frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

We have a year to fix security everywhere

https://jyn.dev/a-year-to-fix-security/
33•saikatsg•43m ago

Comments

hn_submit•24m ago
Or we could just dump Linux and Windows and switch to a microkernel operating system, which is much more secure.

These endless patching cycles are simply not going to work in the long run. Operating systems get orphaned all the time, especially the ones in cheap Chinese stuff.

thunderfork•22m ago
"throw away all software written before 2026" does technically solve this problem, if you ignore everything else the article is talking about (deployment and continuity of service)
simonw•18m ago
Got any leads on good tutorials on how to use a microkernel operating system on a VPS somewhere to host a website?
simonw•17m ago
I don't think we even have a year. The current batch of LLMs are ferociously good at identifying vulnerabilities.
sho•15m ago
> On September 22, Apple is releasing the M5 Mac Studio with 256 GB of unified memory [..] it will probably [..] enough to write this snippet of code in 3 seconds

The author has obviously never ran an LLM on a mac! In 3 seconds, it will have possibly started to think about maybe scheduling a date to contemplate the planning timeline for processing the second token in your prompt.

simonw•14m ago
The difference is memory bandwidth. The M5 Ultra that's coming out on 22nd September can do 1,200GB/s. The M5 Max you can buy today only has 614GB/s.
sho•1m ago
So, that gets us to about where nVidia was with Ampere in 2020. Let's hope the M7 catches us up with at least Hopper.
akmarinov•5m ago
The author put in the numbers, but maybe you didn’t read them.

45 t/s a second is perfectly respectable especially with no limits and 24/7 uptime with very little power draw on the Studio.

Luna is at around 100 t/s for comparison, but it’s a worse model than 5.3 Flash

pmlnr•12m ago
Here's an idea: as a first step, simplify everything, and make sure you're aware how your stack works, and what it imports.

As an example: WordPress is a horrible thing, but the core has been through so much, that it's suprisingly secure. Then plugins and themes come, and whoosh, the security is gone.

We need a new KISS: keep it simple, stupid, secure.

mirashii•8m ago
The "surprisingly secure" WordPress just had a unauthenticated RCE earlier this year. Just simplifying isn't going to be enough.

https://nvd.nist.gov/vuln/detail/cve-2026-63030

dbdr•7m ago
> Invest in formal verification, fuzzing and property testing, and memory-safe languages. LLMs are good at writing Lean and fuzz tests. I don't care whether you use Go or Rust but for the love of god please don't use C or C++ for new code.

How accepted is this thinking in your respective domains?

petesergeant•6m ago
Mmm, a world where a defender-LLM is essentially required is great news for people selling inference.
LoganDark•2m ago
It's just the same advice as ever: be extremely, exceedingly careful in what you expose to any network. When I set up machines for production, they don't respond to pings and they don't even have an SSH port open without knocking. There are also ways to eschew the need for an SSH port entirely.

People who never took that seriously will never take this seriously either, and that's their loss. (And loss of the commons, unfortunately.)

There's just also new advice: you can't afford to expose an unsecured system to the internet even for a moment. Think of those IPv4 address space scanners, except this time any one of them could be capable of developing individualized attacks in mere minutes. They don't sleep, they don't take breaks.

uecker•1m ago
And don't use software written in Rust, it has too much supply chain risks, which is far more important than memory safety in my opinion.

An AI that talks like a friend, jumps in, interrupts and keeps the convo going

https://www.echohive.ai/deep-talk-buddy
1•echohive42•44s ago•0 comments

Electrostatic Cathode Ray Tube Project 1 (2014)

https://labguysworld.com/Project_ESCRTs_001.htm
1•userbinator•6m ago•0 comments

Multi-Agents LLM Financial Trading Framework

https://github.com/TauricResearch/TradingAgents
1•fittingopposite•11m ago•0 comments

