or mendacious, there is a reason for loss of trust.
The “climate crisis” has been taught as gospel for 20 years. Now the average person can see their teachers have been preaching politics in the name of science.
climate change though real is not the civilization-threatening disaster pitched at Davos, Paris, and the media, that hugely expensive climate policies do no good
language models have world knowledge and can eloquently talk about products, but are unaware of our catalog. Also, their recommendations are generic and suffer from popularity bias.
And I think we'll see the same thing with automation where uh basically robot plus human is much better than just human or just robot. Uh and and that just like makes total sense. It also makes it much easier to get all the technology bootstrapped because when it's robot plus human, now there's a lot more potential for the robot to like actually learn on the job, acquire new skills.
So, I think it'll be the same thing that that we'll see an increase in the scope that we're giving that we're willing to give to the robots as they get better and better where initially the scope might be like there is a particular thing you do like you're making the coffee or something. Uh whereas as they get more capable, as their ability to have common sense and a broader repertoire of tasks increases, then we'll give them greater scope. Now you're running the whole coffee shop.
Like if you answer a question, you just like answered it wrong. It's like well it's not like you can just like go back and like tweak a few things like the person you told the answer to might not even know that it's wrong. Whereas if you're like folding the t-shirt and you messed up a little bit like it's pretty obvious like you can reflect on that, figure out what happened and do it better next time.
Uh but it uh I think we kind of like know like roughly the puzzle pieces and it's something that we need to work on and I think if we work on it and we're a bit lucky and everything kind of goes as planned I think single digit is reasonable.
So to me like what what I tend to think about a lot in terms of timelines is not the date when it will be done but the date when it will when like the flywheel starts basically.
I don't say this lightly, but I feel like indie games right now are the best they've ever been, and I want to live up to that new standard with my own work.
This is Russia's war. And it is Russia that should pay. This is why we need to work urgently on a new solution to finance Ukraine's war effort on the basis of the immobilised Russian assets. With the cash balances associated to these Russian assets, we can provide Ukraine with a Reparations Loan.
Battlelines for a new world order based on power are being drawn right now. So, yes, Europe must fight. For its place in a world in which many major powers are either ambivalent or openly hostile to Europe.
This must be Europe's Independence Moment. I believe this is our Union's mission. To be able to take care of our own defence and security. To take control over the technologies and energies that will fuel our economies.
With Web forms, the burden is on people to adapt to databases. Today's AI models, however, can flip this requirement. That is, they allow people to provide information in whatever form they like and use AI do the work necessary to put that information into the right structure for a database.
But in thousands of years of education humanity has discovered no replicable and reliable means of taking kids from one educational percentile and raising them up into another.
We live in a culture that preaches self‑esteem more than self‑knowledge. That leaves people morally flabby-performing virtue while outsourcing the actual discipline that keeps ordinary humans from doing monstrous things.
People who insist they're harmless aren't safe; they're naive. People who know exactly how dangerous they could become-and fear it-are the ones who deliberately yoke that force to good aims.
I use FreeAgent for sending invoices and handling bookkeeping. I've been using it for over five years at this point and I don't know what I'd do without it.
So, giant eyeballs, they can either see very well, they can see a very long way in daylight, or they can see very well at night. And my suspicion is it's the latter. I think they're probably primarily nocturnal when they get that size.
So, I think it's just straight fiction. It works for the book and it works for the movie, but it's... as far as I can tell, it's straight fiction, and Crichton just made it up. If it's buried in some bit of literature, he's done better finding it than I have, and I've had a really good look, and I know how to look.
But this is the absolute rule of thumb for how foraging and growth and predation operate across everything from fish to starfish, praying mantises, all the way up to things like big cats via stuff like crocodiles. It's how it works. So it'd be very weird if it didn't also operate for dinosaurs. And then, as I say, we've actually got the direct evidence for this from bite marks and stomach contents. They're taking small stuff.
But fundamentally, T-Rex is definitely weird, even compared to all the other giant tyrannosaurs that are very closely related to it, because it is by far, ludicrously by far, the largest carnivore in its ecosystem.
Agentic and sync are natural complements because local-first is multiplayer by default, and agents are another sort of "player" in addition to human collaborators.
There are obvious pricing problems in this scenario. With a new fab, a company can’t turn a profit selling a lagging-edge chip at the price that was previously dictated by a fully depreciated fab. Prices have to go up.
Their words now
The COVID super-cycle, plus a new China supply line that does not care about capacity economics, points to a forever glut.
