matherial

Setting a swarm of agents loose for hours to look for software vulnerabilities is far more compute-expensive than fuzzing. The industry has never thrown this kind of compute resources at pure fuzzing, in part because you can't get much VC money for that.

> I don't think Zitron is wrong, exactly. He's just early.

By that metric, so was Nostradamus. The apocalypse is coming for sure, we're just quibbling about the timeline.

Realistically, timing is everything. You don't need to get it precisely right, but you also don't get a pass if you, for example, keep predicting an imminent recession through a decade of unprecedented growth.

If the parent knows electronics, that's already a pretty good predictor that the kid will learn it too. But in the absence of that, the kit itself doesn't teach much. The instructions don't really explain the circuits, they just show you how to put together colorful blocks.

I think it's the same story as with Snap Circuits for learning electronics, or baking soda + citric acid "chemistry sets". Many parents really wanted their kids to have a STEM career, and so we lowered the bar to the point where STEM-shaped educational activities are accessible to any kid. But because there's no bar to entry, playing with these toys doesn't reliably translate into motivation to really learn electronics, programming, or something like that. That motivation still manifests only in a small percentage of kids. Maybe we're making that percentage go up slightly, but it's a small effect.

FWIW, it's not exactly a new phenomenon. Two decades before Scratch, many schools were using Logo to teach kids programming, probably with similar success rates. Stuff like that is probably good even if it doesn't lead to a STEM "success story". It makes kids actively participate instead of scrolling Tiktok all day.

First, as others have said, a lot of translation work doesn't involve high-brow fiction. It's reference material of all sorts. Technical books, manuals, software localizations, etc.

Second, you vastly overestimate the desire that book publishers have to pay for good translations. I've seen books from reputable European publishers that were (badly) machine-translated by a contracted translator keen to dig their own grave. Yes, Harry Potter will get a good translation, but 1,000 lesser books won't.

I strongly disagree, for reasons that are different from what's discussed in the thread.

It doesn't matter if LLMs can't hold a candle to some of the finest human writing. What matters is that before you become a world-renowned writer, you need to pay your bills, often for a long time. And that's what LLMs take away. All the mundane literature-adjacent works: journalism, translations, technical writing / corporate comms, copyediting.

The same goes for many other forms of art. A budding painter or a musician could support themselves off commissioned / commercial work while working on their grand opus... but now, the customers just prompt gen AI.

Although Zuse definitely received too little credit in the West (for somewhat understandable political reasons), there's another Zuse museum in Hünfeld, along with exhibitions at the Deutsches Technikmuseum in Berlin. So this is not as profound as it might seem. I suspect the main problem with this museum was just that it's off the beaten path.

They aren't, they're one of the people who latch on longstanding and neutral syntax (em-dashes, "it's not $foo") as a proof of LLM text. I don't blame them because HN has a lot of people trying to pass gen AI stuff as their own, and you need quick heuristics... but I'd encourage people at least do it right. Pay for Pangram or something.

You're chasing ghosts, son.

Being right is not the most important part. If you're right but don't convince anyone, you've made no difference.

Theo was right that virtualization is a comparatively shoddy security boundary. At the same time, it's flexible and capable in ways that now define the shape of modern IT.

Could we have replicated that by other means? If yes, then it's on Theo and other knee-jerk critics that they never proposed a better approach and settled for insulting people. If not, then maybe virtualization was a necessary evil. Or maybe everyone else is an irredeemable idiot, but again - if we reach that conclusion, is the world better off?

Modern electret and MEMS microphones are almost flat from 100 Hz to 10 kHz. Here's an example plot:

https://www.allaboutcircuits.com/uploads/articles/CCTB_PUIME...

In contrast, small speakers, especially non-round speakers with plastic diaphragms that are commonly found in laptops, have pretty wild response curves, something like this:

https://www.bestartech.com/wp-content/uploads/2020/09/FRC-BL...

