The elusive 100%-coverage formal spec that covers every intended and unintended input, the interests of every stakeholder, and every possible user behavior? In the vibecoding era? No, the page has the decency to acknowledge this doesn't exist.
In the end, they settle on finding crashes with a conventional fuzzer, fixing crashes, and then... writing a formal proof that the crash is fixed? That seems... pretty pointless? Am I missing something here?
> Proving that a fuzz test has no more crashes to find is more significant than it sounds. It’s common for fuzz tests in real-world code to test for violations of meaningful high-level properties, such as encode-decode roundtripping, idempotence, algebraic laws, differential testing, self-consistency, or equivalence to reference implementations. For example, a fuzz test for the LZO compression algorithm asserts that if you compress an input, then decompress it, the result is the same as the original. If the property does not hold, the program crashes—so proving the test never crashes is equivalent to proving the property itself.
However, you're correct that this is not formal verification in the sense of a correctness proof against a full specification validated to capture what the program is intended to do. Pretty impressive though.
In essence this lets you verify any property you want for a given function, as long as it can be expressed as a fuzz test. Specifically, you can express your property with an assert, and then prove that it never fires. This is nice because it keeps everything in one language, rather than having to translate the whole program into Rocq and then express your desired property as a Rocq theorem.
What they did is tried to prove that the conventional fuzzer won't find crashes. Where they found counterexamples, they fixed the bug and tried again.
In essence, a fuzzer takes random samples out of a space of possible inputs that could crash. They proved that nothing in the space of possible inputs could crash.
This is a strictly more powerful result than a fuzzer could give you.
Yes. You are missing something. They are proving that for all possible inputs the program doesn't crash for a given size of potential inputs at the machine code level. The proof extends to an entire set of inputs and all the code it touches, not just the cases the fuzzer covers. In short, they have better coverage than any fuzzer.
Turning buggy software A into buggy software B, with all bells and whistles (and bugs included). This is useful for doing large-scale restructuring of codebases. You can verify you aren’t adding (or removing) bugs.
Actually, bugs might surface during such restructuring. If the statement you're proving can be greatly simplified to something reasonable, if we omit this one weird special case, that indicates a bug in the specification.
Especially when "this one weird special case" pulls in the entire binary layout because your code has a hole that allows an attacker to jump to a specified address
The main thing they focused on in that is just cybersecurity. Formal verification to try and get rid of surface area for vulnerabilities. Seems like a good idea idk. It also sounds like they're just kinda getting started with it - we'll have to see what opus 5.6 or 5.7 looks like after they've RLed on it a bunch
I wouldn't interpret the chart as indicative of the gains in the quality of AI writing; it's more that it's a fairly new phenomenon and there is a dramatic increase in people who try to smuggle AI stories into contests (be it for personal gratification or to "show" human writers), and an increase in the scrutiny that submissions receive.
Fundamentally, I know that HN has plenty of people who will react with "ha-ha, human authors should pack their bags", just like what we keep telling mathematicians, because we see the discipline as just another market that can be conquered by automation or another benchmark to track. But literature, similarly to most other forms of higher-brow entertainment, isn't just about reducing unit-cost or hacking engagement metrics. For the same reason why a genuine van Gogh painting is worth a lot more than a nearly-perfect replica, people who care about the craft usually care about it telling genuine human stories and celebrating the author as much as the result.
Now, the mass consumer may not care; if you have an infinite Harry Potter sequel machine, many will partake. But then, I guess the question is, if you build the sequel machine, are you leaving the world better off or not? Do you want to be interacting exclusively with machines down the line? And if there's no inherent value to humanity, why should the machines be compelled to interact with you?
«And if there's no inherent value to humanity, why should the machines be compelled to interact with you?»
Reading this part made me wish it was more compelling than I think it really is, because I think someone could argue that the value of human consumption isn't predicated on the value of human creation.
I used to find things like machine-run pizza places and coffee shops really cool (even though I've never been to one). I even wrote a small coffee automation program once, but recently, with the promise of automation getting more real by the day, I find myself genuinely worried for the fate of human-run pizza places! They're doing fine now, but what if the machine-run ones become 10x or 100x cheaper? Would anyone still occasionally pay for the expensive human-run ones?
While it's debatable how that will turn out, the value of human creation feels a lot more significant when it comes to art and literature, but I'm starting to realize that for many others there's no fundamental difference between consuming art and eating pizza.
The fine dining scene is doing well despite cheaper alternatives being available (ubiquitous?). So I wouldn't worry about the possibility of experiencing amazing hand crafted pizza disappearing, even if convenient cheap machine made frozen pizzas should become what most people choose experience regularly. And I believe the same goes for other art forms, people enjoy producing and consuming them, and the experience of top notch human productions will be available and no worse regardless of what the masses choose to spend their time and energy on.
Where we might get fucked over is if Bauhaus displaces architecture where some resources are allocated to form instead of all going to function because buyers are cheap. It would be impossible for individuals with taste but little resources to set out with to meaningfully affect the public landscape. But that ship has sailed almost a century ago.
You brought up fine dining and frozen pizza but what about a situation where the automated one is both cheaper and better tasting? We're still functioning on the assumption that that will never happen.
The fine dining was a reference to high art in general, the frozen was directly addressing the worry that pizza places with human made pizza would disappear given cheaper better alternatives. I think the picture still holds. I've been to places that serve human made pizza where I think I would prefer some of the better frozen ones food-wise, but still went and enjoyed the experience (mostly for the social component?).
Similarly, a lot of the fine dining is about showing off an artistic creation that surprises and/or delights, not necessarily the food-artefact beating out others in price nor quality. (Ok, there's the whole show around it as well, but even the places light on the craftsmanship that goes into the ceremony are hard to imagine succumbing to cheaper and better/equivalent alternatives assembled/delivered home, which arguably exist.) I foresee the other arts now competing with LLMs having a similar fate to the fine dining scene. Not capturing the masses, but still doing fine.
