Hacker News Reader

ttctciyfon Show HN: Kinkora – A creative playground for experimenting with video models

Also unfortunately reminiscent of https://en.wikipedia.org/wiki/Kincora_Boys%27_Home

ttctciyfon What will enter the public domain in 2026?

> Why though? Do we really need that many more commercial attempts at Star Wars and Harry Potter?

This kind of baby and bathwater argument could as well be used to ban writing altogether!

ttctciyfon Be Worried

At risk of Godwinisation, there's a very obvious example.

ttctciyfon Ishkur's Guide to Electronic Music

> you nonce. Who was it that said, "every accusation is confession in disguise."

Hmm.

ttctciyfon The death of east London's most radical bookshop

Thought it was going to be about Freedom Bookshop in Angel Alley, Whitechapel, but thankfully that seems to be alive and well[0]!

Being "the oldest and largest specialist anarchist bookshop in Britain" without succumbing to internecine bickering and factionalisation is pretty darn radical, I would have thought.

0: https://freedompress.org.uk/about-freedom/

ttctciyfon Why language models hallucinate

"Hallucination" is a euphemism at best, and the implication it carries that LLMs correctly perceive (meaning) when they are not hallucinating is fallacious and disinforming.

The reification of counterfactual outputs which are otherwise indistinguishable from the remainder of LLM production etiologically is a better candidate for the label "hallucination" IMO.

ttctciyfon AI note takers are flooding Zoom calls as workers opt to skip meetings

Then they should host the meetings on irc! (not completely kidding.)

Or maybe there's an opportunity for someone to create an "interop environment" where AIs representing diverse persons' agendas within an organisation can come together and agree on outcomes and goals on behalf of their sponsors (maybe not completely serious, but ...)

Yeah, I know TFA is just notetaking but that shouldn't stop us from thinking ahead, right? The possibilities for cost-cutting in middle management are spectacular!

ttctciyfon 'Explosive increase' in ticks that cause meat allergy due to climate change

Conspiracists were already on to this in 2023[0]. It's such ripe pickings that if the tick explosion pans out a concomitant explosion in info pollution seems inevitable.

0: https://americanfaith.com/bill-gates-funded-research-into-ge...

ttctciyfon An Analysis of Links from the White House's "Wire" Website

> top 10 domains that the White House Wire links to (by occurrence), from May 8 to June 24, 2025:

    youtube.com (133)
    foxnews.com (72)
    thepostmillennial.com (67)
    foxbusiness.com (66)
    breitbart.com (64)
    x.com (63)
    reuters.com (51)
    truthsocial.com (48)
    nypost.com (47)
    dailywire.com (36)

jfc.

ttctciyfon A man says ChatGPT sparked a 'spiritual awakening.'

The screenshots are quite something.

Maybe one way to think of this is the chatbot and a susceptible user stochastically spiralling each other into a folie à deux[0] together?

0: https://en.wikipedia.org/wiki/Folie_%C3%A0_deux

ttctciyfon Small language models are the future of agentic AI

There really should be some comeback for this type of enshAItification.

We're supposed to think "oh it's an LLM, well, that's ok then"? A question we'll be asking more frequently as time goes on, I suspect.

ttctciyfon Silicon Valley Insider Exposes Cult Like AI Companies [video]

Just the relentless digs at Starmer, and, you know, adherence to the facts, which have a well-known left slant ;-)

ttctciyfon Silicon Valley Insider Exposes Cult Like AI Companies [video]

Despite the clickbaity title (in the original it EVEN has tabloid caps), actually a very interesting long form interview with Karen Hao, an MIT alumnus acquainted with several OpenAI execs since university, whose recently published Empire of AI[0] offers an informed, critical look at the AI explosion in Big Tech.

For example, I think this has the best exposition of OpenAI corporate history and Altman's part in it that I've encountered.

The interviewer, while annoyingly keen to inject his own political slant, at least asks some decent questions and knows how to sit quiet listening to the answers.

0: https://www.penguinrandomhouse.com/books/743569/empire-of-ai...

ttctciyfon Silicon Valley Insider Exposes Cult Like AI Companies [video]

I've given this the (awful, clickbaity) title that it's under on YouTube, but it's actually a very interesting long form interview with Karen Hao, an MIT alumnus acquainted with several OpenAI execs since university, whose recently published Empire of AI[0] offers an informed, critical look at the AI explosion in Big Tech.

The channel is a Leftist UK media outlet with a capital L and the interviewer a self-professed Marxist, but if you can ignore or fight your way through the left spin there's plenty of informative and well presented narrative details about the history of OpenAI, its ambitions and methods and those of its CEO.

0: https://www.penguinrandomhouse.com/books/743569/empire-of-ai...

Title would possibly be better as Aaron Bastani Meets Karen Hao, but there you are.

ttctciyfon Why Are Subway Tunnels Tiled? The Surprising Reason Behind Those Glossy Walls [video]

I've no inclination to click through to the video, but there are undoubtedly many youtube channels that use AI generated scripts spoken by AI generated voices in order to try and make a quick buck, following the advice of many "make-money-fast" videos on youtube, a substantial portion of which, no doubt, are themselves read by AI voices from AI generated scripts.

Whenever I encounter one in my recommends I use the "do not recommend from this channel" option, but I'm clearly fighting a losing battle.

Just one of the many reasons I think AI generated content should be required by law to be marked as such, and passing it off as human produced should be proscribed. When we drown in a sea of slop at least it will be clearly labeled.

ttctciyfon Google Gemini is about to control your messages and calls, even if you say no

Genuine question: does your bank not have a website you can use for that? (Wondering how your bank enables online for desktop/laptop users and/or folks without their app.)

