frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Today, we hacked what we'd set out to hack: the iPhone

https://blog.calif.io/p/iphone
207•pavel_lishin•50m ago•97 comments

Kimi K3.1 Pro and K3.1 Pro Highspeed

https://www.kimi.com/code/docs/en/kimi-code/models.html
42•Jowsey•13m ago•3 comments

Local Wikipedia: Wikipedia, but only for things a half hour walk from you

https://edjefferson.com/localwiki/
940•sohkamyung•18h ago•181 comments

The Real Kevin Bacon

https://pudding.cool/2026/10/kevin-bacon/
33•xnx•3d ago•5 comments

Docker is weird

https://codeandcake.dev/posts/2026-10-22-docker-is-weird
754•psxuaw•17h ago•303 comments

Total Historian

https://www.theideasletter.org/essay/total-historian/
5•benbreen•48m ago•0 comments

The Missed Opportunities of Sonic 06

https://isaacmarovitz.com/posts/sonic-06/
3•ibobev•33m ago•0 comments

The alien space of science: Coherent but cognitively unavailable directions

https://arxiv.org/abs/2603.01092
169•martincsweiss•1d ago•60 comments

Job loss and cortisone mania

https://tilschuenemann.de/job-loss-cortisone-mania
100•speckx•6h ago•40 comments

The foolishness of "natural language programming" (1978)

https://www.cs.utexas.edu/~EWD/transcriptions/EWD06xx/EWD667.html
312•xiaoyu2006•16h ago•211 comments

Show HN: OnKey – I'm a bad singer, so I built an app to show where I'm off-key

https://github.com/dsmurrell/onkey
155•dsmurrell•1d ago•69 comments

Every AI model has the same "controversial" take on humans

https://www.modelcompare.ai/blog/ai-models-controversial-take-on-humans
592•rakannimer•1d ago•562 comments

Ask HN: If the models are so good, why doesn't OpenAI just start 100 companies?

408•ex-aws-dude•20h ago•441 comments

Gemma-4-26B-A4B with 39-69% less thinking at same accuracy

https://bottlecapai.com/post/thinkingcap-gemma-4-26b/
18•armcat•2h ago•0 comments

OpenSSL 4.1.0

https://github.com/openssl/openssl/releases/tag/openssl-4.1.0
157•petecooper•4h ago•45 comments

How does Fable 5.1 hide messages in images?

https://www.lesswrong.com/posts/WHEQps5PK8LrrcTAb/how-does-fable-5-1-hide-messages-in-images
177•gmays•19h ago•57 comments

OpenShell – A safe, private runtime for fleets of autonomous AI agents

https://github.com/NVIDIA/OpenShell
475•niemandhier•23h ago•205 comments

Pentagon grants DHS permission to test heat ray weapon for use at border

https://www.cnn.com/2026/10/21/politics/heat-ray-weapon-dhs-pentagon
50•bushwart•2h ago•51 comments

LLMs are making formal software verification feasible at a much larger scale

https://www.anthropic.com/research/formal-verification
412•johnbender•18h ago•252 comments

An alignment assessment of the UK AISI cybertesting incident

https://www.anthropic.com/research/alignment-assessment-ukaisi
4•olwmc•2h ago•1 comments

Run full iOS 16.5 firmware on Android Devices

https://github.com/oosexgang/hackios-xnu-kernel
497•aghuang•1d ago•181 comments

eMudhra ignored Mozilla Root investigation deadline; blames social media

https://bugzilla.mozilla.org/show_bug.cgi?id=2083449
158•IcePenguin•2h ago•90 comments

WorldTUI: I trained a 360M model to pretend to be a terminal

https://simedw.com/2026/10/21/world-model-terminal/
593•alphax314•1d ago•180 comments

This I Believe (1952)

https://www.awakin.org/v2/read/view.php?tid=468
139•ctoth•14h ago•62 comments

Getting ready to operate GNU Taler in Germany

https://www.taler.net/en/news/2026-13.html
345•dewey•1d ago•157 comments

Show HN: Making a smart TV dumb and then smart

https://github.com/anubhavmalhotra/r2d2-tv
73•animalhot•17h ago•44 comments

The HTML-in-Canvas API origin trial

https://developer.chrome.com/blog/html-in-canvas-origin-trial
111•andrewstuart•21h ago•49 comments

Quad9 to exit France

https://quad9.net/news/blog/au-revoir-france/
1322•farrokhi•1d ago•456 comments

Remove pgrust: the entry is cheating (per-query kernels and result caching)

https://github.com/ClickHouse/ClickBench/pull/2165
167•f311a•9h ago•105 comments

Jank – The native Clojure dialect with seamless C++ interop

https://jank-lang.org/
158•giancarlostoro•17h ago•31 comments
Open in hackernews

Armbian Updates: OMV support, boot improvents, Rockchip optimizations

https://www.armbian.com/newsflash/armbian-updates-nas-support-lands-boot-systems-improve-and-rockchip-optimizations-arrive/
74•transpute•1y ago

Comments

proxysna•1y ago
Armbian is an exceptional project, even if the support might be uneven in some places, being able to roll out the same OS across almost every SBC i have is an absolute game changer. If there is support, Armbian is worth trying 100% of the time.

