skippyfish

It's 100% AI-generated, including nonsense illustrations. And it's just so pointless too - there are many good, better-illustrated, and better-explained human articles about the same circuit.

I very much doubt that. We now have nearly-perfect AI image generation and it hasn't really changed the nature of human expression. I don't see my friends getting wildly creative. I mostly see it used for spammy blogs, spammy books, and cringeworthy corporate marketing - basically, a negative signal, rather than "ooh, AI image, I'm in for a treat". Is your experience different? If not, what changes with moving images?

Ultimately, most people don't have ideas for the kinds of personalized entertainment they want, and they don't want to be in charge of content production (even if you have an LLM do most of the work). I don't doubt that there are niches for it, especially stuff like porn, and I'm sure that pros (game studios, film studios) will leverage AI more and more, but I suspect that most of us will just want to sit on the couch, watch Spiderman XVIII, and then be able to talk about that shared Spiderman XVIII experience with all our friends.

I know that we're discouraged from meta-comments, but what is going on in this thread? It's a nearly 800-page book about the art of programming. A huge amount of work on a topic that should be dear to our hearts. News for hackers, right?

But somehow, the discussion has three themes. It's 50+ comments of "I don't like the first sentence of the marketing copy", "I don't like the tool the author is using", and "what would happen if we train an LLM on this book?". Has anyone read the sample chapter? Did you like it? Anyone here owns volume 1 and has opinions about that?

No one is writing purely for themselves. It's something we like to say, and there's some value in writing down your thoughts - but deep down inside, every blogger is dreaming of having a positive impact on other people's lives. That requires reaching these people in the first place.

I promise you that Bartosz would be bummed out if he had an audience of one. And while relying on HN to accidentally discover you can work, it's not a sustainable model for anything.

> I use https://kill-the-newsletter.com/ to put them in my rss though.

Every Substack has an RSS at /feed.rss.

Counterpoint from someone who's been publishing on the internet for decades: no one will visit your website. It's as simple as that. You need a push mechanism to reach readers.

RSS could be it, but it's too niche. Social media is kinda it, but algorithmic feeds reward political clickbait and just because someone clicks "follow" or "subscribe" is not a guarantee that they will ever see your posts. Substack gives you a way to push content to your subscribers without an intermediary deciding what they get to read.

So no, there's no special value to having a website or syndicating across the web. There's a lot of value in backing up your Substack subscribers' emails so that you can move somewhere else down the line, and the nice thing about Substack is that they let you do that. If they stop, that's a trap.

ACM has been leaning heavily into AI-generated content for their journals in the past year, and this article is no exception: it appears to be 100% AI generated and full of LLM verbiage.

There's something hilarious about that, but also, snake eating its own tail.

With respect, the way you talk about corrections makes me suspect you weren't working in tech in the US at the turn of the century.

A "correction" in this model isn't a reasoned and gradual re-evaluation of the market cap of every company. It's a broad pullback where people panic and no one wants to be left holding the bag. Money shifts into other assets for years and tech employment, incomes, and the availability of funding takes a big hit.

As to your comment about profitability... every unprofitable company is on a path to be profitable. Some even get there.

Plus, look at it this way: Kellogg's is a profitable company and a part of almost every person's life. Does this make them worth trillions of dollars? No, they just provide boring, commodity products, their valuation is basically a low multiple of the assets they hold and the revenues they bring. There's a future where OpenAI or Anthropic are more powerful than all the world's governments combined, but also a future where they're Kellog's.

> Ya, there are already three pure ad hominem comments here with no examples.

You should probably understand Thomas' comment in part as a reaction to Zitron attacking Thomas in a pretty low-brow fashion, essentially for saying "hey folks, AI coding works and you should be using it":

https://www.wheresyoured.at/how-to-argue-with-an-ai-booster/

The bottom line is that Ed Zitron, like Gary Marcus, built their entire brand on being AI contrarians; they can't say anything positive about it without qualifying it with a more damning negative. There's nothing wrong about having voices like that, but it makes them unreliable narrators. I am apprehensive about AI and its externalities, but I don't want to be caught citing either of them for that reason.

> I do think the valuations are quite and they would need to meet the reality, but don’t think there’s gonna be a crash or we’d ever go back to pre-AI era. It would more or less would be a correction to valuations.