Or perhaps ultra small scale brewing of beer would be a better picture now that I've slept on it. It follows the same pattern of most consumption being mass produced good cheap product, but somehow artists putting awesome hand made creations into the world is happening both commercially and by people doing it purely for their love of the craft, despite the prices being higher and the technical quality not rarely lower.
>Fundamentally, I know that HN has plenty of people who will react with "ha-ha, human authors should pack their bags", just like what we keep telling mathematicians, because we see the discipline as just another market that can be conquered by automation or another benchmark to track.
I think it's a bit more nuanced than that, at least some of us are going: "ha-ha, people derive their value from their outputs instead of understanding existence is itself valuable"
Why does everyone value themselves based on their economic output? Why do blood diamonds compete at all with lab grown - one often involves sacrifices to the minors, and the other is cheaper and by all metrics better? (hint: because the narrative of "effort" is what drives that particular example... the blood makes the diamonds shinier in some people's minds... that's not a good look, morally)
It's not the economic output itself, but what it represents. I can value myself as a human being all I want, and I should, but without that economic output, it's a much harder life. Money gets you food and a roof over your head. More money gets you better food and a better roof that doesn't leak and a more secure one as well. (also gets you more problems)
Gets you a car that runs, one that runs reliably instead of breaking down. A parking space for it, even in the center of a big city.
It represents security, happiness, vacations, things/stuff. We can judge however we want about those categories but judging people isn't going to change that do actually want them. So maybe you can get all that without economic output, it just seems like it would be easier to accomplish/achieve with than without.
Diamonds are expensive because that's how much money people pay for them.
People buy diamonds because [they have no taste and] they can afford to, or want to present themselves as being able to afford to.
I'm sure some of them have various dark thoughts about it, too.
People derive their value from their economic status because that is a tool available to them and the world they grew up in forced them to do so. Being able to derive your value from other aspects of life is a luxury and a privelege that is, ironically, unaffordable for the majority.
> Why does everyone value themselves based on their economic output?
I mean this is really the issue with the idea of AI taking jobs, isn't it? If our entire economic system wasn't predicated on this, nobody would be really that miffed about AI use in art because there wouldn't be much of a market for it, because the concept of art as being a marketable widget wouldn't exist in the first place, and the notion of human expression being replaced with a machine would be absurd. But we have tied up art with economic exchange for centuries. It has been a market commodity for almost as long as recorded human history. So the graphic designer who pours their creative spirit into what is, essentially, an economic exchange (rather than going and painting landscapes or something) feels an acute threat because it threatens their existence.
To overcome this and to live a future where it feels more like ML/LLMs live in concert with humanity then you'd have to picture a future where human value is divorced from one's capacity to provide economic value. When exactly does anyone foresee that happening? The entire system would crumble. UBI is clearly not the answer because all it does is entrench a subtle cast system and limit upward mobility and does nothing to address this root cause.
I don't know what the future is, but I know that as long as the current economic system persists (in all its permutations, whether that's socialism, capitalism, communism, whatever outmoded 19th and 20th century ideas we have to structure society -- all alienate the individual and relationships to some degree or another) then AI will always be met with disdain even in areas where it can improve people's well-being (and FWIW I think the arts and even mathematics are NOT part of this subset of use cases).
Maybe the solution is walled cities and shutting off the internet entirely; returning to our neighborhood pubs, sports, expressions of the human spirit placed up at farmers' markets. Seems like we already seem to be trending in that direction; and if last night's apocalypse was any indication maybe the internet's days are numbered anyways.
> UBI is clearly not the answer because all it does is entrench a subtle cast system and limit upward mobility and does nothing to address this root cause.
what. how?
UBI allows bad jobs to end. It allows workers to look for other work without threat to their economic conditions. Nothing about that is a cast system or limits upward mobility. it in fact, increase that mobility.
My optimistic take on literature, compared to code, is that what writers produce is what is exactly consumed. The medium stays the same. The implementation choices of writers, like syntax, matter deeply to how the text resonates with a reader.
Code, on the other hand, is not consumed by end users in the same way. It compiles and turns into applications, which are then used. So users of applications are less scrupulous about what the code looks like, what language was used, and so on.
The post is about ai literature winning awards. Since an award is supposed to be a measure of quality, not unit-cost, a trend of ai winning awards is either indicative of the quality of the writing or of a systemic flaw in how these awards are judged (which seems plausible too).
A van gogh painting is more valuable than a replica because 1. it is scarce and 2. owning one conveys prestige. A perfect replica of a van gogh painting conveys the same artistic expression as the original.
> a trend of ai winning awards is either indicative of the quality of the writing or of a systemic flaw in how these awards are judged (which seems plausible too).
I definitely think this is plausible, and I think we need to start extending the idea of code "smells" to regular languages as well. I have TAed for literature and have reviewed many a manuscript and except for really really finely honed projects with lots of editing in post, written text almost always has many of these obvious smells (cliches, fallacies, illogical juxtapositions, poor cause and effect). AI-generated prose is no exception. Character names, phrases, descriptions of sensory experiences that are off or not quite right, improper negation (LLMs still struggles with the idea of 'scope'), sentence structure being very predictable and one-dimensional (even when using the latest models and mimicking famous authors and poets and even with very careful prompting), so on and so forth, you get the idea.
> A perfect replica of a van gogh painting conveys the same artistic expression as the original.
I think OP's example was a poor one. A perfect replica of a van gogh painting indeed does convey the same artistic expression as the original. However, I think the better comparison is an understudy's mimickry of A Starry Night which seems inspired by it and yet adds nothing to the original and is somewhat deficient in some ways, and misunderstands the artistic intent of the original. That would certainly not convey the same expression as the original, and would be met with widespread criticism.