Tweets from 1765-1799 by a 4’9” hunchback physicist, friends with Goethe and Kant, admired by Nietzsche, Schopenhauer, etc. Such wonderful random thoughts, beautiful perspectives on thinking for yourself, observing nature, language, freedom, philosophy, religion, and more. Hundreds of initial insights, especially inspiring because they’re undeveloped.
film was this high margin product where they had a bunch of unique intellectual property and digital cameras are a low margin um commodity where they were competing with the entire consumer electronics industry with no differentiation.
I think it's actually the opposite which is that everyone's kind of using the same data which is you need such an enormous amount of generalized text that the amount that Google has or that meta has is not actually enough to move to be a kind of fundamental difference in what you can train with.
I'm pretty sure people thought Microsoft had an advantage on the internet and Google and um Meta had an advantage on mobile and everyone thought IBM was going to win PCs. Like once IBM made a PC, that was it. It's all over now. And we kind of forget that like there were PCs before and then IBM made one and that kind of became the standard but then IBM lost it.
It wasn't clear that the browser wasn't where the value capture was cuz Microsoft crowbarred its way into dominance in browsers but that turned out not to matter and then all the value is in search advertising and social which were 5 years later and 10 years later and so like you can be very very clear that this is the thing and then still be completely unclear how it's going to work.
My sort of base case is to say this is kind of another platform shift and all the new stuff will be built around this for the next 10 or 15 years and then there'll be something else and so the impact on employment will be kind of like the impact on employment from the other platform shifts
When AWS ships a service which is half-baked, it diminishes customer trust in AWS as a whole; even if the problems in that service ultimately get corrected (either by fixing them or in some cases by simply getting rid of a service which should never have existed in the first place) the memory of a failed launch will live on in customers' minds for years to come.
While the same does not directly apply in other fields, working with others to produce the best results for everyone will be much better in the long-term than focusing solely on what Amazon needs right now.
Customer Obsession is great, but I often see Amazonians taking this too simplistically: "Start with the customer" doesn't have to mean "ask customers what they want and then give them faster horses".
However, the methods used to produce these estimates are debatable, and seem to have been chosen to give the maximum possible value for data center water consumption.
On this measure, state capacity at the local level looks much stronger in China than in India, which helps explain some of the long-running economic differences between them.
It's also worth noting that chat isn't the only way to integrate AI in software products and increasingly agent-based applications outperform chat-only solutions. So expect things to keep changing.
But they can't do everything that megapapers do, such as conveying intellectual heft and legitimacy, and they are not yet a replacement for real, diverse experts.
It's very relatable; the internet is so toxic but we kind of have no choice but to interact with it if we want to be heard as musicians / writers / artists / citizens.
The band told me about the pressure of creating content in order to promote their music and the ways in which the audience now has power over how songs sound.
Still don’t see how this could be moved to background agents. I steer the models a lot as I notice them drifting off - that’s much harder if they run in the background.
Bigger changes always get tests. Automated ones usually aren’t great, but the model almost always finds issues when you ask it to write tests IN THE SAME CONTEXT. Context is precious, don’t waste it.
Even removed my last MCP, since Claude sometimes would go off spinning up Playwright unasked when it could simply read the code - which is faster and pollutes the context less.
Likewise, self-diagnosis should be understood as an attempt to secure the therapeutic effect of a medical intervention to which patients do not have official access.
If a clinical diagnosis can have a therapeutic effect, then, at least in some instances, diagnoses are medical interventions in themselves and ought to be treated and researched as such, with an eye to both positive effects and potential iatrogenic harms.
Their experiences highlight a striking, neglected and unchristened medical phenomenon: The therapeutic effect of a clinical diagnosis, independent of any other intervention, where clinical diagnosis refers to situating the person's experiences into a clinical category by a clinician or the patient.
The Weight is Mellia Mendes's award-winning webcomic, now collected in a massive, beautifully made graphic novel from Drawn & Quarterly. It will tear your heart out, it will send you to a dreamy world of pastoral utopianism, then it will tear your heart out.
I suspect that if you decided to be open-minded, then the Being would probably be extremely persuasive. But I don’t think it’s very common to do that.
Their words now
I take it back. Instead of being unsure, I now lean strongly towards the idea that the Being would in fact be very good at convincing people of stuff, and far better than any human.
