deaux

This comment section is the best example of all time of the arrogance of Big Tech and its employees. Please try to take a second thinking outside of your bubble before commenting ridiculous stuff.

Yes, as a wealthy American, you "live in the Apple ecosystem". 99% of the world doesn't. And guess what, they're affected by passkeys all the same. They use a Windows laptop, and either an Android phone or iPhone. A lot of people even have an Android phone and an iPad. And no laptop at all. But at work or school they have to use Windows.

It's quite simple. Besides people "living in a single ecosystem" (discussed above, this is almost nobody), passkeys are only viable (i.e. not very painful to use) if you use a dedicated cross-platform password manager. Yet people who use those - which too is a globally negligible percentage - are exactly the people who tend to have near nothing to gain from passkeys, and only to lose. The majority of them is tech-savvy and they use auto-generated unique passwords. In that scenario, the minuscule improvement in security is meaningless and not worth it.

Ironically, this comment section shows exactly why passkeys are a shit show. Half the people here are exactly those who are coming up with this shit in their FAANG jobs, happily part of the global 1% (of which their tech-illiterate grandma too is part of), and they have no idea or care in the world for the remaining 99%. Unless of course this was simply a land grab for lock-in, which is about as likely.

"Intelligence" being what, math? Coding? Unfortunately there's a billion use cases for LLMs whose performance is not at all captured by the popular benchmarks they're all trying to maxx.

It's literally the meme of the MS org chart pointing guns at each other.

The GCP team wants their slice, the other team wants some otjer slice, and so on. Everyone wants some crap for their promotion package.

It's no wonder Meta has shit the bed even worse.

It's also why Google still releases actually decent, useful models despite the product being such a hilarious mess. A lot of the time Gemini models have actually been better as production LLMs as part of LLM-based production applications than OpenAI and Anthropic models when it comes to the complete cost:quality:latency:adherence picture. And they still are. We have products in prod that use Gemini because they're better than any other model at the specific task. But we wouldn't dare use it for anything coding related, or even just as productivity tool to rely on, because as a consumer product it's a joke.

This is a point that bears repeating: because U.S. open weight model makers must follow the frontier labs’ terms of service, they (1) are worse than Chinese alternatives and (2) end up distilling the distillation, just with a detour through Chinese labs. Wouldn’t it be better if western open weight model makers could go to the source?

This is of course a baseless assumption. Let's say China created GPT 3.5. Then I can guarantee you that Ben would say "Western frontier labs are at a disadvantage when gathering data, because they have to follow the terms of service of Western media, and Western copyright law". Which we now know wasn't true.

And sure, some will say "but Anthropic can more easily block this as it's a single point of failure". But it's doable to overcome this. Without being "state backed".

No, it is not. This is a common misconception that the large companies are very happy about.

1. The century-old Ford decision wasn't about this. It was about him refusing to pay dividends to shareholders he was feuding with. I.e. dominant shareholder using his control to starve minority shareholders of returns.

The court still let him keep spending huge sums on factories and price cuts that didn't clearly maximize profits.

2. Even if the case was about profit maximization (it wasn't), the judgment was a Michigan state decision. It doesn't have force outside of it.

3. US business law is in practice actually the opposite. A business can do pretty much whatever it wants. This is fundamental.

4. Even if it were illegal (it's not), anything can reasonably be framed as being in the long-term interest of the company, including donating to causes, raising employee wages and so on. Courts never question this.

Just imagine this was a thing. It's completely untenable for this reason. Who is a court to judge that something isn't in the longterm interest of shareholders unless it's literally spending all the company money on yachts for personal use?

5. No company has ever been prosecuted for this, obviously, because it's not a thing that exists. The closest you can get is that there's a potential duty to seek the best price if a company must be sold or broken up. But that's a very specific situation.

It's 100% a myth. Feel free to copy this and spread it when you see someone saying this.

Of course not, this makes no sense. It's the opposite, most of LLM generated apps are indeed uninteresting, yet it's not inherent at all. Most photos, most drawings, most human-written articles are uninteresting and have been so since forever. Nothing about that is inherent. And in the exact same manner, no, others can't generate the same thing. They physically can, but that has never been a barrier.

If I make uninteresting art, it's not because I physically lack the ability to use my hands in a manner that would create interesting art. It's because I can't come up with the ideas, and lack the taste, to produce interesting art.

It is. There are many people on here who think of having any LLM generated, never-human-reviewed code in the codebase, as vibecoding. There are also people like you, who believe it's about completely ignoring it, and even human-reviewing 10% of the code means it's not vibecoding. The median definition, at least on here, is somewhere in the middle, around "a large portion of the code is never looked at by a human".