Same for blood diamonds. I think scarce objects with exaggerated monetary value have some extra function for very wealthy people, for accounting or laundering purposes. That's why NFTs were a thing.
Because there are a million literary awards and some of them are awarded by people who are bad at recognizing quality.
Additionally, many of these cases were only partially created using AI or aren't proven to have actually used AI or used AI in a creative and non-conventional way and that's what they won for.
It's a chart that shows the number of things going up over time. The total number of things increasing over time is going to play a hugely significant role in that, regardless of how broken the selection criteria are.
A contest of skill is not necessarily like spinning a roulette.
If I were to play Lebron James in basketball 1v1, it would not matter how many times we ran the experiment. We could play a trillion times, and I would win 0 of them.
If you sampled 10 Hollywood directors and asked whether The Godfather or The Room is a better movie, the Room would win the poll 0 out of a trillion times.
The reasonable naive expectation for contests like these is that, below a certain bar of quality, the chance of win is 0. So if you see wins go up as volume of submissions goes up, it shows that either ai writing is above that bar, or that faulty judging is a widespread problem beyond a few anomalous organizations.
You have better than zero in a trillion odds against Lebron James. What if his phone rings?
"10 Hollywood directors" is hilariously unconstrained -- I'd put money on at least one having never seen The Godfather.
> The reasonable naive expectation for contests like these is that, below a certain bar of quality, the chance of win is 0.
I agree that the layperson's understanding of what 'literary award' means is close to this idea, but the "contests like these" is the problem here. "contests like these" include one that allows and seems extremely keen on "AI" generated text, for example.
There isn't a stark contrast between 'contest of skill' and 'game of chance'. You can eliminate the skill part and get closer to pure random chance, but you can't eliminate the random chance from the contest of skill.
> You have better than zero in a trillion odds against Lebron James. What if his phone rings?
You can easily devise in this thought experiment rules that prevent loopholes, like saying no phones allowed and medical emergencies do not result in a win for the opponent.
You're missing my point. A very good fake of van Gogh can sell at an auction for a lot of money, but if the new owner finds out, they're not gonna be happy. The same applies here: AI entries would not have won most of these contests if it was known that it's AI (discounting the handful of contests that were actually meant for AI literature and are weird to include on the list).
"Scarcity" is a really bad theory of this, and the NFT bros have already tested how that works out.
My point is that the value of most human art derives from something more than the arrangement of pixels / bits, and if we disregard this, we end up in a pretty shitty world.
If someone buys a fake of a van Gogh and find out it's fake, they just lost millions of dollars of wealth. This metaphor seems inappropriate to apply to the typical ways people enjoy art.
> most human art derives from something more than the arrangement of pixels / bits
Many people think art should stand on its own, independent from the artist or the circumstances of its creation. Meaning or quality in art may be conveyed purely in bits even if it is the not individual bits that are meaningful.
> if we disregard this, we end up in a pretty shitty world
This is an inappropriate degree of moralizing. But I think it if we accept policing quiet enjoyment, that ends up in a pretty shitty world.
> people who care about the craft usually care about it telling genuine human stories and celebrating the author as much as the result
No, the human was never a core factor. People who enjoy a van Gogh - or any other work - primarily do so because it has particular qualities they appreciate (no, being human-made isn't an intrinsically appreciable quality; secondarily because many humans instinctively like to possess unique - or "shiny" - things), and the author tends to be a very good means to access other works with such qualities. That "celebrating the author" is just a side-effect, and there's nothing to stop a non-human entity from "telling genuine human stories" (whatever that means).
> why should the machines be compelled to interact with you?
The ultimate purpose of machines is to serve humans, and that's a bit hard to do without interaction. Just as the ultimate purpose of humans - and other living things - is to survive.
> People who enjoy a van Gogh - or any other work - primarily do so because it has particular qualities they appreciate (no, being human-made isn't an intrinsically appreciable quality; secondarily because many humans instinctively like to possess unique - or "shiny" - things)
Speak for yourself. You could show me the exact same artistic image, same exact pixels, but if I know it was made by a human, it will instantly be worth more to me and capture more of my interest. Conversely, having the knowledge that the pixels were created by a machine would make me instantly lose interest in the image.
I think "being human-made" very much is an intrinsically appreciable quality, but you have to include the production of the art as part of the art itself for that to be the case. And I and many other people explicitly do include the production of the art as part of the art itself. Not everyone shares this perspective, but many do.
You said it. Depending on your knowledge of the entity producing the work, you get a different experience. In a vacuum though you'd appreciate the work for it's actual qualities which you find appealing. The potential for appreciation isn't different at all; it's just overruled by stronger feelings regarding the current trend of AI-hate vs AI-love. Like given identical works from 2 humans, if you respect human A and despise human B, then it's a near 100% certainty that you'd appreciate the work attributed to human A and find every possible fault with that attributed to human B. With a fun bit of hypocritic confusion if it's the same work with the attribution going back and forth.
Also the point about humans preferring the unique and "shiny". Human-made things will always be unique (nobody can do the same the the exact same way twice without machine help), while machine-made things are usually not unique but can be made so (and even then a generative routine used to make something unique can be replayed).
I think it all comes down to scarcity, the same thing that's driving pokemon card prices through the roof. That's all there is.
As AI can generate ~any image nowadays, there is potential for an infinite amount of AI art, driving the value of AI generated images to 0, unless artificial scarcity is introduced like in the case of Pokemon cards.
Whereas with humans, there isn't an infinite amount of human art, as only a subset of humans are skilled enough to create good art pieces. Human art should therefore theoretically always be more valuable than any AI art. EDIT: Have to add, human artists always make use of artificial scarcity as well by only making 1-N copies.
