HN Best Comments

Re: Citroen revives the 2CV for an electric age With all the hype around the nostalgia it seems to be slipping by people that this 2CV also has a class leading weight at under a tonne. Lighter weight means better range for a smaller battery, more efficient tyres and lower road wear. It's one of the reasons why the target price is going to be so low. It's a really cheap car but I think to do that they still have to have some good engineering in it. Lio, 21 hours ago

Re: Denmark set to allow return of 'physical force' in school > Starting in August 2027, they will be allowed to 'hold a student, keep them in place, or take them to another room' if the student disrupts the class. Far from what I imagined as 'physical force'. feverzsj, 4 hours ago

Re: AI coding agents generate more code, but not more software I would reframe this as: Quality was always a significant bottleneck and is even more so with agentic coding. Code review is where many of the design & code quality problems are observed/highlighted. cyb_, 13 hours ago

Re: Hackers activate automatic defrosting in supermarket fridges across Europe Was just talking to a lab tech at a hospital. It seems that some of the newer lab machines require internet access because you can't run lab tests unless you have a paid subscription to use the machine. No credits on the subscription, no tests. I mention this b/c a lot of commenters here are saying "but why would you ever hook up a freezer to the internet??" I find that type of comment fascinating as many of the people on HN worked for firms that built the technology required to use internet based subscription models. alexpotato, 9 hours ago

Re: Passkeys are a trap I've yet to voluntarily enable passkeys, for the main reason that I want portability, and not to be tied to any single device that can be lost. I want to be able to backup and lock my credentials in a safe. I also want to control the strength of my credentials for different services, and I don't consider my thumbprint used to unlock my phone to be as strong as a long password in a password manager. The first time I was forced to use a passkey implemented within the vendor's app (3rd party password managers were not an option). I quickly closed that account. The second time was an app that popped up a passkey opt-in in the middle of a bunch of transaction screens. I quickly realized the error, but there was no easy undo. The opt in was one green button in the middle of the screen. Turning it back off required digging through settings to find the security option to disable, and then confirm my choice multiple times. That process also signed out all my other devices. The fact that companies are resorting to dark patterns and forced requirements, where my money is held hostage, should be pretty good evidence of how poorly the passkey rollout is going. bmitch3020, 1 day ago

Re: Hackers activate automatic defrosting in supermarket fridges across Europe If you believe some of the emerging comments out there - try "Danfoss CVE" - you'll see that some of the suspects are 15-20 year old controllers like the Danfoss SM800 (EOS 2021) and SC355 (EOS 2016). Why don't they just--- no margin. But there were CVEs-- no margin. The laptop class doesn't always grok that the shift-work labor class has nothing in terms of luxuries like security reviews and gear replaced on cycles and other basics taken for granted. But I'd setup Tailscale and-- would you, the grocery store manager running on 3% margin, really? What if you're part of a chain and it's centrally controlled elsewhere, like all the other decade-old tech you use but have to call and wait for a dispatched tech to handle? Secure-by-default laws will eventually help, but there's always going to be groaning infrastructure particularly in commodity industries, and we still haven't onboarded most of the disaffected men to the Internet that will commit innumerable more hacks and disruptions. 9x39, 6 hours ago

Re: AI is a tool and using it well takes skill Its probable that AI will roll out much faster than humans can develop that skill of using it in a way that increases not decreases human knowledge over time. The current slew of excellent LLMs will probably cause a decades long dip in human cognitive ability. If social media on smartphones had an impact of ability of students to concentrate, LLM AI will likely have a much greater negative impact on education. There are trillions of dollars of investment pushing for humans to trust whatever an LLM says, and very little money behind encouraging people to tackle hard problems themselves and train themselves to think. I have to say honestly, that AI looks like an apocalypse for human thinking [ Im glad Tao is warning people, I just dont think many people will listen ] jgord, 3 hours ago

Re: Muse Wonder I personally have many problems with Meta, but imo the more companies close to the frontier, the better. If only one company were calling all the shots once they hit recursive self-improvement, it would lead to extreme concentration of power, and their values being enforced across the entire industry. With 3-4 companies at the frontier, there is more urge to cooperate to avoid a catastrophic race condition. atleastoptimal, 7 hours ago

Re: Passkeys are a trap Note that attestation can only be required during credential registration; at login time it's not possible to retroactively require attestation if it wasn't obtained during registration. And browsers will throw up a permission prompt if a website requests attestation. Are Netflix and Twitter currently requesting attestation when people register passkeys, and rejecting the registration if the user declines the permission prompt? I very much doubt it, since Apple's passkey implementation doesn't support attestation. Therefore, the author's scenario that sites will wait until most of their users are using attested passkeys and then "flip the switch and start rejecting non-attested passkeys" isn't very plausible. Attestation is certainly a stain on the WebAuthn spec, and it would have been better if Chrome and Firefox didn't support attestation (unless perhaps the browser is enrolled in enterprise management), but this blog post is presenting a hypothetical and unlikely scenario. That's not a good enough reason to recommend TOTP instead, given that TOTP is vulnerable to phishing and passkeys aren't. If consumer-facing sites start requesting attestation when registering passkeys, they deserve to be vigorously called out, and maybe then the recommendation to use passkeys should be revisited. But until that happens, don't tell people not to use passkeys. agwa, 24 hours ago

