News for hackers. Stuff that matters.

alberto-m

Why should any mathematician invest decades of their life working on one of them, knowing that all their efforts could be vanified by the release of the N+1-th version of a LLM, or by the usage of 10x compute resources by companies with infinite dollars? It was VERY different when the only competition came from other biological brains that had to invest the same amount of time and effort.
Re:REA Reverse – Engineer Anything on Saturday October 10, 2026 @08:12AM (#50032146)
> For the first two, it is irrelevant what someone else has to say or do about it; ignore the noise.

We are not robots and telling people how they should feel and how their mood should not be influenced by this or that is akin to saying to depressed people to just stop being depressed. The divide-and-conquer method you appear to be using is worthless for human psychology.

Re:REA Reverse – Engineer Anything on Saturday October 10, 2026 @05:33AM (#50031173)
> So perhaps then, the sentiment being as negative as it is, is revealing something about the hobby: that many participants, considered the metagame of being the first to do it, to be "smart enough" to have figured it out, the "street cred"

This is a very ingenerous interpretation.

I am a hobby hiker. My mountain hikes are so easy that basically anybody could do them: no street cred involved, it’s not even worth to speak about them. But it is important to me that most goals are only reachable by walking and putting in the effort: if a company offered a 20$/mo service that brings subscribers to the endpoint, carried in a luxury litter by robotized carriers, I would feel that my hobby has been devoided of sense.

Re:Sharing AI progress in mathematics on Wednesday October 07, 2026 @07:31AM (#49991275)
Mathematicians have been unpaid for centuries. The problem at hand is much deeper then just deciding who gets the taxpayer money.
Re:Sharing AI progress in mathematics on Wednesday October 07, 2026 @05:17AM (#49990217)
OpenAI: At long last, we have created the Open Problem Strip Miner from classic Terence Tao tweet “Don't Create The Open Problem Strip Miner”.
Re:Sharing AI progress in mathematics on Wednesday October 07, 2026 @04:45AM (#49989975)
> mathematics doesn't exist for their pleasure, it exists to provide tools to solve humanitie's problems

Who decreed that? Mathematics predates capitalism and publish-or-perish by a couple of millennia. Euclid’s Elements were not written to benefit the weapons or medical industry.

Correct: if one takes as basis not the Bruttolohn but the total cost for the company, the sum of taxes and mandatory social contribution can very well reach the half of the income. The method of splitting employer-side and employee-side deductions seems just there to hide this fact.
More precisely, given a salary of 85'000€, about 17'500 (20.5%) goes into income tax and about 16'800 (19.7%) goes into social contributions, though one can reduce the latter by taking a private health insurance*. The remaining 60% (Nettolohn) can be used to purchase things, usually paying 19% of VAT on them (though rent is exempt and food has a lower tax).

So, “half goes away in taxes” is an exaggeration, but I can totally sympathize with the feeling. Especially since the quality of healthcare is declining and the pay-as-you-go pension system is only built to benefit the current generation of pensioners.

* note that, if one is earning less than 77'000€ today or less than 84'150€ in 2027, one is prohibited to switch to a private insurance. One of the most regressive precepts of the German system.

The German value seems very suspect. Glassdoor, for the same job role, says €53,000 to €80,000 per year which is probably lower than the truth (Glassdoor overweighs junior roles) but looks much more realistic to me.
Re:Let's Stop Debating the Gnu Imp Manipulation Program on Thursday September 24, 2026 @10:56AM (#49831536)
I actually think your comment reinforces the point of the author. Zero interest into making things clear or attractive to outsiders, the main point is to be smug towards those poor ignorants.

Good post, and the advanced use of the font typographic settings was alone worth the visit.

Re:The body temperature of Tyrannosaurus rex on Wednesday September 23, 2026 @08:46AM (#49815270)
Missed that, thanks. I searched "Tyrannosaurus" on Algolia before submitting, but it was not sufficient.
Re:16-bit Intel 8088 chip by Charles Bukowski on Wednesday September 23, 2026 @05:26AM (#49813527)
> but the wind still blows over / Savannah

Is this a citation of a former work? There's a famous Italian song titled “And yet the wind still blows” (1978), built in the same format of this poem.

