frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

The part of Navier-Stokes no one is talking about

https://www.johndcook.com/blog/2026/09/09/formal-method-revolution/
70•ibobev•54m ago•49 comments

Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra

https://cognition.com/blog/swe-2
317•seelos•6h ago•130 comments

More questions about whether researchers can trust OpenAI with unpublished math

https://mathstodon.xyz/@andreasthom/117240535270608201
489•pred_•15h ago•520 comments

OpenAI Agents API

https://developers.openai.com/api/docs/guides/agents-api/overview
42•aquir•2h ago•35 comments

NASA Color Trick Was Meant for Mars. Now It's Unveiling Rock Art on Earth

https://gizmodo.com/this-nasa-color-trick-was-meant-for-mars-now-its-unveiling-rock-art-on-earth-...
225•gumby•6h ago•36 comments

Don't let anyone take away your big box of cables

https://blog.jim-nielsen.com/2026/hands-off-my-cables/
213•Brajeshwar•6h ago•167 comments

NTSB Issues Investigative Update on B-767 Runway Excursion Accident in Miami

https://www.ntsb.gov:443/news/press-releases/Pages/NR20260909.aspx
20•mckn1ght•46m ago•13 comments

Proof of Capture: Apple Reference Image, but open source and using steganography

https://merybenavente.me/blog/proof-of-capture
27•merybenavente•2h ago•28 comments

Music Theory for the 21st-Century Classroom

https://musictheory.pugetsound.edu/mt21c/MusicTheory.html
115•aanet•5h ago•56 comments

Shopify moves back to Native from React Native

https://shopify.engineering/back-to-native
629•fnthawar2•8h ago•425 comments

Forgejo <=16.0.3 Critical RCE

https://codeberg.org/forgejo/forgejo/src/branch/forgejo/release-notes-published/16.0.4.md
122•weierstass•6h ago•47 comments

What happens when a GPU writes memory

https://blog.doubleword.ai/what-happens-when-a-gpu-writes-memory
27•ibobev•2d ago•1 comments

Hitachi launches CO2 heat pump water heaters with solar-friendly tariff controls

https://www.pv-magazine.com/2026/09/07/hitachi-launches-co2-heat-pump-water-heaters-with-solar-fr...
266•thelastgallon•1d ago•211 comments

Rust is tier-1 language at Microsoft

https://rustfoundation.org/media/guest-post-rust-is-tier-1-language-at-microsoft/
556•mmastrac•8h ago•303 comments

Bodily Oddities

https://vester.si/bodily-oddities/
10•vesterde•1h ago•12 comments

Neki

https://planetscale.com/blog/introducing-neki
178•simon_weber•6h ago•72 comments

Show HN: Vertumnus – printable posters of farmers' market produce seasonality

https://vertumnus.fyi
12•perspectivezoom•2d ago•3 comments

JEP 544: Ahead-of-Time Code Compilation

https://openjdk.org/jeps/544
53•Skinney•4h ago•22 comments

Silicon Valley Is Transforming the Military-Industrial Complex

https://costsofwar.watson.brown.edu/paper/how-big-tech-and-silicon-valley-are-transforming-milita...
129•paimapi•6h ago•242 comments

Douglas Hofstadter: Analogy as the Core of Cognition [video]

https://www.youtube.com/watch?v=n8m7lFQ3njk
114•tosh•4d ago•64 comments

Creativity is the New Moat

https://www.inventbuild.studio/blog/genuine-creativity-is-your-new-moat
98•virgil_disgr4ce•3h ago•52 comments

Detecting and countering misuse of AI: September 2026

https://www.anthropic.com/threat-intelligence-report-september-2026
42•garo-pro•4h ago•21 comments

Cognition's SWE-2 achieves 92.8 on Terminal-Bench 2.1

https://tokenstead.ai/models/swe-2
45•cdnsteve•5h ago•21 comments

What algorithm did Windows XP use to choose your initial user picture?

https://devblogs.microsoft.com/oldnewthing/20260909-00/?p=112683
329•soheilpro•13h ago•160 comments

iPhone Duo

https://www.apple.com/iphone-duo/
1400•thecosmicfrog•1d ago•2416 comments

List of references on Sony websites to players "owning" their digital games

https://consumerrights.wiki/w/Sony_PlayStation_digital_game_ownership_lawsuit
336•haunter•9h ago•110 comments