Something to consider: would your description also apply to the dot-com boom of the late 1990s? The internet was real, the ideas for internet business were real, and we were not going to the previous reality. But the valuations weren't quite right and a "correction" happened at some point.

When people talk about AI crash, that's what they mean. Not that AI is a hoax, but that the correction could be quite violent and have effects on the broader economy.

That's like saying "politicians are crooked, so every bad thing you can think of about politicians must be true". It doesn't and we shouldn't fall for it.

In fact, the one industry where enshittification is widely embraced is our own, so I guess maybe it's projection?... Most other goods and services are not deliberately designed to get worse over time. We're the masters of giving people something cool for free and then progressively worsening their experience until they're forced to pay.

Things like appliances don't get worse on purpose. They sometimes get worse as a combination of market pressures and regulations. Take fridges: you can buy a full-sized kitchen fridge for less than $600, that's kinda crazy? And yes, it probably won't last a lifetime, and they won't make it serviceable because it's not worth servicing given today's labor costs - if your bonded-and-certified repairman quotes you $500, you're going to toss it out either way.

It's not just emotions, it's also a matter of time and uncertainty.

I was in a buying situation where my old lease was expiring soon, I had a second kid on the way, and the last thing I'd have wanted is to buy a place where I'd have contractors milling around for weeks or months to make various repairs. I was willing to pay a hefty premium for a place that's in a move-in condition.

I don't think this extends to "you'll get $10k less if you don't vacuum", but it's a common courtesy on a million-dollar transaction to maybe not leave a mess.

The page is entirely AI-generated and each piece is written with the same silly thesis (things used to be better and then they ruined everything). I see why this works, but it's frankly disappointing to see so many smart people fall for it over and over again. There's no human on the other end of these essays that believes these things. They're not true and not meant to be true. Yes, they're what we want to hear, but that shouldn't be enough.

As to why it exists: when generation of content costs next to nothing, I think that question is harder to pin. The basic answer is "clicks".

Indoor GPS reception is spotty, especially in urban / high-rise environments, so I don't think that's a realistically marketable product. Not unless you some have some window- or balcony-mounted probe, which would then require battery changes (although I suppose you could make it solar-powered).

I can't decide how I feel about this article. On one hand, it's great. It's great that the author wanted to get something done, and thanks to the marvels of modern tech, they could do it with very little effort.

But on the other hand... gosh, this was about the simplest software engineering challenge imaginable (toggle some I/O pins) and the simplest design task imaginable (make a featureless rectangular box). And the author - a technologist! - isn't just saying "I wanted to try out some new tech". They're saying "this was too hard, I needed help from an LLM" ("would absolutely not be able to meet this challenge", "hit another wall").

And that... I mean, I just don't know what to make of this. Were we always like this? Are LLMs making us like this? Is it good? Is it bad?... and don't give me the calculator analogy...

That's not relevant at all. I'm asking for human-written text that gets flagged by Pangram as AI-generated, not an artificially-constructed case that contains AI text (and was explained by the founder of the website as essentially a block resolution issue).

Discourse as we know it doesn't work if the cost of producing an infinite supply of arguments is essentially zero, while the cost of reading them and arguing with them remains high. Unless you want my LLM to get back to your LLM.

And this is a problem on HN today. There are powerful incentives to generate provocative opinion pieces just for clicks. I've seen websites on HN that seemingly took the human entirely out of the equation and just post a nearly identical op-ed every day on a fixed schedule. What's the point of engaging with that?

Large displays are not driven by a general-purpose CPU that bit-bangs a single serial data line, and never were. They're driven by separate circuitry with direct memory access, often integrated on the die of chips meant for these applications. The actual bus to the display controller may be a parallel dot-clock RGB bus, or an ultra-speed differential multi-lane serial (MIPI DSI, HDMI, etc).

And train stop displays certainly don't need 60 fps.

The main constraint for high-resolution displays is memory, not CPU clock speed. Your (odd) 1920x1920x24bpp frame buffer takes up more than 10 MB.

