frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

I extracted the safety filters from Apple Intelligence models

https://github.com/BlueFalconHD/apple_generative_model_safety_decrypted
190•BlueFalconHD•2h ago•80 comments

LLMs should not replace therapists

https://arxiv.org/abs/2504.18412
41•layer8•1h ago•34 comments

Show HN: I wrote a "web OS" based on the Apple Lisa's UI, with 1-bit graphics

https://alpha.lisagui.com/
209•ayaros•4h ago•65 comments

Building the Rust Compiler with GCC

https://fractalfir.github.io/generated_html/cg_gcc_bootstrap.html
28•todsacerdoti•59m ago•0 comments

More than 1 in 5 Show HN posts are now AI-related, get > half the votes/comments

https://ryanfarley.co/ai-show-hn-data/
169•rfarley04•2d ago•98 comments

Why English doesn't use accents

https://www.deadlanguagesociety.com/p/why-english-doesnt-use-accents
26•sandbach•1h ago•6 comments

A non-anthropomorphized view of LLMs

http://addxorrol.blogspot.com/2025/07/a-non-anthropomorphized-view-of-llms.html
6•zdw•19m ago•0 comments

Jane Street barred from Indian markets as regulator freezes $566 million

https://www.cnbc.com/2025/07/04/indian-regulator-bars-us-trading-firm-jane-street-from-accessing-securities-market.html
154•bwfan123•8h ago•81 comments

Async Queue – One of my favorite programming interview questions

https://davidgomes.com/async-queue-interview-ai/
68•davidgomes•5h ago•36 comments

Get the location of the ISS using DNS

https://shkspr.mobi/blog/2025/07/get-the-location-of-the-iss-using-dns/
245•8organicbits•10h ago•75 comments

Functions Are Vectors (2023)

https://thenumb.at/Functions-are-Vectors/
140•azeemba•7h ago•71 comments

Intel's Lion Cove P-Core and Gaming Workloads

https://chipsandcheese.com/p/intels-lion-cove-p-core-and-gaming
7•zdw•18m ago•0 comments

Backlog.md – CLI that auto-generates task files (took my Claude success to 95 %)

https://github.com/MrLesk/Backlog.md
45•mrlesk•2h ago•12 comments

Opencode: AI coding agent, built for the terminal

https://github.com/sst/opencode
82•indigodaddy•5h ago•18 comments

Nobody has a personality anymore: we are products with labels

https://www.freyaindia.co.uk/p/nobody-has-a-personality-anymore
3•drankl•29m ago•1 comments

I don't think AGI is right around the corner

https://www.dwarkesh.com/p/timelines-june-2025
84•mooreds•2h ago•95 comments

Lessons from creating my first text adventure

https://entropicthoughts.com/lessons-from-creating-first-text-adventure
12•kqr•2d ago•1 comments

Corrected UTF-8 (2022)

https://www.owlfolio.org/development/corrected-utf-8/
29•RGBCube•3d ago•20 comments

Metriport (YC S22) is hiring engineers to improve healthcare data exchange

https://www.ycombinator.com/companies/metriport/jobs/Rn2Je8M-software-engineer
1•dgoncharov•5h ago

Cool People [pdf]

https://www.apa.org/pubs/journals/releases/xge-xge0001799.pdf
60•ilamont•5h ago•14 comments

Mirage: First AI-Native UGC Game Engine Powered by Real-Time World Model

https://blog.dynamicslab.ai
13•zhitinghu•21h ago•8 comments

Hannah Cairo: 17-year-old teen refutes a math conjecture proposed 40 years ago

https://english.elpais.com/science-tech/2025-07-01/a-17-year-old-teen-refutes-a-mathematical-conjecture-proposed-40-years-ago.html
314•leephillips•7h ago•69 comments

Toys/Lag: Jerk Monitor

https://nothing.pcarrier.com/posts/lag/
41•ptramo•8h ago•35 comments

Collatz's Ant and Σ(n)

https://gbragafibra.github.io/2025/07/06/collatz_ant5.html
19•Fibra•5h ago•2 comments

Overclocking LLM Reasoning: Monitoring and Controlling LLM Thinking Path Lengths

https://royeisen.github.io/OverclockingLLMReasoning-paper/
46•limoce•9h ago•0 comments

