I tend to like the ethos/logos/pathos model. Arguments from clever people can sound convincing because ethos gets mixed in. And anyone can temporarily confuse someone by using pathos. This is why it's better to have arguments externalized in a form that can be reviewed on their own, logos only. It's the only style that can stand on its own without that ephemeral effect (aside from facts changing), and it's also the only one that can be adopted and owned by any listener that reviews it and proves it true to themselves.
The whole point is to make many such predictions and experience many outcomes. The goal is for your 70% predictions to be correct 70% of the time. We all have a gap between how confident we are and how often we're correct. Calibration, which can be measured by making many predictions, is about reducing that gap.
Would o3 pro be the first one that can reliably understand a gigantic congressional bill, to the point where it could analyze and warn of side effects?
Would require the bill to be short, or otherwise made ingestible. And also would require an analysis of relevant inter-related statutes and precedents.
Legal analysis is challenging because it's like wordier code.
the "Big Beautiful Bill" is 350K tokens. O3 Pro's context window is 200K, but you also lose performance as you get closer to the max.
It could analyze a section but you still have the challenge of finding relevant laws and precedents.
Weird article, it mentions three apps at the beginning, doesn't even test two of them in favor of two others it didn't mention, and then declares all the apps failures.
One of my favorite things to wonder about is if our historical tech tree has ever been truly limited by anything other than human ingenuity. Like, if not for the random placement of smart people in the right places, who's to say we might not have started launching rockets 500 years ago, or 500 years from now? And how disjoint can branches of a a tech tree really be; are there branches of our tree that could conceivably be completely dark ages and undiscovered given where we are in the other branches? But it's the first question that is most fun: have we ever been externally gated, like where a certain idea or technology was just completely impossible to invent discover until something external happened like a meteor hitting the ground, or a volcano erupting, etc?
In the book Guns, Germs, and Steel, Jared Diamond develops a theory for why Europe and Asia developed vast, powerful civilizations, but Africa and the Americas did not.
In Africa, the co-evolution of primates/humans and other life resulted in that other life becoming extremely dangerous to humans. My impression is that if Africa were the only continent, humans would still be living in caves, because nature had enough time to adapt. Even today, we do not have the technology to tame the wildlife and diseases of Africa, nor will most crops grow there. So the ability to escape the continent would have been the first “external gate”.
In the Americas, the geography and north-south orientation of the continents, as well as the lack of work animals and crops suitable for agriculture, gated development until much later than Europe and Asia. Corn took a very long time to go from being an itty bitty little thing to a crop capable of supporting civilization.
Guns, Germs and Steel is a pretty terrible attempt at a theory. The largest East-West band of climatically similar areas isn't Eurasia, it's the Sahel in Africa, which incidentally does have quite a bit of native agriculture, including a plant extremely similar to corn called sorghum. The near east where European agriculture originated was actually a complicated patchwork of microclimates leading up to the neolithic, which cultures like the natufians essentially arbitraged. If we go down that pathway you end up at something like the vertical archipelago model, developed to understand Andean agriculture. Corn also wasn't the first crops domesticated in the Americas. Squashes were, around the same time as the near east was experimenting with vetches (because cereal grains hadn't caught on yet there either).
So on and so forth. Diamond didn't put a lot of effort into making his theory actually fit the evidence, he just wrote an engaging book.
Could you suggest any good books which go over this kind of global history? Until now I’d also accepted Diamond’s work as the leading theory for global inequality, since it was taught in school and I never looked into it further.
There are few good books of this kind because it is so remarkably hard to construct grand narratives rigorously. So you get people like diamond stepping in it.
Braudel is a combo of geographic and economic history that is still respected (if old).
There's no narrative of world history that's universally (or even widely) accepted by all scholars. Many reject the idea that there's even a single unifying explanation.
With that said, if you're interested in a relatively neutral world history, I'd recommend the Cambridge World History. Failing that, you could do worse than Wikipedia.