Considering that it's thanks to China that even a 1% chance of avoiding this remains, I'd say so. Without their enormous investments in solar and batteries, it would be 0% rather than 1%.

I don’t care about your Show HN unless you have spent a month on it.

100%, you actually stated it correctly. It's not "I don't care unless you've written/reviewed everything by hand". If someone writes something by hand in an hour, including ideation, and puts it on Show HN, realistically it's almost certainly not going to be worth even looking at.

I've been running an app in prod for more than a year now, mostly vibecoded long before HN believed it was possible - HN only started believing this around Jan this year at the earliest. But nobody who uses it cares, or even notices, because it took a month to build. That's what matters, the amount of time and thought put into it. Of course it's true that vibecoded projects have a much lower median amount of time and thought put into it. But that's like Louisiana having a much lower median income than California, it's just a median, not an inherent characteristic of being a Louisianan vs a Californian, there are plenty of poorer Californians and richer Louisianans.

Yes and no.

Let's first settle on the definition of vibecoding so that we're not miscommunicating over definitions. I'm using the one that seems the median definition nowadays: >95% of the code written by LLMs, <60% of the output code human-reviewed, meaning there's a large part of the codebase that no human ever reviews.

As you said it's about time invested in thinking about it, yes. But remember that even pre-AI >90% of software got never used, it was dead on arrival. Look up "success rate of software projects in business".

You can put lots of time into thinking and vibecode everything. You can put very little time into thinking and write by hand. Of course, vibecoding makes the former much more likely. But nothing about it is inherent to it.

This isn't true at all in general online discourse. Maybe on X, in which case I'd recommend getting off of X.

It's overwhelmingly brought up when talking about Japan (and sometimes Korea) in comparison to the US (or EU). With Japan (or Korea) being the high-trust culture in that comparison, and the US/EU being the low-trust one.

I guarantee you can do a search across mentions of high/low-trust culture across online platforms in the last 12 months and the large majority will be these contexts, i.e. Western countries described low-trust, not high-trust.

Exactly the same for me. Sonnet 3.5+Aider made it possible.

I got downvoted though, apparently you and me are liars, haha. Easier to believe that than admit that their take of "I adopted it exactly when it became viable, now it's good, and before that it was a waste of time" was wrong. They're the people who were loudly claiming here last summer it was useless, asking "show me anything useful that has been coded using LLMs". Come to think of it, it has now been a few months since I saw one of those. Used to be every other thread.

Exact same for me as WhatIsDukka.

at this time last year I couldn't even get Claude (using Cursor) to spin me up a service skeleton that would compile, let alone do anything meaningful

I've been using it to do this for 2 years now. And many people with me. The change you mention is one of is primarily one of Overton windows, of vibes.

Do you ask this to all HNers who have worked at Meta, Google, Microsoft and Amazon - the latter three who Palantir relies on to even exist?

I.e. half of HN?

the Dems simultaneously find their balls again and stop being allergic to the use of power.

If only the problem was that they don't have balls. Everything they've done, and not done, is intentional. Hence why most of them see the DSA as a bigger threat than the Reps, and the former is what's awakening some of them rather than 8 years of MAGA.

Wonder if the whole cyber paranoia leads to their models ultimately generating less secure code.

This may be the goal.

Looking forward to the Apple quarterly earnings call where Tim Cook (or soon the new guy) explains why they decided to let go of the EU, leading to a 30% fall in revenue and net profit. I'm sure investors will be delighted.

Not really. EU is actually trying to decouple. But in many cases there are not any homegrown alternatives to support.

If the EU was trying to decouple they'd mandate at least including a hardware token option as an alternative. This is not new technology, it's existing and has been in use for decades.

They're not trying to decouple, so they haven't mandated it.

Clearly it isn’t. This is what techies forget: The mass amount of Europeans don’t give 2 shits about digital sovereignty or open source.

When Trump was invading Denmark, a huge % of Danes would've given a shit about sovereignty from the US. And that's the moment to pounce.

What he's correctly saying is that if even one major country adopts an EUID Wallet implementation that only allows for Google and Apple, this on its own has a magnitudes bigger impact than the pockets you're talking about. That's a regression.

The EU legally forbids member states from making a smartphone mandatory to access public services.

Yes, I'm sure they'll still allow for mail-in of obscure forms to access public services, which will then take 3 weeks to be processed.

If the EU actually wanted to "anticipate" this danger they'd have made it mandatory to include a physical form factor in EUDI wallets. In reality, they don't mind this danger, so it's optional, and you can bet most countries won't include one and make Google and Apple the only options.

It's even more expensive to have your country's digital ID held hostage by the US or its big tech players.

