Top 30 stories from Hacker News. Top 10 include comment highlights. Compiled at 20:10 UTC.
677 points by dwwoelfel · 341 comments
Because I am known to write using a fountain pen on paper, a number of people have pointed me to this post and its underlying research. I won’t rehash what is said in those sources, but the gist of it is that when you write things down by hand you’re recruiting more of your brain...
What HN said:
postalcoder: A corollary to this is, when reading books, mark them up! Don’t be precious with them. Destroy the spine so it’s easier to read, fold pages, write the important stuff inside the front cover of the book.
Wowfunhappy: I've always been pretty skeptical of the data which says handwriting is better for learning than typing. Yes, handwriting causes more brain activity, but just because your brain is doing more does not mean it's doing the right thing! For example, if I asked you to code FizzBuzz w...
apparent: I find his dismissal of iPad writing to be unconvincing. He talks about how we have calibrated our writing utensils over centuries to involve the proper amount of friction, and how that knowledge goes out the window when you're writing on an iPad.
chris_money202: My thoughts on this is there is a deliberate act going on that is doing more for retention than just the simple act of using a (fountain) pen. If I am in a lecture and I am writing down everything I hear as fast as possible in one ear, on paper, out the other style I don't think...
525 points by theanonymousone · 498 comments
What HN said:
capevace: I‘m not even sure what the argument for banning Chinese models/open weights even is supposed to be? 1. if it’s to stop hackers doing hacking things with „uncontrollable models“ then, well… they’re already doing something illegal to begin with, why would they care about breaking a...
avaer: Besides serving no purpose other than an attempt to enforce software by fiat (which would hurt legitimate business, make US AI companies noncompetitive, and do nothing to stop illegitimate use)... it's literally impossible to enforce.
vkaku: Read this thread when you get the time: https://x.com/_vkaku/status/2080352797606744209 Open Data+Open Models gives everyone else an advantage and bringing regulatory capture here should be appealed and brought to the FTC and the courts to challenge such regulations.
GodelNumbering: Proprietary model weights are IP, their outputs are not IP and I can't imagine a court decision that would rule otherwise because that would set an extremely far reaching and dangerous precedent, even for American businesses.
464 points by technewssss · 225 comments
AI companies are pouring tens of billions of dollars into enormous data centers. They're being built on top of a mountain of hidden debt.
What HN said:
senshan: As long as this debt does not make it into life insurance and pension funds, we are fine. The trouble is that private credit is taking control of some life insurance companies and off-loads this debt to these. When these fail, it will become everyone's problem.
wongarsu: Do they? Is a company with $200 billion annual revenue and earnings (EBITDA) of $100 billion having $420 billion of off-balance-sheet debt really staggering? In many other industries that would be a perfectly normal amount of debt to have.
chasd00: Are they really "trying to hide" this debt? I think it's pretty common knowledge that a lot of these companies are using debt/bonds for funding. The debt not showing up where the author wants is a reporting formality not an attempt to hide it.
jimnotgym: If I were nearing retirement and had a decent pension pot where I could control it in fine detail...I would be diversifying away from tech stocks and holding some cash for immediate needs. There probably won't be much time when it unravels...
186 points by mpweiher · 31 comments
What HN said:
articulatepang: I went through this a few months ago in Rust. I wrote all the code by hand, no LLMs. Then I went ahead and added a small "game" on top, plus some special effects like pixelization shaders and chromatic aberration at the edge of a flashlight. https://github.
nkanaev: This resource, along with Mathematics for Computer Graphics by John Vince [1], was truly indispensable when I wrote my own software renderer [2]. This was long before LLMs, so the whole process took me at least a couple months - most of it trying to wrap my head around math behin...
bob1029: I wish we could have just one of these tutorials properly cover the concern of triangle clipping. This is the part that I struggle with the most in a software renderer. If you are going to be building a practical one, this is something you will eventually have to deal with, even...
espetro: Finally an engineering feat that's not built in Rust
180 points by nickthegreek · 67 comments
If you work in or around the film industry, there is a decent chance you have used the work of The Numbers this month, whether you realise it or not.
What HN said:
primitivesuave: A couple years ago, I was running a website which allowed the public to view all the US government handouts to small businesses during the COVID-19 pandemic. It also tracked all the fraudulent loans being prosecuted by the DOJ, and allowed anyone to run structured queries over th...
ethagnawl: At the risk of oversimplifying things from a distance, this site -- especially the free, public-facing part of it -- seems like it would be an ideal candidate for a rewrite using static site generator/framework.
abetusk: I think people are missing one of the points of the article. It's not just that agents are hammering the site, it's that there might be lurking vulnerabilities that allow malicious usage, which is why it went down, then came back up with a fraction of the data and a reduced desig...
ajkjk: It feels like there is fundamentally missing infrastructure here that is needed to make these problems go away. Basically bots need to be (somehow) paying for the traffic they create, or prevented from creating it, or told to go away and then fined if they violate the request.
161 points by MarcoDewey · 56 comments
Observations made with the European Southern Observatory’s Very Large Telescope (ESO’s VLT) have revealed evidence for a moon-like object in the CD-35 2722 system. Unlike moons in our Solar System, the newly found object does not orbit a planet, raising questions about what to na...