1945 TV Console Showed Two Programs at Once

https://spectrum.ieee.org/dumont-duoscopic-tv-set
30•pseudolus•1d ago•9 comments

Serving 200M requests per day with a CGI-bin

https://simonwillison.net/2025/Jul/5/cgi-bin-performance/
294•mustache_kimono•22h ago•251 comments

Hidden interface controls that affect usability

https://interactions.acm.org/archive/view/july-august-2025/stop-hiding-my-controls-hidden-interface-controls-are-affecting-usability
536•cxr•23h ago•382 comments

Show HN: Simple wrapper for Chrome's built-in local LLM (Gemini Nano)

https://github.com/kstonekuan/simple-chromium-ai
17•kstonekuan•4h ago•1 comments

Can we test it? Yes, was can [video]

https://www.youtube.com/watch?v=MqC3tudPH6w
56•zdw•3d ago•63 comments
Open in hackernews

Hannah Cairo: 17-year-old teen refutes a math conjecture proposed 40 years ago

https://english.elpais.com/science-tech/2025-07-01/a-17-year-old-teen-refutes-a-mathematical-conjecture-proposed-40-years-ago.html
314•leephillips•7h ago

Comments

scythe•6h ago
Paper here:

https://arxiv.org/abs/2502.06137

I had the opportunity to take a harmonic analysis course in grad school. I passed it up. It was only tangentially related to my research at the time.

munchler•4h ago
I had never heard of the X-Ray Transform until I happened to read about it in the New York Times today, and then here it is again.

https://www.nytimes.com/interactive/2025/06/30/science/math-...

dekhn•4h ago
This sort of transform (what I think many people call inverse problems) is quite common in reconstruction problems- that is, where you pass light or other EM through an object, the light scatters, and hits a detector. Typically you want to find the minimum error reconstruction. See more here: https://en.wikipedia.org/wiki/Radon_transform
schoen•1h ago
That's a funny coincidence that the "Radon" there is a person's name and not the radioactive element or its emanations.
echoangle•1h ago
Not really related but similar situation and I found it funny when I realized it:

The Poynting-Vector (https://en.wikipedia.org/wiki/Poynting_vector) indicates the direction of the energy flow of a magnetic field, but it’s named after a physicist called „Poynting“, not because it is „pointing“ somewhere. I thought the „y“ in poynting was a typo when I first read it.

rendall•6h ago
https://archive.is/Nr1hH
old_man_cato•6h ago
[flagged]
dang•2h ago
Could you please stop posting unsubstantive comments to Hacker News? we're trying for something different here.
kemitchell•5h ago
Refuted?
qsort•5h ago
The Mizohata-Takeuchi conjecture is a statement in the form "For all <x> (a bunch of math)".

Showing that there exists an x such that the statement is false disproves the conjecture.

She found a counterexample.

mgiampapa•4h ago
She found more than one way of disproving it in the process.
gilleain•5h ago
Yes, found a counterexample to the conjecture.
zahlman•5h ago
Yes, either proving a true conjecture or refuting a false one is "solving" it.
Keyframe•5h ago
Original title is more informative than the edited one here.
leephillips•4h ago
I submitted under an approximation of the original title, and it was edited within seconds.
miles•4h ago
There is too much "helpful" title modification of late. The original title itself fits within HN limits:

"A 17-year-old teen refutes a mathematical conjecture proposed 40 years ago"

The site's guidelines are clear[1] but increasingly ignored by some moderators:

"...please use the original title, unless it is misleading or linkbait; don't editorialize."

[1] https://news.ycombinator.com/newsguidelines.html

miles•46m ago
After several hours, the HN title has been modified again to match the original much more closely, from "Hannah Cairo has solved the Mizohata-Takeuchi conjecture"[1] to "Hannah Cairo: 17-year-old teen refutes a math conjecture proposed 40 years ago".