I think this slop-adjacent realm of dupe products and Instagrammable potions points to our increasingly desperate need to manipulate our own images, and an acceptance of the fact that we're kind of hybridized monsters ourselves in this post-A.I. human-machine moment
My model is that research requires a mix of skills. The day-to-day coding and execution is crucial. But there's also a set of harder-to-learn conceptual skills, collectively called research taste. These skills take a long time to gain because they have poor feedback loops, but they take very little time to use.
I'm more agnostic about the best techniques, things like sparse autoencoders are a useful tool, but easy to waste effort using when a simpler method is sufficient or better - start by doing the obvious thing!
I'm now fairly pessimistic about ambitious interpretability (i.e. complete reverse-engineering), and I'm excited about model biology (studying qualitative high-level properties of models) and applied interpretability (rigorously doing useful things with interp).
I don't think all parts of the economy can absorb intelligence equally. So let's just say we develop fairly generalized super intelligence. I always use the analogy like you can invent a lot of drugs, but if clinical trials still take a long time, you're not necessarily going to get new therapies rapidly.
I just think about like really simple things like access to mental health care, access to education, access to medical advice, access to legal advice. We're essentially taking expertise and making it a commodity. And I think that will is generally democratizing.
if you tell what what jobs are most likely to be automated with the current generation of technology, you would probably put software engineering right at the top. The people building this technology or building the technology that is disrupting their own profession.
If you look at the S&P 500 now and you look at the amount of value from companies created in that, one could argue that actually almost all of the exuberance and hype was totally warranted and it in fact did change commerce in fundamental ways.
So when you make something like chat GPT, you can go from to zero to 100 million users faster than any technology in history because of the other technologies though. like you couldn't have gotten there if not for the build out of the internet, if not for the smartphone.
Seat based. Well, what is a seat? That doesn't make any sense. Consumption where it's like is it per message? Is it per token? Is it per conversation? And none of these things actually mapped very well to getting a job done and getting a job done well.
And in contrast to software as a service or software you'd buy off a shelf at, you know, Fry Electronics, you know, decades ago, which might help you be marginally more productive, help you get a job done. Agents, in contrast, are actually getting the job done for you. And so you're in essence hiring software to accomplish a task and get it done well.
And it came back and half of them were great, a quarter of them were total garbage and a quarter needed some work. But like it was, you know, in a way him working the entire weekend with, you know, guidance from him.
One of the first was we had a bet as to what percent of kind of all incoming customer issues one of our agents could resolve. And Brett was maybe the pessimist and realist favorably charitably. And I bet we'd exceed 80% by the end of the year and we did.
Brett and I like betting and one of the bets we have is the year in which greater than 50% of conversations with agents built on Sierra are with people's personal agents. So agents talking to agents.
Almost all discussions of it center on “data”, but it’s actually a fight about payments, and whether banks have a right to monopolize and charge for all economic activity their users engage in, irrespective of whether the bank operates the payment method.
Do you want technological supremacy and radical breakthroughs? Well, yes. If you do, then you have to tolerate that there are going to be some people who are just going to be uncomfortably rich, like in the United States. You can’t say that you detest the financial system for its greed, but then not accept the fact that it is what fuels innovation, right?
Industrial policy was heavily criticized and again, if I talk to my Harvard professors, they’ll be very, very, very skeptical about the role of the state, especially in such a major way. But if you look at the Chinese experience, the evidence for success is all there. The cities that have pushed the most on supporting that particular sector ultimately did the best in terms of production, but also in terms of innovation, measured by patents. It’s all in the data.
I say now that China’s biggest challenge is consumption, it’s not production. We know that China’s very, very competitive in production and supply. China needs to be a consuming country to be a rich country.
I’ve rarely seen a more capitalist society than China, from the pure economic side. I’ve rarely seen companies that are as competitive as Chinese companies. People as ambitious and obsessed with making money as Chinese people. Kind of ruthless actually. And look, consumers shop, firms invest. If you invest well financially, you’ll get great returns. What is not capitalistic about the Chinese economy? At the same time, the social fabric is highly socialist.
The biggest misunderstanding is somehow that a group of people or even just one person runs the entire Chinese economy. It is far from the reality. It is a very complex, large economy, and even if there is an extreme form of political centralization, the economy is totally decentralized.
Any performance test should be accompanied by an explanation of the limiting factor, since no explanation will reveal the test wasn't analyzed and the result may be bogus.
One performance anti-pattern is when a company, to debug one performance problem, installs a monitoring tool that periodically does work and causes application latency outliers.
A new performance team can likely find enough optimizations to halve infrastructure spend in their first couple of years, even for companies that have been using commercial performance or observability tools.