This is Daily Mail (or Fox News for fellow Americans) level tripe, and you can be almost sure of this just based on the title. Designed for dopamine maximization.

I think most liberals have the intuition that laws should apply equally to citizens and non-citizens, and I think that's where a lot of the discomfort comes from when we talk about immigration. A citizen who doesn't meet those demands imposed on non-citizens (e.g., language, cultural assimilation, etc.) will never be at risk of deportation, simply because they were lucky enough to be born in the country.

What you said makes no sense. Vanishingly few liberals believe laws should apply equally to citizens and non-citizens. If they did, they'd be for the complete abolition of visa/residency conditions altogether, simply allowing anyone to immigrate to the country. I doubt that even in the US more than 5% of people believe this. Any European country it'll be <3%.

Any visa/residency condition is a demand not imposed on citizens. Language and cultural assimilation aren't any different from other such conditions.

I disagree. They should learn the language of their own accord or get kicked out.

I can't agree, even though we likely have the same goals. You want to filter for people who work very hard to do their best to integrate. You don't want to filter for being good at self-studying languages. Most people just suck at self-studying through not much fault of their own, they were raised that way. It can also be incredibly useful for non-language integration purposes, as it gives them hours of direct contact hours with natives which - again, for the dedicated ones, which are the ones we want to stay - often leads to social contact and further integration.

I get where you're coming from, but the tax dollars needed for such language schools are really negligible, and I've seen with my own eyes the benefits to integration that couldn't be had without it. It's just a good cheap investment. I'd still apply the same final bar as you would, FWIW.

I feel like the default assumption for a company in the privacy space is that they are close to politically neutral (0).

That assumption here is wrong, because VPN companies in particular are known to be a lot dodgier than average. Maybe if we were talking bakeries, then I'd agree.

Also also, saying that Proton’s founder is “outwardly MAGA” is really oversimplifying things. I’m not sure what your viewpoint actually is — maybe you’re just expressing someone else’s — but I’d encourage you to read: https://medium.com/@ovenplayer/does-proton-really-support-tr...

One can make very similar arguments for this specific case. Saying that the party in question is far-right is simplifying things. On top of that, a co-founder making a private donation to a party (two layers: 1. private 2. party) is less directly related to the company than a founder directly tweeting an ideological point (1. public 2. direct).

Honestly, the party sounds insane, I wouldn't be caught dead supporting them and would hope that any Swede reading alone avoids them with a ten-foot pole. But the response in this comment section is absurdly exaggerated compared to what any of the Big Tech corps that get talked about daily on here, and whose

I don't believe it's organic at all. Recently there's the biggest push in history of "think of the children" online surveillance, with remote attestation and mandatory identity coupling being rolled out at scale. Mullvad is known to be one of the top privacy-preserving VPN providers in the West, with a large userbase on HN, and now you get this very disproportionately negative reaction to what is realistically a meaningless issue: "$0.01 of every $1 spent on Mullvad might end up with someone who donates to a radical party, if even that" - just like, almost certainly, every other $1 you spend elsewhere does.

https://miro.medium.com/v2/resize:fit:1056/1*9_3JV1BBIZjAYko...

Founder and CEO.

I too don't believe that GP's dream is going to happen, but the current momentum is much stronger than in the OWS days. Maybe it's less visible to some because the world has moved into online algorithmic bubbles rather than people camping in front of Wall Street. If they were more scared back then than they are now, it's not because the current momentum is actually weaker.

Can you share the reasons that you believe this?

Firstly, please keep in mind I'm talking about the entire doctor population of the world here. Not sure which particularly bubble of this earth you have experience with, but note how half the word's population lives in India/China/Indonesia/Pakistan/Nigeria/Brazil/Bangladesh/Russia. Now I do believe that it holds the same for e.g. Europe and non-China East-Asia, but still.

How many patients has the world-wide average doctor seen? How long have they been a doctor?

How many have they seen with the particular condition the patient has?

How much time do they spend listening to and reasoning about a patient? The median in the world is likely under 3 minutes.

How many real-world incentives do human doctors have to deal with?

Given infinite time and resources, and zero external incentives, maybe the median human doctor would outperform the LLM at this task. But this is completely detached from the real world.

What is special about medical imaging that makes AI/LLMs specifically bad?

LLMs: Besides lack of training data as mentioned elsewhere, they're simply not trained for high-fidelity image processing in general. It's not limited to medical imaging. It's a bit like the "How many Rs in strawberry" thing, but worse.

As for "AI" in general, medical image analysis is a very active field. These tend to be purpose-built though, not general-purpose. It seems likely at some point they'll become mainstream, but there's still a way to go.