Hacker News: dudu24

New comment by dudu24 in "I won't download your app. The web version is a-ok"

dudu24 — Mon, 06 Apr 2026 16:55:11 +0000

old.reddit.com

New comment by dudu24 in "RFC 454545 – Human Em Dash Standard"

dudu24 — Tue, 10 Mar 2026 18:38:57 +0000

Hot take: I think the em-dash is just lazy punctuation that can be replaced by the more nuanced pauses, i.e. the comma, semicolon, and colon. I think its popularity stems from people being confused on how to use a semicolon.

New comment by dudu24 in "FreeBSD doesn't have Wi-Fi driver for my old MacBook, so AI built one for me"

dudu24 — Tue, 24 Feb 2026 17:56:14 +0000

My Nintendo Switch 2 Pro Controller didn't work with my Mac, so I had Claude write me a driver. Amazing times we live in. (As long as I still have a job so I can buy controllers in ten years.)

New comment by dudu24 in "LG UltraFine Evo 6K 32-inch Monitor Review"

dudu24 — Wed, 21 Jan 2026 15:57:43 +0000

If this had a higher refresh rate I'd be making unwise financial decisions today

New comment by dudu24 in "On the Existence, Impact, and Origin of Hallucination-Associated Neurons in LLMs"

dudu24 — Mon, 22 Dec 2025 15:46:45 +0000

I'm tired of this pseudointellectual reductionist response. It's not "literally by accident" when they're trained to do something, as if we are not also machines that generate next actions based on learned neural weights and abstract (embedded) representations. Your issue is with semantics rather than content.

Obviously "hallucinate" and "lie" are metaphors. Get over it. These are still emergent structures that we have a lot to learn from by studying. But I suppose any attempt by researchers to do so should be disregarded because Person On The Internet has watched the 3blue1brown series on Neural Nets and knows better. We know the basic laws of physics, but spend lifetimes studying their emergent behaviors. This is really no different.

New comment by dudu24 in "We remain alive also in a dead internet"

dudu24 — Fri, 21 Nov 2025 21:47:42 +0000

This is a non-response.

New comment by dudu24 in "We remain alive also in a dead internet"

dudu24 — Fri, 21 Nov 2025 21:46:04 +0000

I'm also losing my ability to tolerate prose without headings, but I think that's symptomatic of this bigger issue.

New comment by dudu24 in "Google Antigravity"

dudu24 — Tue, 18 Nov 2025 19:49:48 +0000

I cannot stand webpages that hijack scrolling like that.

New comment by dudu24 in "BERT is just a single text diffusion step"

dudu24 — Mon, 20 Oct 2025 16:20:36 +0000

That is not contrary to token-at-a-time approach.

New comment by dudu24 in "Who Invented Backpropagation?"

dudu24 — Mon, 18 Aug 2025 16:39:27 +0000

It's just an application of the chain rule. It's not interesting to ask who invented it.

New comment by dudu24 in "Symmetry between up and down quarks is more broken than expected"

dudu24 — Mon, 31 Mar 2025 19:24:56 +0000

This misses the point of isospin. Isospin is an approximate SU(2) symmetry due to the fact that the up and down quarks (the "light" quarks) have very similar masses compared to the rest of the quarks, so they can be approximated as two different eigenstates of the same particle. It's mathematically identical to the SU(2) symmetry of a spin-half particle. The reason it doesn't include the other quarks is because they are so much more massive.

New comment by dudu24 in "Avoiding outrage fatigue while staying informed"

dudu24 — Wed, 05 Feb 2025 18:06:51 +0000

For better or worse, news flows through social media, so this approach basically amounts to ignoring all the bad stuff going on. If you read HN, chances are you can probably safely get through the next four years doing this. But as the saying goes, "first they came for the communists..."

New comment by dudu24 in "Do AI detectors work? Students face false cheating accusations"

dudu24 — Mon, 21 Oct 2024 20:57:14 +0000

> grand vector space

what.

New comment by dudu24 in "Research in psychology: are we learning anything?"

dudu24 — Thu, 10 Oct 2024 19:31:18 +0000

What a stereotypical hackernews comment, wow.

> I, a completely unqualified internet commenter,

Just leave it there.

New comment by dudu24 in "Nobel Prize in Physics awarded to John Hopfield and Geoffrey Hinton [pdf]"

dudu24 — Tue, 08 Oct 2024 16:45:45 +0000

I am a physicist working in ML and this is an absolutely bonkers comment lol

New comment by dudu24 in "BorgBackup 2.0 supports Rclone – over 70 cloud providers in addition to SSH"

dudu24 — Tue, 01 Oct 2024 16:17:06 +0000

My problem is I learn some tool like this, set it, and then indeed forget it. Then I avoid testing my backups because of the work it takes to un-forget it. Because of this, I'm leaning more and more towards rsync or tools that have GUI frontends.

New comment by dudu24 in "The historical distinction between coder and programmer"

dudu24 — Tue, 27 Aug 2024 18:03:53 +0000

> the programmer he calls what he does "engineering" which he deems equivalent to what rocket engineers do when they "engineer" a rocket that flies into space, but instead of a rocket it's some website.

Bingo. The (very American) dilution of the title "engineer" has been sad to see.

New comment by dudu24 in "Group of 17 London secondary schools join up to go smartphone-free"

dudu24 — Mon, 17 Jun 2024 16:42:50 +0000

Haidt is a reactionary who makes grandiose conjectures about the Kids These Days with little real scientific evidence to back them up. He's the same guy who threw a fit over safe spaces in colleges and made them out to be a way bigger deal than the were/are.

New comment by dudu24 in "Show HN: I built a game to help you learn neural network architectures"

dudu24 — Tue, 21 May 2024 19:04:19 +0000

I have always found neural network diagrams like the RNN one here to be very vague and even slightly misleading. What does it mean that h_t loops onto itself? While I know that it means "take as input h_{t-1} also", the diagram itself does not illustrate the concept to the primary person looking at such a diagram, i.e. someone wanting to learn about the architecture.