I scraped 10,555 ivf rounds rounds from forum posts around the world

https://eggs.help
1•grapefruitjuice•11m ago•0 comments

Clean Web-to-Markdown: Fast HTML Extraction for LLMs and RAG

https://markdown.usemy.cloud/
1•petoz•17m ago•0 comments

What may happen in the next hundred years?(1900)

https://earlyradiohistory.us/1900next.htm
2•o4c•20m ago•1 comments

My Feed, My Way

https://www.pm.gov.au/media/my-feed-my-way
8•dotcoma•21m ago•0 comments

KDE's new Photos app offers a sleek alternative to Gwenview

https://www.omgubuntu.co.uk/2026/09/kde-photos-app-replace-gwenview
2•fork-bomber•24m ago•0 comments

Mistral raises €3B to make sovereign, open-weight AI the technology frontier

https://mistral.ai/news/mistral-makes-sovereign-open-weight-ai-to-frontier/
2•kuberwastaken•25m ago•0 comments

The 100x Engineer Certification

https://100x.iamdevloper.io/
1•Curiositry•26m ago•0 comments

Ask HN: Best Platforms to Orchestrate Agents

1•venkat971•28m ago•0 comments

How One Connection Kills a Database

https://planetscale.com/blog/debugging-live-database-connections
1•vzhou842•32m ago•0 comments

Show HN: TripleFantasy Reconstructed Source Code

https://github.com/Sadpainy/TripleFantasy
1•CMDDestory•36m ago•0 comments

Timeline of Mathematics

https://mathigon.org/timeline
2•o4c•38m ago•0 comments

Autodafe VES: a free vintage emulator studio plugin based on MAME framework

https://synthanatomy.com/2026/09/autodafe-ves-a-free-vintage-emulator-studio-plugin.html
1•bane•41m ago•1 comments

We have a year to fix security everywhere

https://jyn.dev/a-year-to-fix-security/
34•saikatsg•43m ago•14 comments

A Modern Proxmox Docker Architecture with Disposable VMs, VirtIO-FS, and ZFS

https://du.nkel.dev/blog/2026-05-16_rootless_docker_virtiofs_proxmox/
1•Helmut10001•51m ago•0 comments

Cognition Has More Than One Shape: From AI to Synthetic Cognition

https://tvvocold.github.io/2026/09/08/cognition-has-more-than-one-shape.html
2•tvvocold•52m ago•0 comments

The VMs Powering Mobile Agents (Instinct, Claude Code)

https://rohanadwankar.github.io/posts/platforms.html
6•RohanAdwankar•55m ago•1 comments

Canada's counter-tariffs take effect on various US goods

https://www.france24.com/en/live-news/20260908-canada-s-counter-tariffs-take-effect-on-various-us...
9•geox•1h ago•0 comments

Bonds – AI messenger where group chats generate their own apps

https://apps.apple.com/us/app/bonds-the-smart-messenger/id6798100604
1•juanchoabdon1•1h ago•0 comments

RabbitMQ vs. Apache Kafka

https://www.rabbitmq.com/docs/compare/kafka
2•vinhnx•1h ago•0 comments

Evidence runtime for autonomous agent task management

https://grainulator.app
3•aid-ninja•1h ago•0 comments

Show HN: Manhattan Safety Atlas

https://manhattan-neighborhood-atlas.pplx.app/
1•tesseractinf•1h ago•0 comments

Shatranj

https://en.wikipedia.org/wiki/Shatranj
2•teleforce•1h ago•0 comments

Secret screening of Nathan Fielder's new Elizabeth Holmes documentary

https://www.latimes.com/entertainment-arts/movies/story/2026-09-07/nathan-fielder-elizabeth-holme...
1•aanet•1h ago•2 comments

Australia to opt out of social media algorithms, Albanese confirms

https://www.smh.com.au/politics/federal/australia-to-opt-out-of-social-media-algorithms-albanese-...
5•mrkpdl•1h ago•2 comments

Experience is becoming a higher priority for tech hiring

https://www.a16z.news/p/chart-of-the-week-experience-skills
2•deepmem•1h ago•0 comments

GifBook – turn any GIF into a printable flipbook (free)

https://www.mrbray.com/gifbook
1•mrbray•1h ago•1 comments

Uncensored Open-Weight Models: Redistribution as the Persistence Layer

https://arxiv.org/abs/2609.05241
1•sbulaev•1h ago•0 comments