I value this scarcity, that there are people insanely more skilled than me, blessed with good genes and environment that are able to create art that didn't exist before in the space of human made art. It creates healthy role models for people to be inspired by. The scarcity or skills aren't artificial, but natural.
You seem to be conflating "intrinsically appreciable" and "intrinsically deducible". We can and do appreciate artworks because of their extrinsic qualities such as its historical background, the works that preceded and succeeded it, meanings given by author etc. So the "potential for appreciation" is indeed higher for human beings with this web of interconnections. The vacuum example doesn't do much since it strips the very thing that gives the artwork its extrinsic meanings that we appreciate. When we step out of the vacuum the balance changes and the works are not equal anymore and I can hardly understand why we would insist on living in the vacuum when we clearly do not.
To your point that this is merely reactionary attitude on the emergence of AI, people have and do still value hand-made crafts such as carpentry over mass-produced furniture.
> No, the human was never a core factor. People who enjoy a van Gogh - or any other work - primarily do so because it has particular qualities they appreciate
I don't follow. How about the 20k year old cave paintings in Lascaux? Are you saying "What you appreciate about it isn't the craft, or the story and history behind it; you care about only some of it's qualities?"
It's an extremely valuable part of history, for historians, archeologists, etc. The people who created those works too can be considered a "work", something to be studied the same as that which they did/made are studied.
OK, and so other artists can be studied too, and aspects of their creativity can be studied, and this doesn't have to be formal but can amount to just liking the artist's creativity. But you said "being human-made isn't an intrinsically appreciable quality".
That word "intrinsic", what does it mean in this context? You're sure it means something?
Sure, anything that can be sensed/known, can be studied.
It means that one cannot experience (look at, listen to, etc) a work and honestly say "I appreciate how this was human-made". There's no quality that can be imbued into a work that makes it discernable as made-by-human, as opposed to made-by-machine, or even made-by-nature (and eventually, maybe even made-by-alien). Not like one can discern and appreciate the lines and colours in a painting, or the tones and energy of a song. The maker of a work is extrinsic to it.
But there's no quality that makes an object discernable as anything at all; it's all theories. You don't absorb facts about reality through a sense organ, you start with a theory - such as "this is hexagonal" or "this was made by a human" - and test it. I diagnose that you are suffering from inductivism.
What interests me about art isn't just the result, it's the process, and the connection that process creates.
It's more than the "aura of the artist" which Walter Benjamin talks about in The Age of Mechanical Reproduction (highly recommend for anyone interested in past discussions on his topic), it's the fact that the writer is directly processing their world as experienced into an artifact which will allow me to understand their experience and enrich my own.
The loop is:
World -> Human -> Written Word -> Reader
compared to: Corpus -> Engagement Optimized Output -> Reader. This is something David Foster Wallace talks about in an essay E Unibus Plurum - the concern that artists producing art from media consumption rather than direct experience creates a house of a mirrors. Even if the optimization is for literary quality, it's still not coming from the world. Serious art is a kind of news, not just entertainment.
Beyond that, creating art isn't just a technical process, it's an open act of self-shaping and self-discovery that is meaningful to and reflective of a community. I worry about losing the deeply insightful members of society that artistic practises create. Of course if you're a media producer interested in profit that's perfect - they want caged pigs to consume their slop, not thoughtful connected individuals.
For that matter, these first major examples of AI literature—like the first passable Stable Diffusion images, and the first halfway-hummable Suno jingle—have the advantage of cultural novelty. I feel like the shine is likely to come off quickly as derivative AI works multiply.
The first spatter canvas was a new conceptual move and valued more for that than for the paint and canvas; subsequent works tended to ring hollow and command neither interest nor value. The first “ChatGPT, reword this like a pirate shanty” was worth a laugh, now it’s worth an eyeroll.
For that matter, as artists and writers incorporate these new tools into their practice, there’s still a difference between low-effort slop and the products of artistic obsession. AI image gen is commoditized right into the ground, yet you can still find interesting and distinctive works in, say, Midjourney’s community galleries. Or in at least one example from TFA:
> An AI reading machine uses computer vision, OCR, and language processing to find short poetic combinations in physical books, erase the surrounding words, pair them with images, and publish new volumes.
That sure sounds like an artistically interesting endeavor to me, independently of their choice to execute their concept through AI models.
> Fundamentally, I know that HN has plenty of people who will react with "ha-ha, human authors should pack their bags"
And yet they still seem to object to LLM-generated blog posts and LLM-generated comments here, where literary expectations are much lower.
The quality of LLM prose may improve quite a bit, but I suspect LLMs writing fiction will asymptote towards the quality of a writer who has never interacted with other people, but only read other fiction. Even if you dismiss any value of the author being a sapient being to the process, a human author has a far larger training set...
> What is today a moral imperative is not the same as what was moral 50/100 years ago and is not the same as what will be morally acceptable 50/100 years from now.
This is a weird take. If you read Western treatises about ethics / morality from 50 years ago, you'll be hard-pressed to find differences, except that more folks in the West openly embrace relativism nowadays (i.e., "there's no objective morality, the purpose of life is to make money with friends"). Probably because it's been a while since we had a clear-cut enemy to fight.
If you go back a hundred years, chances are, you'll find the writing eerily on-point too. In fact, you'd probably agree with most of what Ancient Greeks believed about the "core" stuff: the value and purpose of human life, the nature of the state, crime and punishment, so on. Maybe not so much on gender roles or slavery, but even that stuff was more nuanced than we imagine.
> But the machine has the benefit of being able to look inward at its state honestly, we mostly do not.
The machine replicates human behaviors and is prone to similar fallacies and judgments. There's nothing to indicate that it has any sort of transhuman morality.
