Top 28 stories from Hacker News. Top 10 include comment highlights. Compiled at 20:10 UTC.
543 points by alternator · 397 comments
What HN said:
Jweb_Guru: This jives with what I've experienced in the brief time I had access to 5.5 Pro. It's the very first LLM that I feel like I can wrangle into solving tedious, but straightforward, problems correctly.
pmontra: It's a very long post with a mix of technical (math) and philosophical sections. Here are the most striking points to reflect upon IMHO. > It seems to me that training beginning PhD students to do research [...
robot-wrangler: A very interesting comment from Baez, I'll just quote part of it. > Where does the value of thinking and having deep ideas come from? We need to think about this now. If it comes primarily from their scarcity – the fact that having certain ideas is hard – then indeed this value m...
ziotom78: I am a physics professor and often use Gemini to check my papers. It is a formidable tool: it was able to find a clerical error (a missing imaginary unit in a complex mathematical expression) I was not able to find for days, and it often underlines connections between concepts an...
443 points by atgctg · 137 comments
Media over QUIC: There are ways to do voice AI without being traumatized by WebRTC.
What HN said:
Sean-Der: Responding to some technical points first, but then after that I do see a future that isn't WebRTC. I don't think it matches where WebTransport+WebCodecs etc is going though.
thutch76: I didn't make it all the way through the post, but I have to say I think he fundamentally understands the purpose of WebRTC. He calls himself an expert, and yeah he's written SFU's in go and rust and different companies ... but his technical credentials do not mean he's correct.
jedberg: I have a lot of experience in this area (and some patent applications). For Alexa, the device established a connection back to the server and then kept that open, sending basically HTTP2/SPDY/Something like it over the wire after it detected the wake word.
fps-hero: > But nope, WebRTC has no buffering and renders based on arrival time. Like seriously, timestamps are just suggestions. It’s even more annoying when video enters the picture. I felt that comment my bones.
420 points by hggh · 64 comments
What HN said:
miki123211: IA needs to do what Usenet has done. Have a bunch of mission-aligned but unrelated orgs (under different ownership and distributed around the world) that peer with each other, distribute all the content obtained by any of the orgs to each other, but that have no technical channel...
input_sh: Relevant blog post: https://blog.archive.org/2026/05/06/internet-archive-switzer... > Internet Archive Switzerland joins a growing group of mission-aligned organizations, alongside Internet Archive, Internet Archive Canada, and Internet Archive Europe.
insom: That website is really struggling. Very tempting to go to a mirror on archive.org to view it :) This seems very distinct from Internet Archive in the US, I wonder how separate it is.
teew: The About Us section states: > We are a team of change-makers who believe that every helping hand can raise a child and create a better future for them. Which I found weird. And searching for this phrase yields many site-hits verbatim, which is even weirder.
371 points by pretext · 226 comments
What HN said:
tmhrtly: My concern here is that by gravitating to HTML you lose the ability for a human (you!) to easily co-author the document with the LLM. If it’s just an explainer for your consumption, that’s not a concern - but if it’s a spec sheet for something more complex, I deeply value being a...
momojo: When exploring a new idea or tool, my go to prompt is
In a single index.html, no dependencies, sparse styling, create an app that <idea>Even before AI, it's how I built small tools, and there's something lovely about being able to email my friends the tool, and tell them...
arianvanp: The irony of this being a Twitter post with pictures of html rendering instead of an interactive html page is not lost on me. Arguing for html on a platform with less rich semantics than markdown is just ultimately funny
SylonZero: This viewpoint talks about consuming markdown natively without reference to how non-technical users view it: as rendered HTML. At best, this makes a case for introducing a few more useful constructs into markdown (without ruining it's simplicity of course).
347 points by homebrewer · 92 comments
What HN said:
niobe: An impressive attempt to summarise Wi-Fi which is a very deep topic. However I think the executive summary already missed the most critical thing about Wi-Fi: only 1 transmitter at a time per channel - across all WLANs, yours and your neighbours, with no deterministic way to avoi...
KingMachiavelli: I'd like to understand why the WiFi spec developed so slowly from G to N and finally to AC but now it's seems like a new version is released every other year yet many of the features/extensions are poorly implemented or have nearly 0 real world improvement.
anyfoo: > Wi-Fi signal strength decreases at an exponential rate as you move further away from a router. This is surprising to me. I'd have guessed it decreases quadratically (i.e. due to the inverse square law), not exponentially.