Compute-efficient pretraining and scaling to trillion-parameter models

https://magic.dev/blog/pretraining#
105•ronfriedhaber•2d ago•58 comments

Schemy Lisp En DOS

https://sled.neocities.org/
74•AlexeyBrin•4d ago•14 comments

Stockfish 19

https://stockfishchess.org/blog/2026/stockfish-19/
256•atiedebee•3d ago•145 comments

Python sets and dictionaries can have quadratic-time performance

https://lemire.me/blog/2026/09/03/python-sets-and-dictionaries-can-have-quadratic-time-performance/
83•ibobev•2d ago•47 comments
Open in hackernews

The part of Navier-Stokes no one is talking about

https://www.johndcook.com/blog/2026/09/09/formal-method-revolution/
66•ibobev•54m ago

Comments

zem•39m ago
not to take away from the author's appreciation of newly accessible formal proofs, but people have been talking about the savings in formalization effort for longer than they have been talking about the AI doing the actual proofs!
parhamn•31m ago
They estimated $40M of agent costs (it was a large fleet of them). Using the number in the post its closer to ~880,000 hours × $150/hour = $132 million for the human case. Still an amazing feat not quite "four orders of magnitude". The comparison is obviously pointless because coordinating 1M hours of intellectual labor isn't easy to say the least.

Very exciting and uncertain times!

pkal•30m ago
IMO the "forty hours per page" rule is not up to date, and more a consequence of lacking proof automation in 2005. From what I understand about Lean, this has been one of the things that they have put a lot of effort into improving, making proof mechanization more palatable to the mathematically inclined, as opposed to just logicians.
Jblx2•16m ago
What is your estimate for the number of hours to formalize one page of undergraduate mathematics? Maybe you are saying this is close to zero, if/when Mathlib eventually covers all of undergraduate math?
boshalfoshal•29m ago
People seem to be talking about anything except the actual results with this particular announcement.

Its still astonishing that any sort of generalized computer program can solve a problem of this magnitude, and we have witnessed it happening in real time.

kpil•23m ago
Unless they just swiped the workbooks of the actual mathematicians that where working on the problem using AI and it's in the "next-gen" training dataset.
sho_hn•22m ago
> People seem to be talking about anything except the actual results with this particular announcement.

To be fair, most people have a fairly good handle on "Does opting out my prompts from training runs actually work?", but not on Navier-Stokes. They discuss what more immediately affects them.

btown•17m ago
Heck, it’s even astonishing that any sort of generalized computer program could even verify a proof of this magnitude that hasn’t already been codified in a formal verification language. If, and it’s unclear that we’ll ever get the full story, they did draw inspiration from training on (or even directly accessing) rough notes that had been provided by another researcher in prose… the fact that it could leap so rapidly to a full formal verifiable Lean program for the entire scope of the problem is an incredible result in its own right.
dooglius•12m ago
I mean, I have a bachelor's in math and I don't imagine I could begin to understand either the human or LLM proofs without a massive investment of time and effort.
aabhay•29m ago
Formalizing proofs in Lean has gotten dramatically easier since the formalizations available in 2005. And Lean’s mathlib has done most of the underlying work so that you have its axioms and necessary lemmas baked in. You can think in terms of standard abstractions that look very much like the exact notation in the undergrad textbook.

That said, I am not in any way trying to discount how incredible of an achievement it is to formalize a millennium prize winning algorithm in Lean. I mean just look at the code that OpenAI published. It’s like an encyclopedia of different fluid dynamics concepts.

stabbles•25m ago
It's kinda funny to realize that Lean is apparently so slow that for Fermat's Last Theorem proof verification runs only 1 order of magnitude faster than agents could generate the Lean code (15h verification with 230GB of RAM vs 11 days to generate it).