[1] https://web.archive.org/web/20250706185220/https://news.ycom...

layer8•2h ago
As the submitter, you can re-edit the title after submission (for some limited time period).
ledauphin•5h ago
here's a dumb question:

she's starting her Ph.D. this fall - hasn't she already achieved it? What is the theory behind expecting someone who has solved a decades-old problem to do some "second" thing to prove that they have extended the bounds of human knowledge?

nextos•5h ago
A PhD in the US requires a lot of coursework, aside from research. Perhaps, she is interested in that. Otherwise, some universities, especially in EU, offer PhDs by publication. She could simply wrap up her counter-example publication (https://arxiv.org/pdf/2502.06137) as a thesis and possibly graduate. Sometimes, you can even do this without a supervisor.
xg15•4h ago
Sounds as if she even has a potential supervisor:

> “It took me a while to convince Ruixiang Zhang [the professor of the course where the problem had been posed] that my proposal was actually correct,” Cairo says

> At the University of Maryland, she will continue working under the supervision of Zhang. “He helped me so much, and I’m really grateful. Beyond his class, which I loved, he spent countless hours tutoring me,” she recalls.

pclmulqdq•3h ago
PhD by publication usually takes a bit more work. I think they tend to want 3 related papers in a field.
stogot•3h ago
What level or type of publication is required?
pclmulqdq•3h ago
They must be peer-reviewed journal papers and I believe they tend to prefer if at least one is well-cited or significant, especially if you have only three papers. It is generally harder to get a PhD by publication than to get a PhD the normal way.
almostgotcaught•3h ago
That's a rule of thumb for applied sciences. Plenty of theory PhDs graduate with 1 or 0 papers.
pclmulqdq•3h ago
Nobody gets a PhD by publication with 0 publications. This is usually a backdoor for people who have done a lot of work in a field, certainly far more than a PhD thesis, and have just never gotten the credential.
almostgotcaught•2h ago
> Nobody gets a PhD by publication with 0 publications.

Do you have a PhD from a theory department? I do. You're wrong.

pclmulqdq•1h ago
Lots of people get PhDs with no publications.

Nobody gets a PhD by publication without publications. It's literally axiomatic.

almostgotcaught•51m ago
It's amazing how many people on hn are experts on things they do not have the qualifications to be expert on.

> It's literally axiomatic.

You've made up some axiomatic definition of "by publication" that does not bear any resemblance to the actual definition. Consider that it's possible to

1. Submit a preprint to arxiv and have it count

2. Submit a preprint to a journal and defend before it accepted (or rejected)

3. Not submit anything anywhere and have the PhD itself count (almost all PhDs get an ORCID)

pclmulqdq•45m ago
Are you aware that "PhD by publication" is a real thing that is a separate path than a normal PhD? It is relatively common for schools in some European countries to offer these, but not that common outside Europe.

This is a process where you can write your "dissertation" by putting an intro and a conclusion on ~3 papers you have already published and get a PhD that way. You enroll in the school for ~3 months, write the missing parts, and that's it. This is a flexible path to a PhD for industry researchers or other people who have a lot of expertise and have pushed the boundaries of a field but did not do a formal PhD program.

I have never heard of anyone doing this with ArXiV preprints or any school accepting this path if they are not referreed papers. I would love to see an actual counterexample if you have one.

pcrh•22m ago
>You enroll in the school for ~3 months, write the missing parts, and that's it.

There are degree mills that do what you describe.

There is also the format in countries such as Germany or the Netherlands where one typically "bundles" one's publications into a thesis. However, the work is typically done in the context of supervised doctoral programmes and no less rigorous than that done under different PhD studies formats.

pclmulqdq•4m ago
TIL Cambridge University is a degree mill.

https://www.cambridgestudents.cam.ac.uk/exams/students/postg...

pcrh•1h ago
A PhD thesis is itself a publication.

PhDs "by publication" refers to not having to submit work additional that already published to the examining committee.

