frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

The Power of AI: Mass Collaboration

https://www.cppdepend.com/blog/ai-real-power-mass-collaboration/
1•cppfirst•9s ago•0 comments

Show HN: DeutschDNA – German Tutor Skill for Claude Code and Codex

https://github.com/ahmtsahin/deutsch-dna
1•impala64•50s ago•0 comments

Show HN: Busbar self-hosted execution boundary for Agentic Apps

https://github.com/GetBusbar/busbar
1•mangatgoyal•1m ago•0 comments

Show HN: Whiteboard (YC W26) – an open-source IDE for thoughtful software design

https://github.com/devdotfast/whiteboard
4•sidharthkmenon•1m ago•0 comments

Show HN: Devc – an alternate CLI for launching devcontainers

https://github.com/rameshvarun/devc
1•varunramesh•6m ago•0 comments

Open AI in the Wild: Adoption and Adaptation of Open Models on R/LocalLLaMA

https://dl.acm.org/doi/abs/10.1145/3805689.3812421
1•theanonymousone•7m ago•0 comments

Return-to-office policy in 2026: 8 of 75 top UK employers require 5 office days

https://www.wfhjobs.co.uk/wfh-jobs-research/rto-policy-facts-and-statistics
2•FreddieSO•7m ago•1 comments

Paper2agent Turns Static Papers into Live AI Tools

https://spectrum.ieee.org/paper2agent-ai-agents-research-papers
2•rbanffy•7m ago•0 comments

35 years ago I did it all myself

https://twitter.com/jasonfried/status/2103170272467767753
2•tosh•7m ago•0 comments

For Computer Use, the harness matters as much as the model

https://www.stagehand.dev/evals
7•MiguelG719•9m ago•2 comments

Meta employees don't like to be filmed with there own glasses [video]

https://www.youtube.com/shorts/6Abwd13odL4
2•pixvi•9m ago•0 comments

Show HN: Guardmcp – I scanned the official MCP registry with a config scanner

https://github.com/BerkantACUN/guardmcp
1•melalonsra•9m ago•0 comments

Developer ported Word for Windows 1.1a from 1990 to run natively on Windows 11

https://www.neowin.net/news/developer-ported-word-for-windows-11a-from-1990-to-run-natively-on-wi...
2•bundie•10m ago•0 comments

Does Georgism Work? Five Years Later

https://www.astralcodexten.com/p/does-georgism-work-five-years-later
1•iamnothere•10m ago•0 comments

"Antedisciplinary" Science (2005)

https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.0010006
1•the-mitr•10m ago•0 comments

Why does a Nix store path survive garbage collection?

https://www.labcraft.dev/blog/nix-gc-roots
1•anandsuresh•10m ago•0 comments

Show HN: ThimbleDb, an encrypted lightweight low latency open source database

https://github.com/Jason-Doyle/thimble
2•randomlurker2•11m ago•0 comments

We Have Named Arguments at Home

https://corrode.dev/blog/named-arguments-at-home/
2•dimonomid•12m ago•1 comments

OpenRouter: Server Tools Marketplace

https://openrouter.ai/tools
1•theanonymousone•13m ago•0 comments

Shared Memory for Coding Agents

https://getbelay.vercel.app/
1•doplexlabs•14m ago•0 comments

Lifestreams: A storage model for personal data: ACM SIGMOD Record: Vol 25, No 1

https://dl.acm.org/doi/10.1145/381854.381893
1•rbanffy•15m ago•0 comments

US plan for 90 day diesel export ban could have significant impact on Europe

https://www.politico.eu/article/eu-launches-diplomatic-offensive-to-stop-trumps-diesel-export-ban/
1•ck2•15m ago•1 comments

Open OCI spec for agent sandboxes

https://www.linux.com/featured/docker-commits-to-bringing-the-sandbox-kit-spec-to-the-cncf/
1•pploug•15m ago•0 comments

LLMs can write tool workflows. Will they choose to?

https://github.com/kukushko/agent/blob/main/agent_workflows_exploratory_benchmark.md
1•kukushko•16m ago•0 comments

The International AI Race May Not Be Won by the Smartest AI

https://phroneses.com/articles/leadership/notes/the-ai-race.html
1•speckx•18m ago•0 comments

S.F. Democratic Party stands behind Flock surveillance cameras in vote

https://missionlocal.org/2026/09/san-francisco-democratic-party-flock-cameras/
17•mikhael•18m ago•12 comments

Gemini 3.8 Live with Live Avatar

https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-live-with-live...
2•xnx•19m ago•0 comments

One day of AI reading: 5.7M crawler reads of a 60M-company index, by engine

https://engagemii.com/research/one-day-of-ai-reading
1•gp_engagemii•19m ago•0 comments

The Download: a bid to scrap the virtual wall and AI hits Climate Week

https://www.technologyreview.com/2026/09/24/1145064/the-download-bid-scrap-virtual-wall-ai-climat...
1•joozio•21m ago•0 comments

Pinocchio, make copilot agents real by giving them memory

https://github.com/chiedo/pinocchio
2•chiedo•21m ago•0 comments
Open in hackernews

GitHub has not removed malicious imitation software after 3 weeks

https://successfulsoftware.net/2026/09/24/github-has-not-removed-malicious-imitation-software-after-3-weeks/
98•hermitcrab•1h ago

Comments

someonebaggy•53m ago
If it's your software send a DMCA. They have a legally required timeframe to process those. If it's open source, however, then you don't have any valid DMCA claim.
93po•48m ago
Code can be open source while the name and logos are copyrighted and still enforceable via DMCA
hnlmorg•41m ago
That’s not how open source works.