Re: Research Firm Calls Anthropic's IPO 'Unprecedented Test of Investor Gullibity' New Constructs is bearish on every tech IPO. Easy to reinforce your bias by selectively looking at instances where they were right, like WeWork, but they also predicted failed IPOs for Spotify (+219%), Uber (+60%), CrowdStrike (+3136%), Cloudflare (+2306%), Palantir (+1991%), Snowflake (+51%), DoorDash (+93%), Robinhood (+187%), ARM (+422%), Instacart (+59%), Reddit (+370%), CoreWeave (+105%), SpaceX (+20%) and lots more. So whatever your stance on Anthropic this particular analysis is safe to ignore. binlog, 12 hours ago

Re: Thinking will become a hobby I imagine this is like how microwaves put restaurants out of business, how TTS obviated literacy, how the automobile age made it unnecessary to walk, how Zoom caused every office to close, how helicopters put mountaineering firmly behind us, Amazon obsoleted brick and mortar stores, and payment cards made cash disappear. It's not only ahistorical to fail to consider preferences and politics, but it's also an indictment of the author's ability to judge the quality of a machine output versus that of a competent person. No doubt that like any of the examples above, some displacement of resources and effort will happen. But that doesn't indicate that something as central as "thinking" (lol) will fade into the past. lsy, 3 hours ago

Re: Deloitte falsified documents to evade tax on Abramovich's wealth > Deloitte Cyprus said it could not comment on the specific allegations because the documents were more than 20 years old and no longer accessible. That's convenient. Also, I received a similar response when inquiring about a financial account that had been overlooked (there are reasons, I'm not getting into it) for a couple of decades. "Can't help you. Too old for us to have records." Well, my records are printed documents. Why can't you tell me anything? Reminds me of when the USA government stole my money from a bank transfer claiming suspicion of whatever (my landlord had a Middle Eastern sounding name and it was the 2000s) and then said they had no record of it when I provided all the required documentation to recover it. In both cases, unfortunately, I concluded that the odds are so stacked against me that I gave up. Convenient system if you're fabulously wealthy or a massive org (such as corporations and government). alsetmusic, 7 hours ago

Re: Fingerprints are the new cookies > the consenst banner is required either way. Consent is required. The banner is just one mechanism, but not required. A website could hold off on all tracking until a suitable point in the usage of the site (like creating a profile, for example) and ask at that point. The banner is an artefact of the hostile way most sites choose to try and hoodwink the user into being monetised and ultimately to piss users off so much that they complain about the legislation rather than the lack of online privacy. louthy, 6 hours ago

Re: The gap in shared understanding of LLM capability is widening I feel like if the gap were so monumental that he could just point to a monument of it, rather than hand-waving at "the vertical professionals". Alternative interpretation of the numbers; there's only a subset of productive work that LLM's are 'capable' of doing at this point. That is, work that's verifiable by a machine or doesn't require verification. Thus, the cleavage points are along that axis and-- naturally-- most human beings don't have that kind of work. usernomdeguerre, 16 hours ago

Re: Fingerprints are the new cookies > Remember the big fuss about third-party cookies? It was so big that the EU had already rolled out one of its greatest inventions - consent banners. It was never about cookies. Thats what the industry made you think. You don't need to show consent banners for cookies (like auth sessions). You need consent for tracking. This includes profiling. And it doesn't matter what you use, fingerprinting or tracking cookies, the consenst banner is required either way. littlecranky67, 5 hours ago

Re: Terence Tao: Math 2.0 [pdf] > Before injecting this cocktail into your bloodstream, would you want to know that there is at least one human cancer researcher who understands the mechanism behind this cure. Tao is out of his lane; lots of medicines don't have understood mechanisms. We don't even have the mechanisms behind general anesthesia nailed down; do you want to forgo it when the docs cut you open to remove your tumors? squidmaster, 22 hours ago

Re: Passkeys are a trap > at login time it's not possible to retroactively require attestation if it wasn't obtained during registration. Of course it is, the site will reject the passkey and require you to create one that is attested. > Are Netflix and Twitter currently requesting attestation when people register passkeys, and rejecting the registration if the user declines the permission prompt? They could start tomorrow. > a hypothetical and unlikely scenario. You have not been paying attention, that scenario is where this whole thing is building towards. > If consumer-facing sites start requesting attestation when registering passkeys, they deserve to be vigorously called out It's going to be the banks first. Why called out, it's all for your safety, to make sure you are not using an insecure client that might leak your data. throwawayffffas, 10 hours ago

Re: The gap in shared understanding of LLM capability is widening https://xxcancel.com/karpathy/status/2109361546505966046 frereubu, 9 hours ago

Re: I paid people to try and follow my README >I hadn't actually explained what the software would do. Half the posts I see lately are like, "Gleam 2.0. What we learned" and then you go to the homepage and it's "Gleam is a Tribble for your Fork! (Scroll down) See if you qualify for Gleam Enterprise!" andai, 17 hours ago

Re: The gap in shared understanding of LLM capability is widening Don't think the 0.2% really are in agreement that "large, complex projects that used to take them weeks/months can now be completed by agents with a prompt". The divide seems to be between people who think you may need to understand the codebase for maintenance, potential future token costs, optimisation, code erosion & being wary of the assumptions/mistakes that they have seen LLMs make, and people who don't review code and are confident that the AI has made no mistakes and will be available for next-to-nothing forever. It's certainly allowing the latter to move fast... but will their clients be happy if/when they break things. We'll find out soon, I'm sure. rico735, 7 hours ago