I am skeptical about measurements done using the laptop's microphone, though. These microphone arrays are often designed to be very directional, some laptops have bottom-firing speakers, etc.

For most cheap computer speakers and headphones, you can make them sound better for music and gaming simply by boosting low and high frequencies. And for speech, you often want the opposite...

For a typical program, I bet that the overall impact of the glibc allocator is well under 0.1%. If you can choose between < 0.1% and < 0.12%, I guess it matters in some sense, but not in any practical way. We almost never spend time on other sub-0.1% optimizations. You could probably squeeze a lot more by optimizing CPU branch predictor performance, minimizing CPU cache misses, or fine-tuning the scheduling strategy, but we also don't bother. 'Tis is the era of "native" apps written in Electron.

If your algorithm does a ton of small allocations to the point where the allocator is the bottleneck, you're already doing it wrong. The allocator necessarily comes with a lot of overhead because it needs to accommodate diverse use cases, avoid fragmentation, and ideally, implement a variety of security checks. If you're doing something alloc-intensive, you're probably allocating and freeing a lot of identical structures and you'd be better off grabbing some continuous memory and managing that yourself in a task-specific way.

But the reality is that almost no one actually cares about performance because compute is cheaper than expertise and labor, at least in the short haul. Everything is getting more bloated and slower and we just compensate by adding CPU cores, gigabytes and gigahertz.

No. Malaria epidemics occurred in medieval Europe, both in warmer countries (Italy, Greece) and in France, Germany and England. It was also common in portions of the US until large-scale eradication efforts.

It's unlikely to become a problem again because the climate is not optimal and these countries are wealthy, have advanced healthcare systems, and can spray everything with DDT again if they feel like doing it.

Show me a person who truly revised their priors to put forward a novel solution to AI that isn't just "do the thing that I like and that benefits me in some way". The discourse around all this is a mix of quasi-religious beliefs (pro and con) and blatant opportunism. For Dario and Sam, the solution is "give us a monopoly on this tech". For Gates, it's "create an supranational organization with unprecedented taxation and regulatory powers" - world AI police, presumably staffed by technocrats like him.

What's your take? Do the ideas put forward by Gates strike you as worth pursuing? What are the trade-offs - how can they go wrong?

I don't think it's the same. People usually don't see their "relationship" with celebrities as peer-to-peer. They either look up or look down on them (i.e., idolize or hate). So the attention you get from fans as a celebrity is actually a lot more meaningful. People go to great lengths to tell you about the impact you had on their life and they don't forget you easily. The scale makes it awkward, but the emotions are real.

On social media, we usually don't care about the other person at all. Likes and comments are 100% transactional. I have thousands of "followers" on X. If I die tomorrow, they won't even notice. The space on their timeline will be filled by other content to retweet or like.

This appears to be a 100% AI-generated essay, not the first from ACM and not the last. Your call if you want to spend your day reading and arguing with the article; the author presumably didn't.

I'd put it differently. The affirmation we were accustomed to generally had some substance. If someone called, sent you a postcard, or met with you, it probably meant they cared at least to some extent.

Social media strips all that away to the point where affirmation costs no effort to give but also means nothing. Here's a test I like to recommend: don't post anything on HN, X, or Facebook for a month. See how many people notice or reach out and ask if you're OK. The most likely answer is zero. Hundreds or thousands of "friends", zero meaningful return.

This also makes us feel that we constantly need to work for these affirmations. If you don't post, no one likes you. This is not a Facebook problem, but Facebook built its business around it, so they deserve to be singled out.

In the SF Bay Area, I'd bet good money that the vast majority of work like that is unpermitted. When I was adding an EV charging station to my home, I actually paid a (reputable, licensed) electrician and he rolled his eyes hard at me when I asked about permits. He said it would add weeks to the project and double the cost. Without it, it was a three-hour job.

The city isn't in the business of busting doors and issuing citations if you replace kitchen cabinets without a permit. They basically act if someone credibly complains, which is mostly if you're doing some exterior additions / changes that annoy a neighbor.