I'm asking the parent about their claim. I already have an opinion about the parent post, in which the author is claiming to have not used an LLM to author a 2025 essay that's illustrated with completely unnecessary AI art. Although one doesn't imply the other, let's say that this diminishes my willingness to believe that claim at face value.

Can you share your experiment, so that we can draw our own conclusions? I've been paying for Pangram for a long time and have not seen it misbehave even once. I've seen many people attack it, but when I looked more closely, their claims often evidently boiled down to "I want plausible deniability for my own use of LLMs" - i.e., they were clearly posting AI-generated content on social media and just didn't like the possibility of being called out.

I'm not saying that's you, but since a proof is easy to produce, it would be nice if you could share.

I'm not disagreeing with you, I think this goes beyond impractical, it's doomed from the get go.

In my book, impractical means "I built a cuckoo wristwatch". Beyond impractical: "I built a cuckoo wristwatch but there was no room for a working mechanism".

Well, that's sub-$0.50. But yeah, CH32V003 is in that ballpark, and some of the cheapest Microchip and Infineon products are around $0.20.

It's almost never worth it to buy the cheapest chip unless you're making a million of something, but there are very good ones around $1-$2, and $5 is the upscale stuff.

You can buy sub-$0.50 microcontrollers. But even at $5, I don't know why you'd want to run models on them, it's an environment constrained to the point of being useless for this task.

And I hope it stays that way, I don't want MCU shortages...

If you say "yes," you're also not doing science (anymore). Therein lies the problem with this take: most of science isn't done with the objective of getting to any specific finish line, and this is particularly true for math.

As the old saying goes, the purpose of a system is what it does, so there might be some Platonic ideal of science we can appeal to, but the practice of science by humans is about the craft as much as it is about the truth.

The fact this is AI-generated aside - I guess that's to be expected - it's a clever position to take if you completely ignore the realities of LLM-produced writing and imagine some fantasy world where AI blogs bring us profound wisdom and are unfairly targeted by pitchfork-wielding mobs.

Although there are some niche uses where LLMs assist or enhance the creative process, the vast majority of machine writing exists specifically because it takes zero effort and allows you to spam human cognition at an unprecedented scale. There is no redeeming quality to 99%+ of AI-generated LinkedIn posts, AI-generated books on Amazon, and so on.

The style-based heuristics we previously had at our disposal to filter zero-effort content no longer work here, so detecting LLM text is the next fallback.

There's very little that wasn't possible before LLMs, because, well, you still had humans. There are many things that the models promise to make a lot cheaper, if you're willing to accept trade-offs, but these trade-offs can be quite severe.

Many of the most successful applications of LLMs are fields that were already terrible. For example, LLMs are a natural fit for customer support. And somehow, it's also a natural fit for software engineering, which I suppose is an indictment of our field... who cares if a model comes up with a bad architecture or a product that only kinda-works, that's how we always rolled.

If you work for a tech company, I can practically guarantee that it bans a lot of lawful behaviors on its platforms. Paypal bans or restricts online dating, cigarettes, and firearm accessories. Youtube demonetizes swearing. Facebook prohibits the sale of bodily fluids (I don't want to know what triggered that policy) and "used cosmetics".

If you're not interested in stuff like that when your company is doing it, but are up in arms about some other platform deciding that they don't like crypto, I think you should look in the mirror. Their reasoning is probably the same: they don't like it and don't want to put up with the task of sensibly policing the category.

> raining a SOTA model takes a huge amount of resources and expertise

Writing books, building Wikipedia, and answering questions on online forums takes a lot of resources and expertise that scraping didn't. So at the very least, we're already one rung down the "maybe you should've asked" ladder.

> Civil breach of contract. Anthropic's TOS explicitly say you can't do what Kimi is alleged to have done.

Ah yes, I remember when Anthropic crawlers abided by the TOS of the websites they slurped up.

All your other points are downstream from this, which makes them pretty tenuous. Labs don't think that ToS or other explicit wishes of content providers apply to them, but they expect everyone else to abide by theirs.

> I think the explosion in the popularity of Lean probably means that tactic-based proofs have won

The number of mentions of Lean in HN submissions aside, how do we gauge that? HN has odd trends like that - a decade ago, we loved everything "Bayesian" - but they don't necessarily translate to anything that's happening in the mainstream.