I suspect you won't find that answer satisfying because it deliberately avoids "explaining global inequality", which no single book can do. A short answer that is definitively not the explanation, but most would accept as a major part of an explanation is colonialism. Europeans enjoyed some early successes redirecting existing institutions like those of the reconquista into colonial ventures. Those ventures extracted vast amounts of wealth that colonial powers were able to leverage via luck, historical happenstance, and the institutions evolved to maintain those ventures into a transformation from relatively poor backwaters in the 12th century to an extremely wealthy set of nations. Basically, they snowballed. Even this admittedly abstract answer is too reductionist and misses a lot of narratives many academics would be quick to point out. For example, Machines as the Measure of Men makes the argument that we observe great inequality in part because post-colonialism we've basically defined inequality as "differences from the Western model" rather than by any objective reality of poverty.
Energy and materials. In particular, discovery and exploitation of hydrocarbons changed the entire trajectory of human technology development.
It is very interesting to wonder if we may have seen "convergent evolution" towards certain technologies with different physical circumstances, but I think it's safe to say our species' (and planet's) future is forever altered by the massive biogeochemical battery, trickle-charged over eons, which we are discharging as rapidly as possible.
I'd say that the more proximate factor is knowledge. Fossil fuels have been around longer than humanity, and the earliest human usage of them was prehistoric:
The oldest intentional use of black coal was documented in Ostrava, Petřkovice, in a settlement from the older Stone Age on the top of Landek Hill. According to radiocarbon dating, the site falls within the period 25,000–23,000 years BC.
The Industrial Revolution wasn't when people first learned about fossil fuels, but it was when we started turning heat into mechanical work at scale.
Agreed, knowledge of how to use an energy source is as important as its presence. The former will not have the opportunity to emerge without the latter, though the former may be the more proximate cause of a "revolution". (I think it's important to emphasize material inputs because it is a common and very consequential misconception that prosperity is driven by human ingenuity alone.)
In a different "branch", transportation may look completely different.
Electric cars were invented 200 year ago, and the first rechargeable battery in 1859. I wonder how things would have looked like, if we had focused more on EV.
In a related scenario, if the car lobby hadn't fought so hard to prioritize personal cars instead of public transportation, we would probably have a more efficient society where you could live far away from your work, and commute in 500km/h trains. The towns could become more walkable and condensed. Logistics would be simpler with a more hub based distribution where more of your customers live in the same place. More area could be available for agriculture or preservation of nature instead of being used for highways and spread out metropolitan areas
Sure. Someone could have invented gunpowder at pretty much any time. It's not particularly complex, just no one figured it out until (probably) some Chinese monks discovered it accidentally in the 9th century. But in some plausible alternate history the Roman legions were using explosives and cannons to conquer the world a millennium before that.
Wooden cannons do work, albeit with much reduced lifespans and maximum power. Even once metal cannons took off wooden cannons remained common for city and fortress defenses for quite a long time because they could be made cheap and as-needed and in great number, but obviously a wood log with a few wood straps around it aren't going to survive or be preserved for hundreds of years. The straps get recycled and new logs are found after they are either used or sit around for a few years to rot. Metal cannons did eventually win out eventually of course thanks to their accuracy, longevity, and improved metallurgy that made them more eventually more powerful than wooden equivalents, but it did take quite awhile for them to completely outclass and outrange wooden cannon, and wood cannons still had limited use up into the 1800s.
The oldest bell foundries do seem to happen at around the same time as the emergence of what we'd call a cannon, but before that there were indeed bamboo tubed contrivances and similar. If the Romans did have gunpowder earlier though, I bet they'd find a way.. I mean you don't hear about stone cannons and they'd be heavy AF but why not? These people basically invented fracking in BC, and literally tore down mountains to shake out all the gold. https://en.wikipedia.org/wiki/Las_M%C3%A9dulas
More than just human ingenuity than human political organization. Building modern technology requires all the amenities a moderns state affords it, from capital financing, to complex supply chains, to modern bureaucracy, to rule of law, to a willing populace that can conform to rigid roles. Stuff like that takes hundreds of years to build, not so much from innovation but the accumulation of trust and conformity, and the necessary defences to prevent outside invasions from nomadic warlords like the Mongols.
I’m being pedantic but while external factors definitely can cause big jumps in tech, everything is always interconnected. We can’t discover rocketry without previous leaps in material science, those were impossible without manufacturing and mining infrastructure, those not possible without agriculture and sociological structures and organization, not to mention all the associated advances in mathematics etc
That's exactly the right level of pedantry. ;-) Right, I'd argue all of that is in the human ingenuity category.
