Hacker News: dash2

New comment by dash2 in "JetZero"

dash2 — Wed, 29 Jul 2026 04:03:28 +0000

There’s lots of evidence for costly signalling across multiple domains! For this specific case, I can’t be certain about any individual’s motivation, of course - I simply raised the possibility and pointed out why it would be a bad idea to think that way.

New comment by dash2 in "Which Odyssey translation wins a blind reading test?"

dash2 — Tue, 28 Jul 2026 13:43:22 +0000

The original is οινοπα ποντον which indeed is very close to wine-dark sea. Literally "wine-faced" where -faced is a metaphor for "having the appearance of".

New comment by dash2 in "Canceling "Hey""

dash2 — Tue, 28 Jul 2026 06:57:17 +0000

> okay for DHH and other Westerners to find the Roma value system disgusting.

The Nazis murdered about a million gypsies so “finding a whole value system disgusting” can lead to some quite dark places.

New comment by dash2 in "Canceling "Hey""

dash2 — Tue, 28 Jul 2026 06:52:41 +0000

If people are systematically more likely to be in area X, that may be because they are systematically more likely to make the relevant choices - for example because they are under different constraints or have different preferences. Choice is part of social science.

New comment by dash2 in "JetZero"

dash2 — Tue, 28 Jul 2026 06:01:58 +0000

> That term lacks all credibility

I disagree. There's coherent economic theory about how costly actions can be used to provide credible signals. It's been used in a wide variety of disciplines including biology (peacock's tails). No mind-reading is needed. I understand that sometimes it is used by right-wingers to attack left-wing positions. My comment wasn't driven by reactionary politics, and my point was to encourage people to think about what actually does well for the group - which, at the risk of repeating myself, is about minimizing the cost for a given level of benefit.

New comment by dash2 in "The Hideous Duty of Isaiah Berlin [pdf]"

dash2 — Mon, 27 Jul 2026 19:38:44 +0000

And here's the outcome and you wonder why they made such an effort: https://en.wikipedia.org/wiki/Wolfson_College,_Oxford#/media... Most Cambridge colleges have some sixties monstrosity and I suppose Oxford ones do too. I've seen worse.

New comment by dash2 in "Jacobian Conjecture for Baby"

dash2 — Mon, 27 Jul 2026 05:16:05 +0000

one thing I didn't understand. If the conjecture is false for real numbers, why isn't it false for complex numbers too? Why can't you just take a real number counterexample and use it for complex numbers?

I guess the answer is that a polynomial map whose Jacobian determinant is constant non-zero over the reals may not also be constant non-zero over all the complex numbers.

New comment by dash2 in "JetZero"

dash2 — Mon, 27 Jul 2026 04:55:44 +0000

I think you've misunderstood my comment. I understand that people can do things together. My point was that the best thing to do together is to solve our problems at lowest total cost, and that choosing a high-cost solution can be individual "virtue signalling".

New comment by dash2 in "JetZero"

dash2 — Sun, 26 Jul 2026 09:07:25 +0000

I hope I'm not overinterpreting, but I get the sense that you think that's morally better than simply avoiding contrails. If so then this is surely a mistake! While an individual who imposes heavy costs on himself to do something socially valuable is certainly demonstrating commitment to the cause, the best solution for society is the one with the lowest costs.

New comment by dash2 in "Show HN: Brolly, a plain-text weather forecast site"

dash2 — Sun, 26 Jul 2026 07:30:58 +0000

Searching for and clicking on Sheringham gives a 400 error.

New comment by dash2 in "The Silurian Hypothesis (2020)"

dash2 — Sat, 25 Jul 2026 15:10:41 +0000

I can't speak to dinosaur people, but the population genetics of our own DNA reveals that there were never many humans on the planet till recently, so that rules out any very large human civilization. (Unless they genetically altered themselves to hide the evidence! Ahaaa...)

New comment by dash2 in "ARC-AGI Leaderboard"

dash2 — Sat, 25 Jul 2026 15:06:13 +0000

You're welcome to speak up, but saying models haven't meaningfully improved in three years, when the most recent models are solving top-level frontier maths problems, is indeed going to strike most people as weird. I still have no idea what you mean.

LLMs can hide text in other text of the same length

dash2 — Fri, 24 Jul 2026 14:53:02 +0000

Article URL: https://arxiv.org/abs/2510.20075

Comments URL: https://news.ycombinator.com/item?id=49036583

Points: 5

# Comments: 0

New comment by dash2 in "Buz – A fork of Bun using modern Zig, with sub-1s incremental builds"

dash2 — Fri, 24 Jul 2026 13:04:51 +0000

I'd also like to see a (kind of) opposite approach: a fork of zig using only AI contributions. This is more as conceptual art or an experiment, rather than because I strongly support AI. But it would be fun to see how the two projects evolved....

New comment by dash2 in "Euro Banknotes Design Proposals"

dash2 — Fri, 24 Jul 2026 09:19:39 +0000

Seems like the major step forward is that the buildings on some designs are real buildings, not just imaginary ones. On the other hand, it's a shame to just use EU buildings! Those may be fine, but they are hardly the most inspiring parts of European built cultural heritage. No Chartres, Parthenon, Versailles or Bauhaus?

New comment by dash2 in "Police removed prominent scientists from American Diabetes Association meeting"

dash2 — Fri, 24 Jul 2026 08:16:53 +0000

I've had a checkup with the NHS once or maybe twice in my life, and it was not with a doctor. (I'm not saying that it should be, maybe a nurse and a blood pressure check is all you need?)

New comment by dash2 in "Police removed prominent scientists from American Diabetes Association meeting"

dash2 — Fri, 24 Jul 2026 05:02:19 +0000

> In places with universal healthcare, you go to the doctor

NHS experience: you might have to wait two weeks to see the doctor, and it will be a ten minute appointment. (Which in this case might be fine.) If you can use a nurse things tend to be much easier.

New comment by dash2 in "Quality non-fiction books are the antithesis of AI slop"

dash2 — Thu, 23 Jul 2026 08:38:22 +0000

I would really like this, but using academic reading lists instead of prizes. What are the important historians of (e.g.) medieval Italy?

New comment by dash2 in "Terence Tao's ChatGPT conversation about the Jacobian Conjecture counterexample"

dash2 — Thu, 23 Jul 2026 08:30:22 +0000

Yeah, as a scientist I can read most Science or Nature articles in most fields, and get a rough idea. Not every piece of jargon, but I can understand roughly what they did, read the graphs and figure out the result. A friend sent me his maths PhD and I did not understand a single sentence.

New comment by dash2 in "The Growing Compute Shortage"

dash2 — Tue, 21 Jul 2026 12:54:59 +0000

Compute would have to be very expensive not to be cheaper than a junior lawyer.