AdityaAnuragi: With Wi-Fi 6's OFDMA handling the hidden node problem better, does it actually make a noticeable difference in dense apartment buildings? Curious if anyone has measured this in practice
308 points by ingve · 179 comments
What HN said:
gobdovan: >He argues that conceptual integrity comes from both simplicity and straightforwardness If you want to follow this advice, I think Leslie Lamport's formulation is the most concrete. His version is: - Write down a small model of what you are about to build before building it.
nvader: Fortunate to be reminded of this right now, especially the pull-quote about conceptual integrity. This is the reason why AI-assisted programming has not turned out to be the silver bullet we have been hoping for, at least yet.
nemo1618: It's interesting to revisit Brooks' "surgical team" in light of AI. For example, I frequently have Claude act as a "toolsmith", creating bespoke project-specific tools on the fly, which are then documented in Skills that Claude can use going forward.
ilitirit: "Conceptual integrity" is something I always try to adhere to when I design and build systems. I also use it also a bit of an indicator when gauging someone's experience.
276 points by rbanffy · 105 comments
Large Language Models (LLMs) are poised to disrupt knowledge work, with the emergence of delegated work as a new interaction paradigm (e.g., vibe coding). Delegation requires trust - the expectation that the LLM will faithfully execute the task without introducing errors into doc...
What HN said:
simonw: I'm suspicious of their results with regards to tool usage. It's unsurprising that round-tripping long content through an LLM results in corruption. Frequent LLM users already know not to do that. They claim that tool use didn't help, which surprised me...
causal: Yeah I've been saying this for a while: AI-washing any text will degrade it, compounding with each pass. "Semantic ablation" is my favorite term for it: https://www.theregister.com/software/2026/02/16/semantic-abl...
timacles: Least shocking thing I've read about LLMs recently. They are essentially like that one JPEG meme, where each pass of saving as JPEG slightly degrades the quality until by the end its unrecognizable. Except with LLMs, the starting point is intent.
buffaloPizzaBoy: I typically tell my agents to only treat document writing as a last "rendering" pass. LLMs are so good at taking sparse knowledge and compiling it, that I prefer to store knowledge as composable ideas/facts.
189 points by ColinWright · 145 comments
I like the Internet. I am old enough to remember the pre-Internet era and despite the younger generations pining for those simpler days, I was there. Paper maps were absolutely horrible, just you and a compass in your car on the side of the road in the middle of the
What HN said:
schoen: I was a great admirer (and later friend) of Barlow, and I'm still very deeply influenced by the Declaration and many adjacent phenomena. I agree with some fraction of this post in terms of seeing many people shelving these principles when it gets inconvenient for them.
randallsquared: > examples of the ideology that powered and continues to power tech Would that it were so. Semi-connected rant: What happened to so many startups to kill the mood was the pattern of: Do something technically legal (or technically illegal!) in a way that seems fixable at first, sc...
jancsika: > Once when driving from Michigan to Florida I got so lost in the middle of the night in Kentucky that I had to pull over to sleep and wait for the sun so I could figure out where I was. Not sure what's going on here, but this reads like 90s cosplay.
loloquwowndueo: Dunno man, those things you say were “horrible” before the advent of mobile phones, media players and gps (not even the internet; usable incarnations of those inventions were entirely independent from the internet) - I was also there and it was fine.
174 points by Georgelemental · 51 comments
GrapheneOS has released a new update that fixes a recently disclosed Android VPN bypass flaw capable of leaking a user’s real IP address.
What HN said:
fg137: Side question: what's a good way of getting a GrapheneOS phone? I have been interested in using GrapheneOS but hesitant about actually getting a Pixel phone. Used phone prices are usually >$300 even for "a" series unless I go back several generations.
nottorp: > Because system_server operates with elevated networking privileges and is exempt from VPN routing restrictions So a VPN isn't a VPN on Android? Regardless of this bug. Do other locked down operating systems act the same?
hedora: > Google maintained its position, authorizing public disclosure on April 29. I'm surprised they honored the embargo at that point, and delayed the fix until May. Why not just release immediately?
idovmamane: The technical detail that makes this egregious is that the leak happens in system_server, a privileged process. Android’s own lockdown mode explicitly promises that no traffic bypasses the VPN.
168 points by Tomte · 42 comments
Armed with a slew of new instruments, physicists are closing in on one of nature’s oldest mysteries — and finding that storm clouds are seething with violent and unexpected phenomena.
What HN said:
nomilk: That 7 second video of a small rocket shot into a cloud to induce a lightning strike (about half way down the article) is incredible. https://www.youtube.com/watch?v=5BJIiX9_c_M Any ideas why the lightning strike appears mostly green (and momentarily purple and orange)?
wpollock: I found this article interesting but lacking. Lightning also sometimes travels from the ground up to the clouds. Storm clouds produce red sprites (there are some theories about these) and blue jets, that shoot upwards towards space. Then there's ball lightning.
raulparada: If that new theory turns out to be somewhat right, there'll be something humbling about ancient greeks stories of Zeus sending Hephaestus bolts from ~'heaven/the cosmos' being closer to it than our modern explanations all along
somedrag: There's a Feynman lecture abput electricity in the atmosphere that is interesting to read alongside this article: https://www.feynmanlectures.caltech.edu/II_09.html
Business/Tech
Other
Programming
Science
Security
Stories and comments sourced from Hacker News public API. Not affiliated with Y Combinator or Hacker News.