But in contrast, if we knew that some invention like grinding lenses or whatever was completely impossible until some meteor or fungus hit planet Earth in year N and introduced a new element we were able to use, then that's an external factor.
I suppose there more pedantry is possible, like we could have invented space travel meteor-hopping tech by that point and discovered the "new element" for ourselves, but that's probably the wrong level of pedantry.
I think the closest practical answer is probably more along the lines of population density, and arguing that certain inventions would not be created until the density was enough to create a problem justifying its existence.
The answer is clearly yes IMO. We probably wouldn't have nuclear fission if it weren't for Shinkolobwe, which by sheer luck had uranium ore at a concentration far greater than any uranium ore ever found anywhere else on earth. Surely there are other technologies we're missing out on simply because we don't (or didn't at the time) have the raw materials.
We can certainly imagine a limiting factor based on our prior development. Imagine a world where large amounts of coal or oil never formed, or formed in a manner that made it inaccessible to a pre-industrial revolution society. Without easily accessible chemical energy, the technical progress made during the industrial revolution probably takes 1000 years instead of 100.
The age of discovery was a similar scenario. Imagine a different world where 2/3rds of the continents are uninhabitable desert/tundra/arctic, and there was no economic benefit to better ships, clocks, astronomy, cartography. Delays social development of joint stock corporations etc...
> ... if our historical tech tree has ever been truly limited by anything other than human ingenuity.
I feel more like our tech tree was pretty much directed by resources availability and human ingenuity didn't play much role at all and any species of roughly 50-150 human IQ would follow pretty much the same path when placed in similar situation, only perhaps slower or faster.
Another interesting idea that I'm not sure if I believe in is that rate of our technological progress might be mostly explained with population growth, as if technologies didn't really accelerate progress on future ones.
> tech tree was pretty much directed by resources availability and human ingenuity didn't play much role at all and any species of roughly 50-150 human IQ would follow pretty much the same path
Huh? There are inventions that happened by mere coincident, or two "right" humans at the same place and the same time.
Penicillin, superglue and microwave ovens are just some examples from the top of my head, that probably wouldn't have happened if they weren't "accidentally" discovered. It would seem many inventions are driven by "Hmm, what happened there?" accidents, and since the right person actually paid attention, they dove into it rather than dismissing it.
But that proves the point: coincident discovery implies that ingenuity was far less important than environment. There are smart people everywhere, what there is not is the prior knowledge or resources to turn that into new discoveries.
Notably we see way more coincident discovery after the printing press and more distant trade became common place (and centers of trade usually originated innovations wherever they happened).
I doubt we’d be able to avoid discovering microwave heating across 200 years of using RF in the relevant frequency area. Once discovered, applying it to a countertop application seems inevitable.
> Once discovered, applying it to a countertop application seems inevitable
I guess hindsight is funny like that, most inventions/technology seem obvious to us today. Like putting wheels on luggage, of course that's easier and better in every way, but how many years was it between "luggage" was invented, and someone put wheels on them?
It took a little while, but the window isn't quite as wide as it might at first seem. Before solid smooth paved and floored surfaces are ubiquitous, wheels on luggage are of only limited use, and if you are carrying your own luggage you want something that's easy to carry on your back. In a time period where most people rich enough to be doing travelling with luggage have servants to carry it, making the luggage easier to move around isn't very high priority. When cost of labour is low and porters in railway stations and hotels are ubiquitous, wheeled luggage seems less necessary. And at the other end of the window, Wikipedia reports ad-hoc attempts to do wheels on luggage in the 1920s, and actual patents or products (either for a wheeled suitcase or a "put wheels on your existing suitcase" add-on) by the 1940s.
I think I'd be happy putting this in the "inevitable" category -- the idea floats around and maybe it takes a few decades for the market to be wide enough and somebody to have a good implementation and marketing to make it take off as a consumer product, but if the specific people who did that in our timeline hadn't managed to, somebody else would have.
It's not the placement of smart people, it's the political climate.
When an idle ruling class of parasites siphons all the productive surplus of society, there's not a lot of room for innovation... Unless they come up with a gentleman's compact to dabble in gentleman scholarship.