(ETA:) My banks probably have apps, but I only use them via their websites, even on my phone (in a desperate pinch).

ttctciyfon Google Gemini is about to control your messages and calls, even if you say no

Can we say "enshAItification"?

ttctciyfon Fedora Linux devs discuss dropping 32-bit packages – bad news for Steam gamers

Another solution is Conty[0] which is a download-and-run containerised Steam based on Arch; I use it to run games on Slackware without multilibs, and it's worked flawlessly so far.

However, it seems Arch are also dropping multilibs as a dependency of Wine, and moving to WoW64[1], with "reduced performance for 32-bit applications that use OpenGL directly".

What this implies for Steam on Arch (and hence for Conty) I'm not sure, though as of May some Proton versions have a PROTON_USE_WOW64 env var according to [2], so maybe multilibs can already be avoided running Steam natively anyhow.

0: https://github.com/Kron4ek/Conty

1: https://archlinux.org/news/transition-to-the-new-wow64-wine-...

2: https://github.com/ValveSoftware/Proton/issues/6889 (comment from artemyto on May 3)

ttctciyfon Smartphones: Parts of Our Minds? Or Parasites?

> Humans being manipulated is not, in and of itself, a hint that things are going wrong.

Well, if this is meant to counter the abstract's claim that:

> Modern smartphones are designed to manipulate the attention and behaviour of users in ways that further the interests of the corporations that built them. In this they are importantly different from resources typically associated with the extended mind—such as notebooks, Scrabble racks and maps—which are not designed to manipulate or exploit users

I think it falls a little short of refuting a claim that in this particular case things are going somewhat wrong.

ttctciyfon Research suggests Big Bang may have taken place inside a black hole

What about "bit"? Doesn't that come before "it", Mr Wheeler? :)

ttctciyfon Introducing stronger dependencies on systemd

I admit I'm an edge case, but last year I installed Slackware on my main desktop specifically to escape the clutches of systemd, and by god it feels great to have a working /etc/rc.d again!

I do use containerised solutions for a few things (e.g. conty[0] for Steam, and a couple of rocm containers for messing with ML things).

KDE is the default desktop and does everything I need. I haven't used gnome for a decade anyhow.

0: https://github.com/Kron4ek/Conty/

ttctciyfon Big Bang theory is wrong, claim scientists

Press release the Telegraph story is based on:

https://www.port.ac.uk/news-events-and-blogs/news/new-theory...

Arxiv version of the paper:

https://arxiv.org/abs/2505.23877v1

ttctciyfon The Spying Scandal Rocking the World of HR Software

https://archive.ph/rBAKV

ttctciyfon Despite Rising Concerns, 95% of Organizations Lack a Quantum Computing Roadmap

Also lacking AGI roadmaps and fusion power roadmaps, I bet. Slackers!

ttctciyfon What Happens When People Don't Understand How AI Works

> Demis Hassabis, the CEO of Google’s DeepMind, said the goal is to create “models that are able to understand the world around us.”

> These statements betray a conceptual error: Large language models do not, cannot, and will not “understand” anything at all.

While I find lots to agree with in TFA, this pedantry over "understanding" irritates.

While the "understanding" of LLMs is (I believe) a very different affair from human understanding, I wonder if the author can find a better word to succinctly convey what Hassabis is saying?

The problem here is not word choice but failure to recognise that the meaning of a word has changed when the context of its use changes. If I say "it thinks the db is down because it doesn't know the cache is out of date", no-one (or at least very very few!) would take issue with "thinks" and "knows" on the basis that a simple computer program doesn't have human-level knowledge and thought. The words serve to convey meaning, and the context is clear.

If the author doesn't recognise this typical feature of language, extending the meaning of existing words when they are used in new contexts, the "conceptual error" is on them.

ttctciyfon Musk says SpaceX will retire Dragon spacecraft amid bitter Trump dispute

... and now he doesn't

> Elon Musk said on Thursday that his rocket company SpaceX will not decommission its Dragon spacecraft just hours after stating that the company would begin decommissioning

- https://www.reuters.com/science/spacex-will-not-decommission...

"Trump/Musk/J.Random.Blowhard says XYZ" stories are highly ignorable IMO. Reporters should concentrate on what they actually do.

ttctciyfon Firefox now allows you to add custom search engine manually by default

> filter out AI generated sites in the web results

If only!

ttctciyfon Firefox now allows you to add custom search engine manually by default

> is the AI avoidance thing very effective?

Yes, it's just a way of going directly to google's "web" search instead of "all" so you don't get the AI summary and half a screen of assorted guff before the search results themselves. In that sense it's 100% effective.

ttctciyfon Firefox now allows you to add custom search engine manually by default

I see tbs=li:1 activates "verbatim". Duly applied, with thanks :)

ttctciyfon Firefox now allows you to add custom search engine manually by default

> This has always been possible.

No. Starting 2-3 years you had to go through about:config to do this, such as detailed at

https://www.reddit.com/r/FirefoxCSS/comments/qqkw5x/enabling...

and

https://connect.mozilla.org/t5/discussions/custom-search-eng...

etc.

Heaven knows what they were thinking when they made this option only available by the user adding an undocumented preference!

I needed to do this on desktop linux a few months ago to use the AI-avoiding &udm=14 google tweak[0]

0: https://news.ycombinator.com/item?id=43002403