To what extent can you optimize Lean? It has to be simple enough to be auditable, does that mean you cannot use opaque optimizations to make it run faster?

redox99•20m ago
Can you use Lean to... prove "Lean-fast" is equivalent to Lean?
gcgbarbosa•17m ago
Maybe, but how many centuries would it take to prove it?
calebkaiser•13m ago
Yeah, in essence. This is actually a pretty cool part of working in Lean. It's a somewhat normal convention to write something in a human readable way and then write a second optimized implementation with some kindness of correctness theorem connecting them. There was a whole open "competition" for writing a faster Lean kernel/proof checker that didn't sacrifice on soundness called Lean Kernel Arena. Fun reference point: https://kim-em.github.io/blog/2026-7-24-why-lean-is-faster-t...
stabbles•6m ago
Great read, thanks for sharing
AndrewKemendo•23m ago
People are exhausted from being told/shown the thing they thought was special or unique or could make them relevant, is another mechanical puzzle that can be solved without joy.

I don’t see that doing anything but intensifying in the short term

QwenGlazer9000•15m ago
> But you see, now you'll have more time for the actual important things!

> Like what?

> Cleaning shit out of clogged toilets!

bethekidyouwant•8m ago
How about figuring out how to turn all of our shit into usable fertilizer?
efnx•22m ago
I heard a rumor (on instagram, so YMMV) that the professor who was closest to solving this problem had only weeks ago used Codex, which had slurped up all his notes on the subject. Now OpenAI's agents solve the problem. If it's true that seems like quite a coincidence.
bethekidyouwant•9m ago
How could they possibly included in the previous training run which takes months to complete..
lordnacho•17m ago
How do you know that it's formalizing what you think it's formalizing? If your Lean 4 has a bug, won't you be proving something other than what you thought?
returningfory2•12m ago
Yes, you need to manually verify the statement of the theorem of interest of formalized correctly. But you don't need to anything more than this: you can rely on the proof being correct. And the proof is overwhelmingly the most amount of code.
ramesh31•10m ago
>Its still astonishing that any sort of generalized computer program can solve a problem of this magnitude, and we have witnessed it happening in real time.

I think about this a lot. I'll have to explain to my kids some day that there was long period of time where you couldn't just talk to a computer and have it talk back to you, and that communicating with one required special skills that took years of study to master. It's going to be completely impossible for them to even remotely understand what that was like. Sort of like the pre-electricity days for us, but even more-so.

sho_hn•8m ago
You're assuming you'll be the one doing the explaining :-)

It might also be that they won't even ask or wonder, similar to how most don't really do with pre-machining skills.

Or it could be like our "How did they build the Great Pyramid?!"

TrackerFF•7m ago
It also needs to be said: The amount of compute that went into this is something. From some estimates I've seen, the compute cost alone would be around $10m, +/-

As a reference, for that kind of money one could put together a research group of 20-25 researchers, and keep them salaried for 5 years.

So while it is impressive, absolutely no doubt there, the SOTA access is so expensive that it is sort of unobtanium.

Luckily, the prices have historically reduced by a factor of 5-10 every year...but still, only those that swim in cash can afford this.

Yizahi•6m ago
Aren't you doing exactly the same thing as people you are mentioning? Skipping "talking about actual results" to talking about general capabilities of this LLM and computers in general? because that's exactly what seems like 99% of all people had been doing lately - debating what computer programs can do and what they can't.
andrewchambers•19m ago
If they aren't already, or if its possible, prove that an optimized version matches the simple version...
QwenGlazer9000•17m ago
It's because anthropic vibemathed it. I forgot the name but some other guy is working on a handwritten version of it and I bet it'll be more than just 1 magnitude faster.
andrewchambers•14m ago
They could probably vibe-optimize it if they cared.

What would happen if they give an equivalent agent swarm the proof and a target to reduce runtime .

dist-epoch•15m ago
Nobody wrote 13 mil lines proofs before.

I'm pretty sure you can make Lean at least 10 times faster if you unleash the agents on it.

Somebody ported Doom to run entirely in the TypeScript TYPES (not code). It took 12 days to compile.

https://www.tomshardware.com/video-games/porting-doom-to-typ...

advisedwang•8m ago
But what hardware was the verification vs agents on? Because you are likely comparing verification on a single beefy machine (say XX TFLOPS total) to agents running on a substantial inference cluster (say XXXX TFLOPS). So you're 1 order of magnitude might actually be 2-4 orders of magnitude.
dooglius•8m ago
Weren't the agents massively parallel, whereas the lean verifier presumably is not? Also, I presume said agents were themselves checking their own parts many times.