skissane•54m ago
A PhD by publication sounds very similar to a higher doctorate (DSc, DLitt, etc). Substantive (as opposed to honoris causa) higher doctorates are awarded based on publication record only. To be eligible for a substantive higher doctorate, you generally are expected to have a PhD first - but it might not be an absolute requirement. You’d generally expect a bunch of papers, but in principle a single publication (if sufficiently groundbreaking) could be enough. While this is very impressive for a 17 year old (I wish I could have done that at 17, or at any other age for that matter), it probably isn’t significant enough for a higher doctorate all by itself. If she’d proved P=NP, different story. (Who knows, maybe she shall-well, probably not, but I’d be very happy to be proven wrong about that.)
parpfish•4h ago
But what does somebody do with a PhD at age 17? I can’t imagine hiring them as a prof when they’re so young. It’s not a bad idea to just take a couple years to continue your already productive collaboration while getting mentored on the non-math parts of being a mathematician.
ics•4h ago
IIRC Erik Demaine (https://en.wikipedia.org/wiki/Erik_Demaine) started teaching at 20 and had his PhD. I can't remember if I first saw his name because of the MacArthur Grant or one of those science documentaries but one of his pages was on the frontpage here a week or two ago and it seems like he's been thriving.
daxfohl•38m ago
Noam Elkies too, the youngest ever tenured prof at Harvard. Another parallel he had a pretty famous contradiction proof, of Euler's conjecture. But he didn't find that until age 22, so seems like this girl has a good head start!
tehjoker•2h ago
When she graduates she'll probably be between 20 to 23 years old.
j7ake•2h ago
In math or theoretical fields it’s not unheard of to have young professors. Terence Tao was full professor at 24. Wolfram at 21
parpfish•41m ago
She could still do a few years of PhD training and still be a super young prof like those two.

At 17 she’s so young that a uni hiring her would need to think about child labor laws

beezlebroxxxxxx•2h ago
> I can’t imagine hiring them as a prof when they’re so young

Many institutions would actually jump at the chance. That's way better than a 35 or 37 year old burnt out from just finishing their PhD and getting onto the tenure track suffer-fest. Think of how many years of productive research she has in her. It used to be way more common until academia became so professionalized and bureaucratic.

EvgeniyZh•3h ago
Ph.D. is training in how to do research. Solving one, even very hard problem not necessarily means that you don't need such training. It's especially tricky with counterexamples which sometimes question of raw talent and luck rather than skill.

The next step for someone who has PhD and want to stay in academia is postdoc. After solving one problem, you would not necessarily have what's needed to get a good postdoc, such as clear research agenda or proof of ability to publish consistently.

andy99•1h ago
Modern PhDs are not designed for people that are smart like this. She's a math savant that obviously has a unique and demonstrably effective way of looking at things, why destroy that with "training how to do research".

I hope she's found a program that will support her while realizing she's smarter than whomever is setting the rules, rather than something stifling.

mkl•1h ago
This is not true at all. Maths PhDs are excellent for this kind of person, and foster and grow such abilities. No matter how smart you are, training on how to do research is going to improve your capabilities and success.
eviks•3h ago
There is no deep theory here, bureaucracy doesn't think deep.
daxfohl•3h ago
A PhD is as much a stamp of endurance as it is a stamp of intelligence or accomplishment.
MPSFounder•3h ago
Great question. I have a PhD. People forgot the purpose of a PhD. Hannah effectively achieved what many with a PhD fail to do, and that is contribute novel research. A PhD in the US (only place I can comment on) has lately been focused first and foremost on a) preparing for academia, which entails teaching and a lot of courses, and b) research for industry positions (many students in my cohort were from China or India and this was their segway into a job in the US). I agree a PhD should be purely focused on research and extending human knowledge. In practice, it is a business where students go to conferences to promote their PI's work, where Universities get cheap lecturers in the form of TAs, and where many mediocre students write incremental papers to secure an RnD position (change this by a little and see how it affects your results. This is your paper). I am very impressed by Hannah's work though and she embodies the selfless nature of research that is very much missing. I see too often people seeking to advance their own career and pick a PhD route of least resistance. While they are entitled to maximize profits, and oftentimes do not want to go to academia where solving the impossible is admired, we must remember discoveries often hinge on challenging problems and a selfless pursuit of the impossible. This is just my opinion based on what I saw in my cohort and at 30+ conferences
vladms•1h ago
I do not think there is "one" purpose for a PhD, and not everyone gets the same thing from it - it depends on what are a person's strong and weak points. I have seen very smart people not able to explain at all their work and during a PhD they were forced to improve. I have seen very good presenters that were forced to do some actual work. I have seen people that were convinced they can solve anything (as in some parts of the world the bachelor and master focus to much on solvable problems) understand that sometimes there just isn't a clear, nice solution to a problem.

On average someone that does have a PhD will have a wider set of skills, like understanding of the complexities of the field, resistance to frustration, capability to do research and ability to communicate.