Yeah, wealth has legs, and we increasingly have this wealth meta-economy where different states are good at producing wealthy people (CA, NY) and different states are good at giving them a hospitable place to live. Sources and sinks.
This is probably good in some sense - you can argue that it's not healthy for states like CA to "own" so much of GDP end-to-end - but I wonder where this leads us in a couple of decades. Does it continue to be like this, or does it mean that more tech hubs pop up elsewhere because in the long haul, people probably prefer to work and invest close to home?
A big problem is that some of these places - e.g., Wyoming - don't have the higher-ed infrastructure that would foster new startups and high-tech research. On the flip side, California is now importing most of its entrepreneurs anyway.
They're not 'hospitable,' they're 'advantageous' due to being lower tax rates. If anything, lower tax rates make places less 'hospitable' in the long run.
That’s not true. For example, Washington State has no corporate tax: https://taxfoundation.org/data/all/state/state-corporate-inc.... But companies aren’t exactly flocking there. Florida meanwhile is seeing a huge boom, but its corporate taxes are around the middle of the pack. California has always had pretty high marginal tax rates, even back in the 1970s when it was a solid red state. It used that money to do things like make big investments in higher education.
Being hospitable is more about whether the state makes it a pain in the ass to do business.
Washington State has a "Business & Occupation tax" on gross receipts rather than a tax on net income. Corporations are absolutely paying taxes here, and if you look at it as a % of net income, it's often more than a traditional corporate tax would be. https://dor.wa.gov/taxes-rates/business-occupation-tax
Taxes are the least worst part of doing business. It's all the other subjectively defined and capriciously enforced regulatory BS that drives business away. You can't run a business let alone get financing to create one when a huge chunk of your costs are a question mark.
The type of people who move from CA to WY for the tax benefits are the type of people who pay more taxes than they use in government services (even with very low tax rates). That means WY can offer more government services to the rest of its residents, making it more hospitable.
> who pay more taxes than they use in government services (even with very low tax rates)
They rely massively on government services. They depend on an educated workforce. They depend on law and order to prevent their workplaces from bandits. They depend on massive navies that allow goods to be shipped across the world without concern. They depend on courts to enforce contracts so that they can build their business on contracts rather than having to establish personal trust with everybody they do business with. They depend on a stable currency. They depend on a regulated finance market. They depend on a massive pacified customer base.
I've only scratched the surface, there's lots more.
Their business in California, most of their income is generated in California. California is bearing the cost, Wyoming is getting the income tax revenue.
>The type of people who move from CA to WY for the tax benefits are the type of people who pay more taxes than they use in government services (even with very low tax rates).
Ah yes, the government services the people will need after these locusts move in and price generations of locals out of their homes.
Not always. Low tax can allow enough economic activity as to bring in more income than higher taxes. This is most obvious when you take a tax rate of 90% and reduce it, and there is big debate on where it breaks down when rates are already lower.
A state can justify and sustain higher taxes on the wealthy by offering government services which indirectly benefit the wealthy (the externalities you mention). But it doesn’t follow that an arbitrarily poorly-ran government can sustain arbitrarily high taxes on the wealthy.
There’s some manifold of Pareto-optimal solutions to the problem of taxes and services.
California, for example, spends a lot of money to get not-great public schools, and has housing costs that are out of control. So if you’re importing new hires from other states, and pay them a lot to live here, and personally pay high taxes on top of that… that doesn’t seem like a Pareto-optimal point.
At some point, the wealthy will decide to move someplace that looks more Pareto-optimal. Better externalities, lower taxes, or both.
That's not the issue in question though. I mean, everyone wants zero taxes and everything a good government can deliver them with, of course, none of the bad things government can deliver. The real question is what is the right compromise between everything that fits your needs. Is it really worth having sports stadiums? Maybe the government should provide a professional sports team, maybe not. How much should the government put into your health care? I mean, we always hear about the so-called death panels when you have a government-run health care system, but regardless of how much that is just pure mongering, the reality is there is always going to be more money you could spend in health care. Is it worth it for any particular question is something up for debate.
FTA: For years, policymakers and economists have debated the shape and peak of the Laffer curve, a theoretical model suggesting that if top tax rates increase beyond a certain point—called the revenue-maximizing rate—they will yield negative marginal revenue.
A new paper from economists at the Joint Committee on Taxation finds that the Laffer curve is flatter than previous theories have suggested, meaning the revenue-maximizing “rate” might be a broad range rather than a single point.
Lawmakers looking to raise taxes might find it difficult to raise substantial revenue from top rate increases; however, broadening the base, examining untaxed sectors, and reducing the tax gap represent promising pathways to closing the gap between revenue and spending.
---
I'm glad they highlighted that instead of increasing the tax rate, broaden the tax base. For example, if they treated borrow-spend-die as income. We could keep a lower tax rate. The same argument could be made for eliminating the estate tax. Treat inheritances as income for tax purposes.
Those states will eventually either kill the golden goose or make themselves unattractive for the next big industry. California for most of the 20th century was a place where the government was well managed and efficient, wasn’t afraid to make big investments in infrastructure and education, and made it easy to build stuff without too much hassle. Texas and Florida kind of fit that niche today, but both are a shadow of what California used to be.
Wyoming is an interesting case with Higher Ed. It's actually the legislature that runs UW. There isn't a regent board or some such thing. Hence, with many other priorities, the Cowboys don't get that much time in the minds of their bosses.
It's one of the many many wrinkles that come along with the many college football issues these days. Any formation of a players union or draft or much of any real professionalization of the sport has to go through so many different governance formations (regents, governors, the Catholic Church, etc). As such, the many issues with college ball are likely to continue for a long time and fester.
It doesn't have to be this way, and it wasn't at some points in the past for different states. Seems like the sources keep forgetting what made them both sources and sinks.