Re:Most people prefer traditional architecture on Wednesday September 16, 2026 @04:49AM (#49723731)
Maybe in “Whither Tartaria” by Scott Alexander, which covers a similar topic.
Re:Most people prefer traditional architecture on Wednesday September 16, 2026 @04:45AM (#49723700)
“Form follows function” is attributed to Louis Sullivan, author of very beautiful buildings. Whatever excuse modernist architects have for inflicting misery upon us laymen, this principle is not one of them.
Re:OpenAI’s Navier-Stokes release included a Lean 4 formal proof on Thursday September 10, 2026 @06:43PM (#49651122)
The other part no one is talking about is the applicability. Navier-Stokes is the most “physical” of the Millennium Problems. Is the exploding solution a mathematical curiosity, just like the Banach-Tarski Paradox does not allow me to double my RAM by cutting my memory modules in five pieces and mounting them back appropriately? Or does it have application in the real world, pointing to hitherto unknown resonance phenomena that could allow to prevent the next Tacoma Bridge incident (or, more sadly, to build new marine weapons)?
Re:On the Navier–Stokes Millennium Prize Problem on Tuesday September 08, 2026 @02:00PM (#49614095)
Since you are a very new account, allow me to inform you that copy-pasting the same comment throughout the thread is very bad form.
Re:Formalizing Fermat's Last Theorem on Friday September 04, 2026 @05:20PM (#49570369)
There are hopefully still some Ludi to play before doing that, Magister.
Re:Paint.NET Adds Experimental Wine/Linux Support on Wednesday September 02, 2026 @07:29AM (#49534723)
> Note: The MIT license does not require credit.

The MIT notice starts with “Copyright (c) <year> <copyright holders>”, and who uses the code must include that line in any derived work.

> Also I rather doubt

Do you doubt Claude's rewrite can be defined as “clean room” (I also do)? Or that Claude is able to rewrite it? Or that it would not have been possible otherwise?

Re:How Europe is killing makers and micro-entrepreneurs on Tuesday August 25, 2026 @02:56AM (#49429999)
> Really the chances that someone will ask a small company to be compliant are just very low.

So EU has replaced rule of law with a system where every SME is a criminal and enforcement is left to chance or to the whim of the regulator (“oh, you criticized von der Leyen on your website? We cannot prosecute you for this, but let's send some inspectors to check if you are complying with the PPWR”).

Italy works like that for tax compliance. I've read USSR also worked in a similar way. The effect on economy and innovation is always catastrophic.

Or Plato's Republic, if one prefers older books.
Much less, the average monthly cost is about 123 CHF (less than 150$). Source: https://www.bag.admin.ch/de/newnsb/d2okh_kUK_OFhmMDfpyiy

Same in Germany: private kids' insurance costs a fraction of the one for adults. Apparently it's very unlikely that they require expensive healthcare.

Did they consent? Just because one receives a letter it doesn't mean they “own” it, much less that they are entitled to publish it at their leisure. If Spirit were active in any country with GDPR-style laws, the seller of these data would be most likely investigated.
Re:More German than many Germans on Tuesday August 04, 2026 @04:15PM (#49174429)
> but the chat absolutely needs to happen in German. Why? Work culture.

It's absolutely normal that people speak their mother tongue when they have the possibility. One can be kind and speak English to make a guest feel more comfortable, but if a permanent resident refuses to learn the local language fluently he should not demand that everyone else bends over backwards to accommodate him.

It's incredible how the “Chinese Room” argument, as well as its counterarguments, is still incredibly pertinent despite being now more than 40 years old. Scientific American published many articles on this topic in the Eighties; they seem as fresh as ever.
Not OP, but my interpretation is this.

Quantum waveform collapse has been proposed as explanation for consciousness, allowing to explain how an entity can have free will and yet obey rigid physical laws: https://en.wikipedia.org/wiki/Consciousness_causes_collapse

I think OP is suggesting a similar thing happened in the LLM, implying it gained consciousness despite following a well-defined compute process.

Re:RFC/The future of Abstract Wikipedia on Sunday August 02, 2026 @10:49AM (#49145178)
I've been away from the Wikipedia environment for more than a decade, but this seems the n-th chapter of the WMF implementing something that the contributor community does not want, but will help to pad the CVs of managers and developers of the Foundation itself. Just like LiquidThreads, the Media Viewer, etc.

Older »