I don't know why you're being downvoted. The political climate, and more broadly the culture of a society, is the crucial component to whether or not technological progress happens. Evidence of this is all around us, even today. Look how much more progress there's been in South Korea vs. North Korea over the last 75 years. That is a direct consequence of the differing cultural and political climates.
Guns, Germs, and Steel: The Fates of Human Societies by Jared Diamond addresses this head on - resources, safety and political climate seem to contribute a lot
Reading this, I'm struck with a familiar feeling. I wonder if other dem-leaning readers can identify. I remember the old joke about how economists have successfully predicted five of the last three recessions. Now it's more intense and political. I feel like there's so much doomsaying amongst those who aren't wired to support Trump. Like they/we are successfully predicting 57 of the last 14 emergencies. Here, Krugman points out some real financial data that might be concerning, but then also devotes equal time to the freaking name of the bill, which, come on. Completely irrelevant to understanding what is really going on. And, I know that that is part of the point/strategy, that some of this branding is trolling to distract people. But where is the actual sober analysis that is better able to tell the difference between the 14 emergencies that are real and the 43 that aren't? That's the familiar feeling; just the sense that I don't really know what is going on, and there are fewer and fewer resources that allow me to make sense of it.
People complaining about the name baffle me; the name of the bill is irrelevant, only the legal effect matters.
I’ve been doing a lot of independent research myself since finding useful, non-partisan sources is a huge issue anymore. I’ve found LLMs to be very useful in helping[1] me navigate these areas so I can learn enough to make my own determinations.
For what it’s worth my opinion is that this country is in a very bad position that could very well lead to massive world wide economic problems in the future. When and how that manifests is impossible to say; there are a lot of possibilities but I’m expecting the next decade will show us the path we’re on and what that looks like. The tariffs are having an effect (which is to be expected) but it didn’t start there nor does it end there.
I think one of the biggest problems we face as a nation is that the citizens do not understand what’s going on and end up parroting what someone else said without really understanding it, usually filling in the gaps of their knowledge from their own world view rather than reality. I don’t mean this as a denigration as most people have families and jobs and do not have enough time to know enough to have well-informed opinions about everything, myself included. However, it is leading to absurdities like screeds about the name of the bill.
[1] I don’t trust the LLM of course. I’m not asking it if X is happening or if Y is a problem. More of a way to guide me to finding good primary resources.
So if not graphql, then what's the latest "in-favor" thinking to solve the problem of underfetching and overfetching? Especially in an environment with multiple kinds of frontends?
It's just still so complicated when you are dealing with any level of digestive issues. Sometimes you're supposed to lay off high-fiber fruits like apples. And then there's soluble versus insoluble fibers (sometimes you need more insoluble), gut flora, probiotics; fats are both good and bad for gut "motility"... if your balance is off somehow it just gets really complicated really fast.
Along the lines of predictionbook, metaculus - something that helps you be "well calibrated", but more playful/fun than metaculus.
It doesn't have a lot of upside - predictionbook actually went offline due to lack of interest. But it was a good excuse to try out some vibe coding, and learn react native (I've mostly been a backend programmer).
In an attempt to make it more engaging and fun, I decided to have it focus on sports picks. Also partly because calibration graphs need to have a lot of predictions to yield any reliable information about your calibration.
I got it up in time for March Madness and about 25 of my friends joined and it was a good time. I nagged and reminded them a lot about about 15-20 of them predicted all 63 games, by picking the winner of each match and what their percentage confidence was. I had a leaderboard and live-blogged and gave silly awards.
I later added support for multiple "tournaments" and currently have tournaments going for NBA Playoffs and NHL Playoffs, but interest is waning. Of my friends, only 2-3 others are still regularly predicting.
Maybe it'll be more fun for the NFL season but I might also let it go a bit dormant.
Biggest challenge is that there isn't really a bulletproof way to rank people if people only predict some games in a tournament. I've tried all sorts of things, minimum # of games, bayesian kernel smoothing, but it's ultimately arbitrary when choosing how to penalize someone for not participating.
If I were to continue I'd be looking at things like automatically integrating with sports apis and odds/bookmaking apis, allowing users to create their own tournaments, etc. But ultimately, the UX of the site isn't much more than making a prediction, and then checking back later when the game is over to see your score. Not much more reason to hang around on the site than that.