> 30 years ago you needed PG / YC to have a startup.

What's a "startup"? For the most part, you're just building a small business. You deal with local bureaucracies, learn about accounting, hire first employees, struggle to find customers. People have been doing this for ages without VCs. And 99% of startups never progress past that stage.

The only thing that VC funding changes is that it lets you grow faster than justified. This can turn you into the next Google, but most often, it puts you on a completely unsustainable path where you only have two outs: writing the project off or offloading it to a FOMO buyer for more than it's worth. The VC ecosystem is exceedingly good at that second part.

To be fair, I wouldn't pin the blame on YC. Few people go to Ivy League universities to ponder ethics or philosophy. These schools are widely seen - by students and by parents - as an incredibly expensive way to "10x" your earning potential and influence. They were producing Mark Zuckerbergs for a good while.

Yes, they accept some number of gifted people from underprivileged backgrounds, but these people are not the ones defining the school's culture. For most people, success at Harvard is to become a politician or to get very rich (or both).

> TUIs and GUIs are both mostly bad for the same reason, you have to learn a new “language” of interaction for with each program ... Stop building user interfaces in general. A command line interface is best.

A command-line interface is a "language" like any other. Each tool has its distinctive keywords, many have complex non-standard grammars (e.g., openssl, ffmpeg). And for more complex tools, the learning curve for this language is hilariously steep. I've been using ffmpeg on and off for two decades and I still feel quite helpless with it.

I think your argument makes sense, but it's just an argument for a standardized language, whether that's GUI, TUI, or CLI. Unfortunately, every other program or class of programs tries to innovate in some weird way. There's no reason why switching from Photoshop to GIMP should be so confusing, but it is.

Then ask an LLM. I'm serious, if you like the style. This article, despite the disclaimer, is almost certainly LLM-written. It's one of many ephemeral blogs that have been cropping up on HN lately and that post single-theme content with striking regularity.

The editor is not the one producing the text. For this book, the sample chapter seems to make heavy use of LLM output. Pangram agrees. I guess that for books about AI it's fair game? Eat your own dog food so to speak. People who don't like AI are not the target audience.

There's no reason they would preferentially travel through the lens and hit the current frame because most of the camera and film housing is about as transparent to gamma radiation as glass. Or maybe more transparent, because glass lenses in pro cameras tend to be rather bulky.

So this just happened somewhere between arrival to and departure from the zone.

The police have SWAT teams, guns, and jails, Flock does not. So I think the state security apparatus is a part of the problem, because if they don't like you and have access to surveillance technology, the outcomes can be far more severe than if Facebook doesn't like you, or the CEO of Flock doesn't like you.

Winona, MN has a population of 20k. I've never been there, but if it's like any of the small towns I lived in, most people trust the local police. If anything, the usual small-town complaint is that the police are not doing enough about <x>, where <x> are the homeless dudes in front of the library, or package theft, or whatever.

In small towns, people distrust the federal government and big tech a lot more.

I'm not criticizing, I just found it amusing.

It's funny how so many of the articles are hyper-specific grievances, e.g.:

https://consumerrights.wiki/w/Bose_QuietComfort_Sleepbuds_do...

https://consumerrights.wiki/w/Tyre_warranty_sold_via_mobile_...

And there's also https://consumerrights.wiki/w/Mr._Clinton_the_cat

Right. If anything, the 1990s were the decade of apex celebrity superstardom? You had singers like George Michael on the covers of mainstream magazines and in TV ads, you had all the boy bands (New Kids on the Block, Backstreet Boys, NSYNC)... the Rolling Stones got paid millions for the promo song for Windows 95, Britney Spears was selling Pepsi (https://www.youtube.com/watch?v=5fugLhNbwoY), Celine Dion had her line of cosmetics, etc.

The author was a teen back then, and teens to be countercultural, but I don't think they're drawing the right conclusions from that.