Edit: Also if you don't like/want Ubuntu/Debian their build documentation is pretty great.

dima55•1y ago
Their website doesn't answer the obvious question: what is it, and how is it different from vanilla debian? Do you know?
qwertox•1y ago
Vanilla Debian will not run on your nice and shiny Radxa Rocks 5B or Banana Pi whatever.
dima55•1y ago
Why not? What's missing?
qwertox•1y ago
Different boot process, U-Boot needs to be compiled for the exact board, drivers for the specialized components are needed, DTB (on ARM systems, the kernel doesn't probe hardware the same way a PC does) and other reasons.
RetroTechie•1y ago
> Different boot process, U-Boot needs to be compiled for the exact board

Why? That sounds dumb. And (assuming you're correct), how does Armbian deal with that / get around it?

ajb•1y ago
It's basically the same in the x86 world : your bios is customised to the board

The sad part is that on ARM the kernel is usually also custom compiled for the board. So what happens is that Armbian ship a different image for each board.

If you go and look in https://github.com/torvalds/linux/tree/master/arch/arm you see a zillion "mach-xxx" directories for different SoC architectures, even if they all use Arm.

Device-tree is a partial solution, but no-one seems to have an incentive to finish the job and let a single image run on any (sufficiently recent) arm board. It's difficult for the community to fix because most people have only their own board. Someone would need to pay for a CI rig with every board, and some kernel devs to do the work of building a single kernel to run across everything. (I think that's originally what Linaro was for - not sure why they didn't finish the job)

qwertox•1y ago
Right, the x86 BIOS/UEFI is baked into the motherboard firmware and handles early hardware init in a mostly standardized way. But with ARM boards, there's no universal firmware, it usually needs to be part of the image you download for that specific board.
FlyingSnake•1y ago
How does Armbian compare to DietPi?

FWIW: I’m running dietPi on my OG Pi Zero W and it doesn’t even hit 30% resource usage.

apple4ever•1y ago
Completely agree. I use it on my old PINE64 and it keeps on ticking.
chris37879•1y ago
I just stumbled across armbian recently and I must say I really like it.

I wanted to use UEFI, but my orangepi cm5 modules don't seem to have the SPI chip needed to store the UEFI there, so I'd have to load it on a partition and lose out on some features like persisting variables across boot.

The arm ecosystem really needs to settle on some sort of universal boot loader / firmware layer and stop just hacking up the linux kernel and not contributing back to it.

Nexxxeh•1y ago
I'm not an Arm dev and am just a consumer so I may be misunderstanding, but isn't Arm SystemReady pretty much the thing that's intended to solve the problem you're talking about (among others)?

https://developer.arm.com/documentation/107981/0302/SystemRe...

robotnikman•1y ago
It is, but it seems like only servers are adopting it at the moment. Or high end ARM workstations. I can't think of any consumer devices or SBC's off the top of my head that support it.
moondev•1y ago
Raspberry PI and Nvidia AGX

https://github.com/pftf/RPi4

moondev•1y ago
https://developer.arm.com/Architectures/Unified%20Extensible...
yjftsjthsd-h•1y ago
> Why? That sounds dumb.

Good, you understand the situation perfectly.

> And (assuming you're correct), how does Armbian deal with that / get around it?

You'll notice that if you try to download it from https://www.armbian.com/download/ , nearly every board has a different download image; this is because every one of those images embeds its own boot chain. There are efforts (in some projects, I'm not aware of armbian doing this) to build some amount of early bootloader per-board (often uboot), and just make the install steps something like "install this per-board thing, then install the real OS using a standard image" but that's less common and doesn't work super well when that initial bootloader has to go on the same storage device as the main OS.

dima55•1y ago
I believe that's common on ARM devices. But "vanilla debian" generally refers to userspace, and that should just work. Is this "armbian" thing quite literally "kernel + bootloader + vanilla debian"? The website doesn't say that in any obvious place
puzzlingcaptcha•1y ago
Pretty much, plus their little configuration utility for loading dtb overlays among other things.
pabs3•1y ago
The hard work of upstreaming/mainlining all the hardware support code in the userspace drivers like mesa, the Linux kernel core/drivers, bootloaders like GRUB/u-boot, boot firmware like coreboot/Tianocore/u-boot.