Given how much moral progress we should hope to make in the future, and how much we’ll learn about what’s even empirically possible, we should act on the assumption that we have almost no idea what the best feasible futures would look like. Committing today to some particular vision would be a great mistake.
I see this as the strongest argument against the better futures perspective, and the reason why I don’t feel confident that work on Flourishing is higher-priority than work on Surviving, rather than merely in the same ballpark.
But the difference in value between the world today and that common-sense utopia might be very small compared to the difference between that common-sense utopia and the best futures we could feasibly achieve.
Ellen Ullman's Close to the Machine is a memoir I couldn't believe I hadn't read earlier. Many of her observations about engineering culture feel as relevant today as when she was a programmer in the 1980s
I'm getting into screen-time blocking. I use the app Opal on a recommendation from Lucas Gelfond. It won't let me scroll Twitter, Substack, or Instagram for 80% of the day. Gmail is also blocked from 11pm to 8am.
I initially started making Anki (well, Mochi) cards to keep track of Google DeepMind's LLM lineage—Gopher, Chinchilla, Gato, PaLM, Sparrow, Meena, LaMDA, Bard, Gemini, etc.
Yet the production methods still can’t make them below $10 per gram, so they’re still not cheap enough for some “obvious” applications like preventing kids getting malaria in a rainy season.
Going all in on AI is not the highest return-on-investment strategy if you care about human flourishing, even if AI might lead to rapid scientific advances.
Rapid AI progress does not automatically mean rapid medical progress. If the point of AI progress is human flourishing, we must make other complementary investments too.
Those are why I do not believe that AI could end disease within 10 years on its own, nor that 50-100 years of medical progress could occur 5-10 years after powerful AI due to AI progress alone
I think the first point is the extreme loyalty of the people whom Genghis Khan chose. His kinsmen, as we said, had deserted him, his anda was a questionable relationship but all the others that he found were just common people, herders or hunters, very common, and they were loyal to him and never, ever revolted against him, never betrayed him.
I think that sedentary armies, sedentary civilizations, they get stuck in ways, this is how we do it. And we're going to make it a little faster, we're going to make it a little bigger, a little stronger but this is how we think. Genghis Khan had no set way to think and, when he encountered the first walled cities around 1209 after founding his nation in 1206, he went out on these raids and I really think there were raids not wars at first.
Mongols produced nothing. They could produce felt to make their tents, but they were not craftsmen. And so they had to get these items from somewhere, and it was through raiding.
Later, Genghis Khan will realize once he becomes Chinggis Khan, he will realize that the true source of most of the feuding on the steppe is over women. And later he will outlaw the kidnapping of women and the sale of women, in part not only because of what had happened to his mother, but what happened to him next in his life.
This means that working on neural networks is NOT getting us closer to AGI, except indirectly.
Their words now
It’s become very difficult for me to maintain the belief in the stupidity of ChatGPT when every time I laugh at it, it ends up ridiculing me 6 months later.
It’s one thing if residents of Sulawesi like Mr. Jamal and his family have had to risk exposure to toxic pollutants in a successful escape from poverty. It’s another thing entirely if those bearing most of the burden see few of the benefits.
Morowali’s poverty rate has seen a modest decline since 2016, from around 15% to 12%, but this puts it squarely in the middle of the pack of other Sulawesi districts—and not far from the decline in the national average, which fell from 10.9% to 8.6% over the same period.
Poverty reduction is never a purely technocratic exercise, and questions about industrial policy can never be entirely divorced from those of political economy—including those of global geopolitics.
But it seems like you can understand it through passive observation, which is pretty surprising to me. And again, I think hints at something underlying about the nature of reality in my opinion, beyond just the cool videos that it generates.
I think to the extent that it can predict the next frames in a coherent way, that is a form of understanding, not in the anthropomorphic version of, it's not some kind of deep philosophical understanding of what's going on, I don't think these systems have that, but they certainly have modeled enough of the dynamics, put it that way
I think we've proven, and the AI community in general that classical systems, Turing machines can go a lot further than we previously thought. They can do things like model the structures of proteins and play go to better than world champion level. And a lot of people would've thought maybe 10, 20 years ago that was decades away, or maybe you would need some sort of quantum machines to quantum systems to be able to do things like protein folding. And so I think we haven't really even sort of scratched the surface yet of what classical systems so-called could do.
And it sort of fits with the way I think about physics in general, which is that I think information is primary, information is the most sort of fundamental unit of the universe, more fundamental than energy and matter.