What HN said:
delta_p_delta_x: Worth noting that the artist's impression is... not accurate. Both CD-35 2722 b (the brown dwarf orbiting the primary star) and CD-35 2722 b I (the exomoon orbiting the secondary) should be much closer in size.
hdz: A small dense planet, which orbits a star, is orbited by a larger gaseous moon. It's interesting the discovery is in Chile which has some of the best night skies, a Class 1 on the Bortle scale specifically in the Atacama Desert. Hoping to make it out there one day.
pelagicAustral: I know this is besides the article and all, but why would the Chilean flag be the only one to have 8px left margin? This is killing me...
HelloUsername: An important phrase from the article to consider before commenting: "This system is somewhat hard to define using Solar-System-based words like ‘planet’ and ‘moon’"
127 points by ibobev · 72 comments
Learn OpenGL . com provides good and clear modern 3.3+ OpenGL tutorials with clear examples. A great resource to learn modern OpenGL aimed at beginners.
What HN said:
cyber_kinetist: The one and only Holy Bible of Graphics Programming. If you're starting to learn computer graphics, just study through the entire site and do the examples one by one. It doesn't matter one bit that it uses a slightly outdated API called OpenGL - you're supposed to learn how to re...
brasswood: Cem Yuksel's lecture videos for Interactive Computer Graphics at the University of Utah are another free and fantastic resource. https://www.youtube.com/playlist?list=PLplnkTzzqsZS3R5DjmCQs...
AyanamiKaine: If you want to use the learned knowledge I would recommend to use something like Sokol [0] or use the SDL-GPU API [1]. While Sokol is more higher level both can be used. Otherwise learn opengl is a really good introduction. [0] https://github.com/floooh/sokol [1] https://wiki.
the__alchemist: In 2026: Please don't learn this; it's obsolete. Learn Vulkan (Or another modern API; or how to build engines on top of them; or how to write shaders; or how to do GPU compute using CUDA etc)
107 points by dhorthy · 84 comments
Contribute to humanlayer/advanced-context-engineering-for-coding-agents development by creating an account on GitHub.
What HN said:
fishtoaster: There's some good ideas and points in here, but this bit threw me: > # We tried this > In July 2025 we went full lights-off Isn't it pretty well-accepted at this point that the models underwent a step-change in usefulness around fall 2025 / spring 2026? I know that I was able to...
rglynn: To me, the thing that stands out about the whole state we're in here is PR review. Yes, in an ideal world, PRs read well, are a joy to review, reflect what you discussed etc etc. We have to be real; there is only so much we can do to that end.
Altern4tiveAcc: >It's easy to be a little bummed by the core thesis here: "for now we're stuck reading the code". >I was pretty excited for a world where we could just ask for things and let the models cook and not read the code and get beautiful production software I feel so disconnected readin...
maerF0x0: I'm currently learning about Cost functions and Regression in the Machine learning course on coursera[1], and I cannot help but be struck by the similarities of how gradient descent seeks to minimize the error between the model and the training data, how agents do the same (far l...
103 points by r2sk5t · 89 comments
Historic VENOM milestone demonstrates scalable AI development capabilities for the operational fleet.
What HN said:
1970-01-01: All Stealth bombers are upgraded with Cyberdyne computers, becoming fully unmanned. Afterwards they fly with a perfect operational record. The Skynet funding bill is passed. The system goes on-line on August 4, 1997. Human decisions are removed from strategic defense.
SoftTalker: > The kit utilizes a novel interface with the aircraft’s flight controls and mission systems, allowing a pilot to toggle between traditional human control and AI control with the flip of a switch. This ensures a safe, reliable environment for human-on-the-loop experimentation.
Zsfe510asG: The article smells of AI psychosis. They are already "orchestrating". In the first Iraq war, they went in and won. In the second one, it took years to sort of occupy the country. Now they struggle against Iran and "orchestrate" F-16s that can be shot down instantly by AA weapons.
odyssey7: Tangent: is AI a bearable arm?
103 points by jjfoooo4 · 69 comments
The release of Kimi K3 has opened a fresh round of angst and confused discourse. There's a loud cohort of journalists, business leaders, and politicians arguing that open source AI is a dangerous threat. OpenAI's Dean Ball:
What HN said:
petcat: This is not "open source" AI. Photoshop source code + OSI license = open source Photoshop binary = open weight Photoshop SAAS web app = closed model like GPT, Opus/Fable etc. There is nothing "open source" about the Chinese models in question.
nater5000: >Much of the angst around China's models centers on "losing the AI race". But what's the goal of this race? Is it to develop the best model? To sell the most tokens? To destroy humanity first? Some people would say it is reaching some sort of singularity.
MostlyStable: Just to add on to other comments: reasonable people can disagree about the degree of safety concern with near to medium term AI. But to not address the arguments at all is, in my opinion, a serious mark against the value of this article.
elmer2: We have continued to find backdoored Chinese manufactured routers and network devices. Will there ever be a way to fully audit Chinese models?
AI/ML
Other
Programming
Science
Stories and comments sourced from Hacker News public API. Not affiliated with Y Combinator or Hacker News.