California imports it's current entrepreneurs because it chased the previous generations of them away. I was a fourth generation Californian, was...
I know one literally one person in my highschool graduating class who stayed behind.
Sure the older generation are still left for the most part, still not admitting they did anything to cause the situation, but the young and ambitious have all left.
Think about that the fact that the entire population of a place has basically been replaced in under a generation, and with people who are chasing dollars and clout not people who are building communities. (And no this isn't about immigrants, Cali was always full of people from everywhere, it's about the personality that it attracts now.)
We all left and built other communities and business elsewhere because the place quite clearly told us we were not wanted there.
I'm not sure why people keep repeating fabricated GDP like it's some measure of success, especially when its completly measured on non-tangible products like the California economy.
Cali compared to Cali 40/50/100 years ago is wasteland and it pisses me off every day.
I didn't say anything about Florida, because that is "no" government. But I've definitely considered Mass. People forget though that when capital takes flight, international borders are of no concern. Singapore and London are on the docket.
(Before anyone gets on my case, I've lived in Both CA and London, and at least the City government is 10x more effective than any municipality in CA)
Internationally there certainly are small countries that specialize in this. Singapore is famous for low tax + effective (if overbearing) governance. The trains run on time. Come to think of it, Switzerland is maybe low tax (though not compared to Liechtenstein or Monaco, and it even borders on .li), and is also known for effective, if expensive, governance.
Good lord - you literally just named the major tax havens of the modern world.
Those places exist for billionaires to go to and not pay taxes or for corrupt regimes to park their money - where it will be protected by commercial law of Western nations - but no questions asked.
Switzerland hosts 1/3 of the world's offshore investment - when Nigerians pilfer their countries (billions go missing from the ministry of energy every year) - they end up in places like Switerland.
They exist with the bounty of the 'free money' of that dynamic.
Singapore is also an ant-democratic authoritarian state.
None of those places are 'self sustaining' they can only exist because of that special leverage they have of 'tax evasion'.
It's much the same in some carribean offshore places.
Trains can run on time if you don't have to worry that-that much about the money.
Dubai also provide many 'material advantages' to residents. Same thing.
Not a very good argument. Even Ireland is a tax haven for US FANG businesses and that arrangement is so much less fair than Switzerland is, because it's literally a tax deal with individual companies, individual Irish people or companies cannot get anything remotely similar in tax treatment. Billionaires only, nobody else, not even Irish citizens or companies. Compared to this Switzerland's tax is egalitarian, in fact Switzerland signed and applies the minimum corporate tax rate, which means FANGs cannot get Irish tax rates in Switzerland. And Ireland cheated: Ireland signed the global minimum tax treaty and, sorry to put it like this but I don't know how to do differently without lying, they lied. They're simply not applying the treaty, with "existing deals" as an excuse, as if they somehow didn't know. Plus Ireland has both natural resources and very easy access to trade, neither of which are advantages Switzerland, Liechtenstein or Monaco have). Netherlands is another example of a country trying to be a tax haven, despite their location, money and ... so no, I don't think Switzerland counts as a tax haven.
A lot of countries will never have a geographic advantage, and need to find business and support somewhere else. Switzerland is one of those countries. So is Singapore. The real EU tax havens (Ireland, Monaco, City of London) don't need anything.
And to add insult to injury: the current Irish political party in power Fianna Fáil explicitly campaigns on their accomplishment of improving Dublin, not for Irish people, in fact they admit their policies are causing rapid deterioration for normal people, but for FANG companies.
So no, I don't feel this is a very good argument at all.
There's also another sticky, malleable substance that could work for this analogy...
But more seriously: the question, the same as before, is "what are you going to sculpt". Sculpting for the sake of sculpting achieves nothing, and if it's too cheap and too easy, it can cost you time and money than if you had to pause and think beforehand.
I implore you to look at the past several months of vibecoded games and Github projects submitted to HN. It's amazing that you can prototype something in hours. It's bleak when you realize that half of these projects weren't even working properly - we often upvote based on vibes too - and are now abandoned in a half-broken state because the author has no attachment to it whatsoever. It was just some money exchanged for internet points.
There's no doubt that we're going to have more software; I think it's very much an open question whether there's going to be more good software, or more software that can make money.
I agree. My hope is that the half-broken made-in-a-day vibecoded demos getting attention is just a phase. Those things look impressive because they would have been impressive 1-2 years ago, but when people's brains adjust to the new normal they won't be any more interesting than the food you cooked for yourself last night or your creative writing class assignment.
Except because it's some sort of a vibecoded atrocity, nothing happens if you try to toggle from "all" to "last 3 days", it only works if you change the value (e.g., to 2 or 4 days).
Stuff like that is really frustrating because it's not like we didn't have one zillion decent, human-tested graphing frameworks before.
I think it's basically an alignment problem. LLMs tend to think it's better to just do something small and bespoke than to just pull a mature library and use it correctly, which, to be clear, they are perfectly capable of doing.
And this is in part a reflection of our own bad preferences. This is what always annoyed me most about the left-pad Discourse a decade (oh god, a Decade™) ago; "why do you need a library for something so simple in the first place!?" people would scoff. And of course the answer is that even the reference class of left-pad-like problems includes a bunch of examples where the problem has bizarre and sticky edge cases you'd never consider, so in general, "just pop in a popular and well tested library" is actually a pretty good policy (assuming your package repository doesn't allow anything really silly like deleting entire packages and breaking everything downstream).
Of course charting is more obviously complicated, but again, it might seem fairly simple if all you're thinking about is one kind of chart in one style with limited settings. The difference is that for a human coder, "just whip up some chart code" is enough of a pain in the ass that they'll reach for a library by default, whereas a coding agent can just spit something out which mostly works in a few seconds.