And I think the reason that's possible is that in nature, natural systems have structure because they were subject to evolutionary processes that shape them. And if that's true, then you can maybe learn what that structure is.
We contend that games provide particularly rich and controlled environments uniquely well-suited for systematically evaluating rapid model adaptation and the process of world model induction.
We expect that building and evaluating AI systems capable of this kind of rapid world model induction will be critical for achieving robust, general AI capable of functioning effectively in the complex and fast-changing real world, and especially in human worlds - the environments that human beings have evolved in, created, and are continually changing and re-creating.
However, current understanding and evaluation of world models in artificial intelligence (AI) remains narrow, often focusing on static representations learned from training on massive corpora of data, instead of the efficiency and efficacy in learning these representations through interaction and exploration within a novel environment.
We argue that this profound adaptability is fundamentally linked to the efficient construction and refinement of internal representations of the environment, commonly referred to as world models, and we refer to this adaptation mechanism as world model induction.
the fortress balance sheet is that you run a company serving clients well, you have good margins, good liquidity, good capital. I'm as conservative in accounting as you can find. I don't up-front profits when I can spread them over time.
if you look at the history of banks from up until 2007, a lot of banks were earning 30% equity. Most of them went bankrupt. We never did that much. Okay, but in '08 and '09, we were fine, and they weren't.
And so, I always look what I call the fat tails and manage that we can handle all the all the fat tails. And not the stress test the Fed gives us, but all the fat tails.
I've always been very risk-conscious. And risk-conscious does not mean getting rid of risk. It means properly pricing it and understanding the potential outcomes.
The main argument they trotted out in that article was that we don’t need more high-yield agriculture because we don’t need more food, just better distribution of the food we already have. That’s wrong
I think we need to correct that ratio somewhat by some testimony of just how God damn amazing it is to literally create life
Their words now
those of us who've chosen to do the traditional thing to get married and to have children have an obligation to kind of talk it up a little bit, which would've seen ridiculous again 50 years ago that you'd have to talk up something so fundamental of that.
it's a luxury good like you're not paying the penalty for that the people who embrace it and think that they can just ask AI for everything or they can just look everything up but they don't have to understand the copy pasta that they slam into their project they're the ones who going to suffer
Their words now
Where I get fired up, and this ties back to the AI discussion, is when that's turned into this meme, that programmers no longer have to be competent. I mean the AI is gonna figure it out. The generators is gonna figure it out. I don't need to know SQL, active record is gonna abstract it away from me. No, no, no dude, hold up. The path here is competence.
static typing is the one thing I will literally die on the hill for for Ruby
Their words now
It's the essence of what makes Ruby Ruby. This is why I don't fully understand when people call for Ruby to add static typing. 'Cause to me, it's the bedrock of what this is.
I enjoy it even almost like as a sort of pair programmer AI pair programmer who doesn't drive who's just there to give suggestions to know the API to do all these other things but the second it starts wanting to autocomplete my code I'm like yeah I'm out bro
Their words now
I love chiseling my code and the way I use AI is in a separate window. I don't let it drive my code. I've tried that. I've tried the cursors and the wind surfaces and I don't enjoy that way of writing. And one of the reasons I don't enjoy that way of writing is I can literally feel competence draining out of my fingers.
I used to think it required more churn to get progress to stay on the leading edge of new stuff. And I wrote this before I experienced the indignity of the 2010s in the JavaScript community.
Bitcoin is not going to catch on as a charity. In order for spending it to catch on persistently at scale (i.e. not just billions of dollar-equivalents in annual global medium-of-exchange volume, but trillions), it has to solve problems for spenders and/or recipients that other solutions are not doing.
If a new monetary network with its own unit (i.e. not pegged to an existing currency as a credit rail on top of it, but rather a fully parallel system to central banks) is going to bootstrap from zero to massive, it requires upward volatility.
Perhaps that Goldman makes, at times, too strong a case for the importune of the Nomonhan war, but the conflict surely deserves much more that the oblivion into which it seems to have been consigned by political historians.
People are more likely to have kids if it’s FUN - relative to alternatives. As entertainment-technology improves, it generates a tantalising menu of ways to spend our lives.
We need better beef - which means more efficient beef, not organic grass-fed beef, which is even worse for the climate and the planet than conventional feedlot-finished beef - but in the rich world we also need to eat less beef.
Finally: tmux, which I had been happily using for many years, but slowly collecting configuration for. I decided to give Zellij a try, and haven’t looked back.
