Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Hacking OpenAI(hacktron.ai ↗)
    52comments
  2. Waymo in Singapore(waymo.com ↗)
    27comments
  3. Astra for Law(openai.com ↗)
    450comments
  4. Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint(prismml.com ↗)
    109comments
  5. Pre-Greek: The lost language hidden within Ancient Greek(linguisticdiscovery.com ↗)
    5comments
  6. Bend – A language that blocks AI mistakes via proof, on CPU and GPU(bend-lang.com ↗)
    191comments
  7. Hister: A private search engine for the pages you visit and the files you keep(github.com/asciimoo ↗)
    142comments
  8. Qwen 3.8 Omni Flash(qwen.ai ↗)
    27comments
  9. The Scourge of x86 Emulation(fex-emu.com ↗)
    discuss
  10. Wax motor(wikipedia.org ↗)
    59comments
  11. Shapelearn Qwen 3.8 27B (13.1 GB VRAM)(byteshape.com ↗)
    2comments
  12. Jemalloc 5.4.0(github.com/jemalloc ↗)
    discuss
  13. Fujitsu launches made-in-Japan next-generation CPU FUJITSU-MONAKA(global.fujitsu ↗)
    207comments
  14. Apple detectives solved mystery of ancient tree and rewrote the history of fruit(scientificamerican.com ↗)
    2comments
  15. Telstra outage: The night a network decided the year was 2006(netnod.se ↗)
    12comments
  16. Ask A Monk – A digital wilderness for thoughts with no immediate answer(askamonk.online ↗)
    13comments
  17. How to Write with an LLM(sockpuppet.org ↗)
    62comments
  18. Flet 1.0 – Build cross-platform apps in Python(flet.dev ↗)
    41comments
  19. Diplodocus, Long Thought Exclusively American, Turns Up in Spain(sci.news ↗)
    30comments
  20. Speeding up gearhash on ARM64 (2× faster)(sam.dev ↗)
    discuss
  21. Code Scans(devin.ai ↗)
    3comments
  22. The most important product decision is what you don't build(liamnugent.me ↗)
    25comments
  23. How Uber Protects Against Retry Storms(uber.com ↗)
    33comments
  24. CrowdSec Source Code Leak(crowdsec.net ↗)
    43comments
  25. Why I didn’t sign the Fields medallists’ letter(gowers.wordpress.com ↗)
    338comments
  26. How do we prevent mathemathics from devolving into the Medieval Era of secrecy?(mathoverflow.net ↗)
    79comments
  27. Khipu (Quipu) Field Guide(khipufieldguide.com ↗)
    discuss
  28. Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data(arxiv.org ↗)
    38comments
  29. Show HN: Snapdrop: Instantly share files between devices. No setup, no signup(snapdrop.me ↗)
    30comments
  30. I Put Nam A2-Lite Inside an iRig HD X(playtaurus.com ↗)
    6comments

Project CHIP gets a new name and so does the Zigbee Alliance

4 pointsby 5y agostaceyoniot.com
1 comments
5y agoHN ↗

It'd be really nice to not have a bunch of proprietary standards defining the home. This work could really help.

Alas it seems like Thread is a key piece of this stack (for low end devices, high end devices use wifi), and Thread is definitely not at all open access. There are some open source implementations! OpenThread &c! But it's really weird trying to work with, when you don't have any idea how the blasted thing works or what's going on. Very much a side note, but interesting that Thread was chosen, even though the work is being done with the Zigbee Alliance, who used to be Threads competitor! (Thread is UDP/IP based though, so kind of obvious versus the much older mystery protocols that make up Zigbee.) Zigbee Alliance's "dotdot" device profiles seem to have been adopted though for this effort. Zigbee was not any more accessible than Thread though, & in fact, it had considerably less code. Both are based on 802.15.4 wireless protocols.

And alas alas alas, it seems that "you can use it, but you can't know how it works" seems to be how the new CSA / Matter specs are being released too, which just feels meanspirited & shitty to me, on so many levels:

The CSA is also focused on releasing as much as it can as open-source code, which strikes me as somewhat incongruous with the fact that to see or vote on the developing standard, companies had to pay to join the CSA and the Matter working group. However, the code is all available on Github, and anyone will be able to see it and download it for their applications and devices.

I watch video's like Peter Norvig's "As We May Program"[1] about building smarter systems, systems that serve us better. But I'm sick to death of the ego, of the closeness, of the claims that users need more help. What we need is a fair shake, we need a world that we have the possibility to understand, that we can approach on an honest level. We need access to the technology about us, need the ability to see for ourselves, & not be coddled & consumerized endlessly. Yeah, more advanced help is welcome & good, but we can't allow technology to proliferate endlessly unchecked unseen unintelligibly forever like this, can't keep piling complexity on & claiming, oh, things are so complex that we need to build more automation & more control into the technology for your sake. There has to be some advancement from the other side, to make the technology apparent & truthful & learnable & visible. CSA is not helping. Thread is not helping. They are continuing the war against general purpose computing, new frontiers to allow product makers to flood our lives with things that do not respect us & which do not give us a fair shake.

The world doesn't just need products that interoperate. The world needs to be intelligible, legible. Witholding the standards, not letting people know how technology works, the technology we are literally filling our houses with: that is the enemy of understanding & civilization. We need to free ourselves from these intellectual tar pits, need to stop building such walls between "consumers" and makers, and start to have a more collective approach to socialization of technology. I wish IoT would not stand as such a foe of the Enlightenment.

[1] https://www.youtube.com/watch?v=J573tlzlSC4