All else equal, I'd rather have the richest possible functionality built into the language. At least I know that it will be updated and maintained in one place, and there's very little ambiguity about setup. You don't have to start picking apart the fact IN THIS PARTICULAR PROJECT, libfoo was imported by pasting it in the code base in 1977 before package management existed and globally namespaced, but libbar is a first-class citizen in the depency-injection system.
If you know you can just call bargraph($data), even if the stock bargraph library is boring, then people and LLMs will favour it.
I suspect they used an underpowered model, or did it a while back. I would be surprised to see 6.1 Sol or Astra produce those bugs, even one-shotting. Instead, I'd just expect it to have broader product issues rather than out and out bugs.
I've seen many posts like this. None of them stood the test of time. Ease is irrelevant because you only need expertise once and then everyone can reuse the result.
It's also not easy to pop an iPhone, but reliable, weaponized exploits are available in trade and governments use them in a plug-and-play way.
The specific controversy here is about the disclosure timeline. Some people allege that the disclosure and patch release of this vulnerability were mishandled; others dispute this. The dispute turns crucially on how available exploits were during the embargo period. Nobody is arguing that the vulnerability doesn't need to be patched at all. (The PHP core team didn't initially treat it with much urgency, but they've acknowledged that this was a mistake.)
I think it's entirely possible that the released exploit and the early released video are independent of each other. The video contains a couple of hints and stack traces, so someone could've used that to pinpoint the problem and write an exploit of their own (or, based on the website it appeared on, had an LLM generate one).
On the other hand, Chainguard did release their patches before the agreed upon time, proving they can't be trusted if you're working on coordinated patch releases. That was an absolute mess.
> The dispute turns crucially on how available exploits were during the embargo period.
I sure hope thats not true.
All signs indicate that writing a self-replicating exploit that'd flood the internet without warning is possible. All we know is that nobody did.
A lot of websites - especially the ones who made the call to go down first like government / healthcare - do not have the luxury of taking any risk.
Writing this blog post with the "hindsight" of no exploit going viral for several hours after the patches is - let me be frank - complete bullshit; and i dont want this person anywhere near an important system.
If by next _year_ nobody has proven a successful consistent hack, then we can start to doubt the videos. If there is, then all of this has been warranted.
Time is the whole point though. Everyone was terrified and shutting their websites down because they believed the exploit would be running wild on the internet in between it hitting git and them being able to update the server.
If it takes a week to come up with a working exploit code, the critical services have time to update.
Time is part of the point, but it is not the whole point. The whole point involves the fact that this impacts two decades worth of PHP versions. Even if it took months to exploit, there is going to be a long, long tail of breaches against ill-maintained legacy sites.
Huh, I thought we were talking about this because it's a massive security breach offering approximately unconditional RCE against half of the servers on the open web.
The specific reason for the current degree of consternation, to the point where many websites went down preemptively which didn't happen even with Log4Shell, is that for about half a day there were multiple alleged proofs-of-concept circulating on Twitter while the PHP Foundation refused to release the patch. If that hadn't been the case, we'd still be having a conversation, but a different one.
Suppose hypothetically there was no disclosure period at all. Suppose that at the time the PHP team made the announcement that there was an unmitigable unconditional RCE against ~every version of PHP with advice to take servers offline, they had just been made aware of it themselves, and simply did not have a patch ready yet. The pre-emptive taking of servers offline until a patch was available, the OP blog post that was written, and this comment thread would all still apply the same. Nothing in this thread was in any way related to the embargo period. It is extremely weird that you came into a thread irrelevant to the embargo and told me my comment was irrelevant to the embargo, and then have stuck to your guns arguing this incessantly. The embargo and the severe vulnerability are related. They are not the same thing, and much of the conversation would in fact be exactly the same regardless of the embargo.
There's no need for an embargo before the patch is available, because the core team can just not tell anyone. It's only once there is a patch that the decision has to be made whether to embargo it.
You can never rule out the possibility that someone interested in attacking you has a non-public exploit, but it's dramatically less likely if there aren't already working exploits circulating all over the internet.
I don't think that's very true. Certainly, modern LLMs make it much easier to discover a vulnerability given only limited clues—but if it's easy to find that way, then probably someone will do that and post it for clout. So if no one has, then it probably isn't.
Time is not the whole point. How are people in this thread this clueless?
Risk is the whole point. The websites going down first were government / healthcare.
Do you expect these websites to take the risk? Risks that you now feel is justified because hindsight seemed to have collapsed the possibility-space?
We can talk about this again next year; if still nobody has demonstrated a working consistent attack that _could_ have pwnd my gov and medical records, then we can reflect on the severity score and the response.
That "if" is load bearing and increasingly flat-out false.
Crowdstrike's average "breakout times", the time gap between a first successful infiltration and maximal exploitation, has gone from days or hours prior to 2024 to minutes or seconds now.
ease is somewhat relevant initially, since it determines how quickly you have to patch before exploits start popping up. I would be extremely curious to see if anyone has been pwned by this already, especially before the public release of the patches (not that I doubt it, but I heard some people saying they saw attempts in their logs, but not really enough information to tell whether they were understanding their logs correctly. The situation just seems a bit murky and unclear to me).
> Ease is irrelevant because you only need expertise once and then everyone can reuse the result.
Not if your LLM relies on existing code and you collectively create a million repos with hallucinated broken implementations that do not work before a good public exploit exists, which later LLMs rely on. LLMs are self-defeating that way.
BioWare hasn't published a successful game in a very long time. Their last releases were Anthem (2019) and DA: Veilguard (2024), both of which tapped into the studio's reputation but underperformed in the long haul. Veilguard was also so "meh" against the backdrop of Baldur's Gate 3 that it probably harmed the brand and the franchise more than if it never came out.
