Old Fucking Punk @lwriemen@librem.one

Big fundraiser for @osnews, in honour of my 21 year tenure and 20000th post. Since my wife can barely work due to health issues (caused by her work, ugh), my family has to rely on my income much more heavily. And that means OSNews donations.

https://ko-fi.com/thomholwerda/goal?g=2

I've also got a few fun incentives:

Make me use Windows 11
A photo/video tour of my office and all my computers/devices
Make me use (and review) macOS
A tattoo of the OSNews logo (my first tattoo) on my shoulder

More details:

https://www.osnews.com/story/144982/the-21-years-and-20000-posts-osnews-fundraiser-a-euro-for-every-post/

Your support keeps OSNews free from ads and corporate influence, and allow me to continue to run what is one of the very last people-focused, independent technology news sites left in the world. Also, OSNews is militantly "AI"-free, in every step of the process.

“AI Software Development – What Does The Data Say? – Codemanship's Blog”

https://codemanship.wordpress.com/2026/08/12/ai-software-development-what-does-the-data-say/

> If anything, the average team is taking longer to ship worse software

Anecdotally, this is what I've been hearing a lot too. Higher costs, more volume, worse outcomes. But still “inevitable” for some reason

I know that people say that AI is great for soft skill nonsense like, summarizing chats and emails and making calendar appointments, but this is the first time I'm on the job market since the AI craze really took off and it seems to have made everyone worse at their jobs. All recruiters are late or no-show to meetings. They have no idea what we talked about over linkedin or email. They forget who I am. They didn't read my resume. They seem to have gotten worse in every way, and their whole just was just to be a secretary.

If you're going to be this bad at your job, why even use recruiters anymore?

So I’m writing an article for a new publication who said they’ll be using an AI checker on my writing. As those things sometimes give false positives for me (since LLMs non consensually used my writing for training), I screen recorded my writing process just in case. The future is so ridiculous.

📢 INTRODUCING THE ESPHOME STARTER KIT! 🎉

Designed and built by our commercial partner Apollo Automation, the ESPHome Starter Kit gives you the tools, skills and confidence to stop just using smart home tech, and start creating it.

Build, Learn, Automate.
https://esphome.io/starter-kit

“While the world frequently reads about Puerto Rico’s blackouts, the quiet collapse of the water system has pushed the island to its absolute limit.”

Coenia Sanchez reports on the federal policies and funding delays behind the crisis.

https://inthesetimes.com/article/puerto-rico-water-crisis-trump-congress

Dockworkers have refused weapons cargo, shut down ports and joined mass strikes to demand an end to the war on Gaza. On October 30, unions representing workers at 40 ports will act together. Sarah Lazare reports: https://inthesetimes.com/article/dockworkers-six-countries-announce-port-strikes-october-30

BREAKING: FPF, The Intercept sue Trump for selling early access to his Truth Social posts.

This scheme is extraordinary, corrupt, and unconstitutional, and we are bringing this case to stop it.

https://freedom.press/issues/lawsuit-early-access-fee-for-trump-truth-social-posts-unconstitutional/

🔴 Brand new comic now up on Boing Boing
-
SUPER-FUN-PAK COMIX, feat. Mank & Mook, Uncle Cap'n's Ant Farm, Tim Tripp Time Traveller, & MORE MORE MORE
-
https://boingboing.net/2026/08/12/tom-the-dancing-bug-super-fun-pak-comix-feat-super-pals-mcsm.html

Everyone wants free candy

From Riot to Insurrection is a vivid, electrifying account of the 2001 Genoa G8 protests, an ode to audacity, and proof that revolution remains possible when we fight for it.
Get 20% off this new title with code August here: https://pmpress.org/index.php?l=product_detail&p=1956

RE: https://toot.cafe/@baldur/117081821641896200

The lesson I learned from this post is that people have absolutely no clue how widespread smoking used to be or how much of a toxic toll on society it was.

GNOME Foundation is vital to the project's wellbeing, and without it the project would not be able to continue as it does today.

Please become a friend of GNOME to keep Foundation alive! 👉 https://donate.gnome.org

#FLOSS #FOSS #OpenSource #Programming

NEW: General Dynamics opened a Texas factory to churn out artillery shells for Ukraine, but the machines kept breaking.

Robot arms smashed into things. Hunks of metal crashed to the floor. There was even a real-life dumpster fire.
https://www.propublica.org/article/general-dynamics-artillery-factory-failed?utm_source=mastodon&utm_medium=social&utm_campaign=mastodon-post

#News #Defense #Military #Ukraine #Government #Technology #Tech

You are telling people that it's okay to type confidential information into Microsoft Copilot.

That. Is. Not. True.

Microsoft admits that they *will* violate local laws in order to share confidential information hosted on non-US servers with the US government.

There is no version of Microsoft Copilot that isn't subject to American courts.

That is the most important thing to know about LLMs!

This article starts with a story from someone who tried Claude Code and found it amazing, but then switches to the same person a few months later seeing what a disaster it’s been. This quote is key:

The problem is that code produced by an AI agent looks reasonable, but can contain ‘hard-to-spot bugs’ that end up causing major problems

LLMs, by their nature, generate statistically plausible output. That is often a set that overlaps with correct output. But the things that are not correct look exactly the same as the ones they are.

Learning to review code is hard. You look for the bugs that you expect to be possible implementing it by using your theory of mind for the person writing the code and the kinds of things that they might overlook (not necessarily a specific person, but the kinds of things people miss) and also common bug classes.

And the big help is that the person writing the code is not thinking adversarially. They are not trying to sneak bugs in. Normally. Unless they’re a supply-chain attacker, and we’ve a depressing amount of evidence that code review doesn’t catch supply-chain attacks.

An LLM is not trying to do anything. It has no intent. But it is a machine that is trained on code that made it past code review. The kind of bugs that it will generate are ones that look like code that appeared in production. This is exactly what an attacker would do: try to write code that looks correct but is subtly wrong.

I’m only being slightly flippant when I say LLMs are a mechanism for brining supply chain attacks in house.

https://xcancel.com/aakashgupta/status/2087036537930424351?s=46&t=

Maybe the Pentagon should start a GoFundMe to get more missiles. https://patreon.com/BrianMcFadden

As I wrote on Destructured, AI companies' current investment in data center and generator construction is the single best metric for assessing AI's future energy demand: https://futurism.com/artificial-intelligence/amazon-data-center-largest-pollution-source-texas-usa Anyone who suggests that the energy consumption of AI is overstated is ignoring exhibit number one.

«they simply replace skill barriers with dependence on proprietary services, while users of these services often trade full community participation and all its benefits for interactions mediated by LLMs trained on community-built open source projects. Our communities are already noticing a shift away from collaborative stewardship and toward transactional interactions with projects»

If FLOSS keeps accepting this state, it's digging its own grave.

https://greatscottgadgets.com/2026/06-17-llms-generate-code-communities-build-projects/