I took inspiration from Omakub, which is where I discovered Tactile - a GNOME extension for tiling windows to a custom on-screen grid using the keyboard. I’ve found this to be invaluable, and easily the best window management experience for GNOME.
I’ve grown to like Space Bar - it provides desktop workspaces akin to those in Sway/Hyprland which enabled me to use the muscle memory I’d developed over years.
I switched to zsh around 2013 and have used it every day since: starting with the infamous Oh My Zsh, powerlevel9k, and subsequently powerlevel10k, and finally settling on Starship with a couple of zsh plugins such as zsh-autosuggestions and zsh-syntax-highlighting.
You should remember that the best AIs can perform at the level of a very smart person on some tasks, but current models cannot provide miraculous insights beyond human understanding.
r1, a Chinese model, is very capable and free to use, but is missing a few features from the other companies and it is not clear that they will keep up in the long term.
Little is lost by preventing a 14-year-old from accessing TikTok or Snapchat, or telling a 10-year-old they cannot have unrestricted access to the internet through their own smartphone, but so much will almost certainly be gained.
But, it’s also now clear that this field is no longer actually divided on the question of whether, generally speaking, smartphones and social media are bad for kids.
Shoulder generation (in late summer) of any given year is generally a higher share of global electricity demand than the peak generation (spring) of the year prior.
A surprising number of my favorite books were not meant to be published. They were private notebooks, or fragments from works-in-progress that never got finished: among them Pascal's Pensées, Simone Weil's private essays, Lichtenberg's waste books, Herzog's Conquest of the Useless, Kafka's diary, Joubert's Carnets, Wittgenstein's Brown and Blue Notebooks, and Sei Shōnagon's Pillow Book.
A surprising number of my favorite books were not meant to be published. They were private notebooks, or fragments from works-in-progress that never got finished: among them Pascal's Pensées, Simone Weil's private essays, Lichtenberg's waste books, Herzog's Conquest of the Useless, Kafka's diary, Joubert's Carnets, Wittgenstein's Brown and Blue Notebooks, and Sei Shōnagon's Pillow Book.
A surprising number of my favorite books were not meant to be published. They were private notebooks, or fragments from works-in-progress that never got finished: among them Pascal's Pensées, Simone Weil's private essays, Lichtenberg's waste books, Herzog's Conquest of the Useless, Kafka's diary, Joubert's Carnets, Wittgenstein's Brown and Blue Notebooks, and Sei Shōnagon's Pillow Book.
A surprising number of my favorite books were not meant to be published. They were private notebooks, or fragments from works-in-progress that never got finished: among them Pascal's Pensées, Simone Weil's private essays, Lichtenberg's waste books, Herzog's Conquest of the Useless, Kafka's diary, Joubert's Carnets, Wittgenstein's Brown and Blue Notebooks, and Sei Shōnagon's Pillow Book.
A surprising number of my favorite books were not meant to be published. They were private notebooks, or fragments from works-in-progress that never got finished: among them Pascal's Pensées, Simone Weil's private essays, Lichtenberg's waste books, Herzog's Conquest of the Useless, Kafka's diary, Joubert's Carnets, Wittgenstein's Brown and Blue Notebooks, and Sei Shōnagon's Pillow Book.
I'm much more comfortable with the fox paradigm. Yeah. So yeah, I like looking for analogies, narratives. I spend a lot of time… If there's a result, I see it in one field, and I like the result, it's a cool result, but I don't like the proof, it uses types of mathematics that I'm not super familiar with, I often try to re-prove it myself using the tools that I favor.
If something holds 99.99% of the time, that's good enough for most things. But mathematicians are one of the few people who really care about whether really 100% of all situations are covered by it.
I don’t believe it’s providing a kind of formal explanation of the different positions. It’s just saying which position is better or not that you can intuit as a human being, and then from that, we humans can construct a theory of the matter.
A korrent is a belief a person has stated in their own words: one
sentence stating the claim, backed by a quote and a source, kept at
korrents.com.
Under a name here, the quoted block is what they actually said.
The korrent beneath it is the claim those words support, in
korrents' wording — tap it to see the record, its source, and who
else holds it.
Nobody here wrote their own korrents. They are compiled from public
statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they
do, this site shows a machine translation beneath the post, in
this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the
serif above is exactly what the person published, and it is what to
quote them on. A translation can be wrong in ways that matter,
especially about tone.
Only the post's own words are translated. A quoted post, a linked
article and a belief on korrents.com
are left in their original language.