YeGoblynQueenne•2h ago
A PhD can be an opportunity to learn, or an opportunity to brag. I'm guessing the teen in the article is going to go for the first one.
paulpauper•5h ago
Trying to do anything original and novel in math is extremely hard at any age. to do it at 17 is insanely talented. congrats
pillefitz•4h ago
Anything original, for that matter.
marvinborner•5h ago
There's a video by Hannah Cairo that explains the conjecture and her results [1]

Also, Terence Tao hinted at some further advances some time ago [2], does anyone know more about that?

[1]: https://www.youtube.com/watch?v=3ZeH_8sTyKA

[2]: https://mathstodon.xyz/@tao/114003793236630744

mellosouls•4h ago
Yes, presumably this:

https://terrytao.wordpress.com/2025/02/25/the-three-dimensio...

tomjen3•5h ago
>One day, he proposed proving a special, much simpler case of the conjecture as a homework assignment. As an optional part, he included the original conjecture

There is a lesson there: always give people an opportunity to excel, if you can.

sshine•3h ago
I remember at first year of university being presented with a bunch of “simple” problems early on, such as the Collatz conjecture.

I remember wanting to spend time trying to explore what a solution might look like, because such simply formulated problems must have equally simple solutions.

Maturing and getting a better understanding of my intellectual capacity, I have opted to solve practical problems with a much bigger chance of success and absolutely no groundbreaking qualities.

But I liked being taken serious from the start, and I think it’s important to try and solve hard problems before you grow stuck in the real world.

gopher_space•51m ago
I give all of my hard problems to juniors.
raincom•4h ago
Great achievement. Now Princeton Math department will ask her to join their school for Ph.D.
bradly•3h ago
While extremely talented, I am not surprised to find this coming from a teen. Major mathematical discoveries often have come from those in their mid 20’s with the greater discoveries being skewed towards the younger 20s and teens. I think this because pure mathematics is just so creative.
fhdkweig•2h ago
The Fields Medal has a cutoff age of 40 years old.
jebarker•55m ago
My understanding is that that’s more about encouraging younger mathematicians rather than an expectation that older ones won’t produce anything worthy.
TechnicalVault•1h ago
It doesn't help what we've designed a rather silly academic system where principal investigators are forced to spend a good deal of their time thinking where they'll be applying for their next grant. We also optimise the system for short term thinking rather than long games. There are some exceptions in research institutes but I think young people are the ones who have the clearest minds because of it.
jebarker•56m ago
Whenever I hear this claim about younger mathematicians I wonder if it still holds true (or really did historically). For example, Andrew Wiles proved Fermat’s Last Theorem in his 40s and there are numerous examples of productive older mathematicians. But also I think the claim skews towards big flashy problems rather than the work of building mathematical frameworks, finding structural insights and finding connections between disparate areas (which requires broad experience rather than just young intensity).
lordnacho•2h ago
How often does someone produce work that is normally taught to people who are older than the person who discovered it?

Euler was 41 when he discovered his famous identity, the kind of thing people learn in school.

Even Newton was 21 when he invented calculus, the sort of stuff that you might find late teens learning.

Galois by a couple of years? He died at 20, and I suppose they teach that stuff sometime mid uni?

louistsi•1h ago
Galois Theory was a third year second semester course at my university when I studied there, which corresponds to 20-21yo in the UK system.
chengiz•2h ago
> The conjecture was widely believed to be true — if so, it would have automatically validated several other important results in the field — but the community greeted the new development with both enthusiasm and surprise: the author was a 17-year-old who hadn’t yet finished high school.

This article is quite poorly written. Case in point above. If the conjecture was believed to be true, refuting it would be news in itself, deserve more than half a sentence, and have nothing to do with the age of the refuter. It should have been simple to add a line about the "other important results" and not violate show not tell. AlsO I fail to see the relevance of mentioning the Spanish academy? The researcher is from Bahamas/USA, it's just the writer is from Spain?

MarcelOlsz•2h ago
>This article is quite poorly written.

Her last name is misspelt in the very first paragraph as well.

libraryofbabel•2h ago
Oh come on. This is in the Spanish newspaper El Pais. Context and audience matters. It’s simultaneously news about a math problem, an article about a young mathematician, and an article about things that happened at a math conference in Spain, which is where they presumably interviewed her.
andrewinardeer•2h ago
"Ciaro says it required several tools, including fractals, and she had to arrange everything very carefully."

Article could at least spell her name correctly.