They basically used to make amazing games in the 2000s and never quite lived up to that reputation again. Whether the acquisition by EA was the cause or the symptom of the decline, I don't know.
Dragon Age: Inquisition (2014) was their last critically acclaimed and successful game, and even that oen was considered a drop in quality for the series.
I'm playing ME: Andromeda now on a handheld and I don't get the hate. Sure it's not ME2 but graphically more polished though the storyline lacks the impact of the first trilogy. Otherwise, I think it's a fine game.
Andromeda got critized a lot for admittedly unnecessary flaws, but at its core it was a great mass effect. It even corrected many mistakes of the original trilogy, like having a proper ending that used the allies won during the game.
I think Andromeda was actually more of a victim of modern "outrage" culture. A few bugs, memes, and suddenly a game is "trash". The game wasn't terrible and I recall enjoying playing it at the time. Sure, some of the animations were stiff and the main character design wasn't stellar but those were _really_ blown out of proportion in games media at the time.
Nowadays if your game has even one thing wrong and the wrong influencer makes fun of it, suddenly it's trash. At the same time, your game can be trash and if an influencer laughs their ass off at the bugs (in a good way), everyone wants it. It's a sad culture. /rant
I still think Mass Effect Andromeda was one story DLC away from being my favorite Mass Effect game. The team behind Andromeda was the team that made all the best story DLC for the previous three games. Their wheelhouse was taking the game past the last stages of polish to telling the small stories that really tie the whole thing together. I still wonder what we might have gotten from the cancelled DLC if the technical bugs of the Frostbite engine switch hadn't created such a bad first impression.
(Side note, the ME: Andromeda novels are really good and hinted at a lot of storylines that we may never see closure on, but were certainly building blocks the game was working on getting to.)
Andromeda had a lot of good in it, tbf. Their big mistake there IMO was hard-resetting the universe the way they did. Like I was excited for more mass effect but I fell off that game so hard. Mechanically it was solid but I just… didn’t care.
Anthem frankly killed them. A live service looter shooter which were already out of vogue by the time it was released. Obviously made by marketing entirely, no soul whatsoever. They never recovered from it.
I will always defend Andromeda as not being as bad as everyone said. I enjoy going back and playing it, and they were a few better writers away from having a really good game. I cared more about the Irish redhead copilot than about any one of the actual companions.
Yeah, Andromeda has some cool ideas (the whole Remnant thing felt like a very classic Scifi mistery) but it was stuck in classing BioWare cliches with companions loyalty quests and romances, the main character being weird chosen one, the main plot was dumb and it was barely a Mass Effect because of the disconnect from the main universe and missing some of the favorite Mass Effect things (like the Geth).
It might have been successful if it weren't a Mass Effect. Or maybe not, who knows.
> missing some of the favorite Mass Effect things (like the Geth)
Some of that is interestingly handled in the Andromeda novels and there were hints that even some of the cancelled DLC was going to reintroduce some of those elements, but EA diverted the DLC teams to Anthem instead.
I think a large part of the issue is that the sum total of ME trilogy is a hard act to follow, without aiming to sound hyperbolic it's a very bright light that outshines a lot. Anything trying to be a sequel has a difficult balancing act of being familiar versus being different and bringing new ideas in.
I didn't say it was bad, I didn't think it was. The combat as I said was the most interesting iteration. I loved the gun play, I just didn't give a fuck about the characters. I found most of them either offputting or just meh.
I genuinely only remember Ryder and PeeBee by name.
> Mechanically it was solid but I just… didn’t care.
This is a great summary of the entire issue with the game.
I really tried to like it but it didn't inspire any enthusiasm. I didn't really care for the plot, the stakes seemed so low compared to the previous title and none of the races/planets really scratched that explore itch.
I liked the fresh reset because it was good excuse to get back to the "Exploration" roots of Mass Effect 1.
That said, the rumor is the in development Mass Effect 5 had found a story way to bridge the gap between the two and make it seem a lot less like a hard reset and find a way to continue past the ME3 endings.
Of course we likely will never get to hear the ME5 story the way it was scrapped along with the studio in this announcement.
They where by than "owing the don" as in there existence was company internally subsidized and you do not recover creative control from that. And those external management initiated "next shiny thing" chases kill you.
> It's not exactly uncommon, it's basically C's equivalent to exceptions.
longjmp is very uncommon, I don't recall seeing any high-quality C codebase using it, and if you want exceptions, you should pick a different language because longjmp doesn't really give you comparable semantics. It's basically saying "ok, computer, f--k my s--t up".
The fact that OOM conditions are an enduring source of exploitable vulnerabilities in PHP really should give everyone a pause. Like, that's not a thing in any sane project.
> And they botched this as someone talked. To me this looks like they used this for a PR stunt and aimed for an RCE.
This is a pretty constant theme in the infosec industry. The idea is that you want to do the "responsible" thing and let all the major, multi-billion-dollar companies patch first, so you form a secret club. But then, you share a very juicy, gossip-worthy thing with hundreds of people and ask them to sit on it for weeks. And guess what happens: they don't. Having worked for a big tech company, I've been involved on a couple of processes of this sort, and they all suffered leaks.
And rest assured, we'll learn nothing and try the same thing again next month or next year. Because what's the alternative? If you 0-day everyone, you get more hate mail, not less. It's all of sudden about you, not the vulnerable code. Microsoft or Cloudflare calls you irresponsible when asked for a PR statement, industry bloggers bully you, etc.
The elusive 100%-coverage formal spec that covers every intended and unintended input, the interests of every stakeholder, and every possible user behavior? In the vibecoding era? No, the page has the decency to acknowledge this doesn't exist.
In the end, they settle on finding crashes with a conventional fuzzer, fixing crashes, and then... writing a formal proof that the crash is fixed? That seems... pretty pointless? Am I missing something here?