Open source code is still copyrighted. What the license defines is rights that people have in distributing that code. If an unofficial repository is using open source code to ship malware, and the license that software had didn’t allow that, then the unofficial repository is still breaking copyright law despite the code being open source.

NewJazz•28m ago
Also open source license doesn't grant use of trademarks, but I'm not sure that means DMCA applies.
joshuat•52m ago
Not exactly the same, but I've noticed a pretty sizable uptick in the number of spam/scam PR comments being left on GitHub (and a longer delay before they're removed after report).

Not the worst thing in the world, they're easy to spot, but I'd like to see GitHub invest more time in protecting their users from falling victim to these bad actors.

kg•51m ago
In the future just issue a DMCA takedown right away for cases like this, IMO.
hermitcrab•48m ago
Author of the post here. Github finally took the offending page down approximately 10 minutes after the post appeared on the front page of HN. Total coincidence. I'm sure!

Moral of the story. If you want even the most basic level of support from Github, you need to get on the front page of HN first.

And it seems they are able to do things very quickly, when they want to. Bastards.

ajmurmann•47m ago
It might not be a willingness issue as much as a bandwidth issue.
cyanydeez•45m ago
unwilling to provide proper support?

Just seems like a silly rational response to the same problem.

locknitpicker•44m ago
Yes,evidently bandwidth from HN unblocks takedown requests of malicious content.
ajmurmann•41m ago
Prioritization and escalation exists in most companies. I guarantee you that once an issue hits the HN front page, even engineers who might have totally different talks will get involved. (Never worked at GH or have talked to anyone there in years but this is how everything works pretty much everywhere)
toomuchtodo•37m ago
The public shaming will continue until the internal incentives improve. Make sure to drop that HN thread link into the internal task tracker y'all. Don't forget to report to journalists if the severity warrants it (Brian Krebs, 404media, etc).

"Show me the incentive and I'll show you the outcome."

jay73763•43m ago
why would anyone host commercial binary software on github or any other third party domain?
hermitcrab•37m ago
Pirates and crackers generally don't host stuff on their own domains. They don't want to pay for the bandwidth and they don't want to be traced.
sgskinner•5m ago
I think they’re alluding to OP not hosting their own downloads.
Havoc•28m ago
They’re presumably too busy with keeping availability above nine sixes
iAMkenough•42m ago
Good thing HN provided them some bandwidth to do their jobs.
post-it•40m ago
Bandwidth can be bought with money, of which Microsoft made an extra $133.7 billion this year.
NobodyNada•38m ago
Sounds like they can afford elite customer support.
ajmurmann•37m ago
We know that coding agents have been pushing GH to its limits. Scaling is hard - especially staff. Maybe they aren't trying to scale support but I think it's reasonable to give them the benefit of doubt here, given what we know publicly
iAMkenough•31m ago
That’s a self-inflicted issue they should have properly planned for.
cj•31m ago
You’re saying this is a problem money can’t solve?

There’s no need for benefit of the doubt when it comes to the level of support provided by tech companies.

Bad support by tech companies is a conscious profit-preserving choice.

cwillu•17m ago
Microsoft is not some plucky upstart company with 12 employees and an unexpectedly popular product. We do not, in fact, need to give them the benefit of the doubt here.
kstrauser•20m ago
Never thought I'd see the day when Microsoft is elite.
koolba•28m ago
> Moral of the story. If you want even the most basic level of support from Github, you need to get on the front page of HN first.

This also works for Google support.

> And it seems they are able to do things very quickly, when they want to. Bastards.

I wouldn’t chalk any of this up to malicious intent. I’m sure they are swamped with such requests.

It was already a problem before agents could automatically perform these actions.

And it’s not something you can really automate on their end either. At least not the judgement call on the removal. Imagine if there was a fully automated process and it inadvertently took down a legit project.

debugnik•20m ago
> I wouldn’t chalk any of this up to malicious intent. I’m sure they are swamped with such requests.

Stalling in the hope that reporters won't escalate, instead of allocating a tiny bit of their billions in profit to hiring for this, is malicious in my book.

majorchord•14m ago
Can you provide evidence of these claims?
csomar•7m ago
I mean we are in the thread of evidence right now?
hsuduebc2•27m ago
Just send DMCA if you want their attention, they act harshly and quickly. Even when it's false one.