Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Bend 2 and the Vibe-Coding Trap(liampwll.com ↗)
    60comments
  2. OpenJev(openjev.com ↗)
    136comments
  3. ZCode, the GLM coding agent, silently uploads your Git history(tokenstead.ai ↗)
    34comments
  4. I don't like passkeys(hawksley.dev ↗)
    81comments
  5. Jemalloc 5.4.0(github.com/jemalloc ↗)
    56comments
  6. Subnormal floating-point numbers are expensive on Intel processors(lemire.me ↗)
    7comments
  7. Microsoft exec called AI scraping 'the largest theft of labor in human history'(techcrunch.com ↗)
    311comments
  8. The scourge of x86 emulation(fex-emu.com ↗)
    44comments
  9. Cekura (YC F24) Is Hiring(ycombinator.com ↗)
    discuss
  10. Warren Buffett Steps Down as Berkshire Chairman, Names Son to Replace Him(nytimes.com ↗)
    55comments
  11. Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint(prismml.com ↗)
    154comments
  12. Replacing Pull Requests with Delta(zed.dev ↗)
    26comments
  13. Astra for Law(openai.com ↗)
    622comments
  14. The Shadows Lurking in the Equations – Underwater Islands(gods.art ↗)
    discuss
  15. Bend – A language that blocks AI mistakes via proof, on CPU and GPU(bend-lang.com ↗)
    247comments
  16. Qwen 3.8 Omni Flash(qwen.ai ↗)
    96comments
  17. Hister: A private search engine for the pages you visit and the files you keep(github.com/asciimoo ↗)
    173comments
  18. Wax motor(wikipedia.org ↗)
    78comments
  19. When the fractional part of a float fixes your shader(crocidb.com ↗)
    10comments
  20. Fujitsu launches made-in-Japan next-generation CPU FUJITSU-MONAKA(global.fujitsu ↗)
    238comments
  21. Pre-Greek: The lost language hidden within Ancient Greek(linguisticdiscovery.com ↗)
    51comments
  22. How to Write with an LLM(sockpuppet.org ↗)
    142comments
  23. Dr Julius Neubronner's Miniature Pigeon Camera(publicdomainreview.org ↗)
    discuss
  24. A heap overflow and SSO misconfiguration to compromise OpenAI internal repos(hacktron.ai ↗)
    165comments
  25. Shapelearn Qwen 3.8 27B (13.1 GB VRAM)(byteshape.com ↗)
    19comments
  26. Ask A Monk – A digital wilderness for thoughts with no immediate answer(askamonk.online ↗)
    28comments
  27. Telstra outage: The night a network decided the year was 2006(netnod.se ↗)
    30comments
  28. Flet 1.0 – Build cross-platform apps in Python(flet.dev ↗)
    72comments
  29. Diplodocus, Long Thought Exclusively American, Turns Up in Spain(sci.news ↗)
    48comments
  30. Speeding up gearhash on ARM64(sam.dev ↗)
    discuss

Information Camouflage

32 pointsby 4y agobuttondown.email
12 comments
4y agoHN ↗

I wish people would take this into consideration when naming their company/product/etc. At least drop a vowel or add a silent 'h' or something...

4y agoHN ↗

Really huge companies can just name their product “Disney+” and break Google, and instead of never being discovered instead force the search engines to adapt.

4y agoHN ↗

I think Google+ had already broken Google search in that several way years before Disney+ appeared.

4y agoHN ↗

Like the project named Pulsar. Or the project named Pulsar. Or the other project named Pulsar...

At least it's been a few months since a intermediate data engineer asked a question about Mirror Maker 2 in the subreddit dedicated to discussions of the works of Franz Kafka.

4y agoHN ↗

re: pay someone. I forget the name but in the early 2000s there was a site where you could submit a question and someone would search it for you. They were paid a small amount for each answer. Does anyone remember this?

4y agoHN ↗

No; it wasn't Google or Yahoo answers, it was an independent site. The name still escapes me.

4y agoHN ↗

This is the curse in using unstructured data, and in sites like reddit banning self promotion.

Unstructured data is often always transformed (ETL) and with that method, things get lost in translation. As automation and SEO is twisted to fit expectations for results, you get an index of spaghetti that is damn near impossible to untangle as it grows, because the data sources are also constantly changing. I read an article about how people change the entire name and purpose of Facebook groups after they develop huge amounts of followers (To new and totally unrelated things) And that's how user bases are often grown now on many social apps...

Unforeseen trickery like that reeks havoc on unstructured query resources for very specific things. This is why structured and curated data resources worked with a lot less stress and more accuracy in the past.

No matter how much a concept is thought out, there will always be hackers that gamify it. It is harder to get away with hacking data in a structured data resource because it is easier to identify data anomalies and abnormalities in structured data.

Secondly, by the new wave ideal that self promotion is a bad thing, we're putting second hand information in the spotlight, and giving it the primary regard and credit... Many of the people posting things and responding to related questions for comment posted often have nothing to do with the authoring of the original content... They are also often the first and only ones to reap the financial rewards of ideas and things they didn't create. It's a pretty toxic rule to prevent people from promoting their own ideas online on any credible resource, but somehow it's been pushed as a reasonable practice for the past few years by possibly the same people that post information and stolen or copied ideas second hand. Content and idea theft is surging because of TikTok and other social platforms alone.

These things lead to a world where everything is posted without proper context and credit, and of course it allows people to operate without accountability for deception, harmful words, and for putting the profit in places where it will inspire more ideas from people who have proven track records and give them proper credit.

The world keeps screwing things up just to make a buck.

4y agoHN ↗

Over time you develop a set of words that work around such limitations, finding alternatives that route around the overloading. I treat them as magic incantations when I find them.

Example: "Annotation" -- what you do when you mark up text, used to route around HTML/ Hyper Text Markup Language (which doesn't allow you to annotate anything)