Old Fucking Punk @lwriemen@librem.one

the reason that every consumer model takes great pain to hide raw "thinking" output is that if you actually see it then the entire illusion is broken and you see that what is actually happening is the worlds most expensive brute force process exploring every statistical linguistic possibility like an amnesiac dictionary before eventually taking a wild leap of chance back into its next hail mary execution. Every skill prompt for muse that's about introspection describes exhaustively how no thinking/reasoning text can be returned from it, but it's just right there in the filesystem. Claude code never even returns the true reasoning output to the client - it actually returns another LLM summarizing the thinking output as the commentary text. For every token you see, a thousand random wander tokens are emitted, and the token economy of Muse is the next escalation, if the amount of heat energy needed for a unit computer thing used to be a gas stove, Muse burns an entire star to deliver something sorta kinda resembling a unit computer thing

#WritersCoffeeClub – 16th Oct. How do you feel about people reading your works without buying a personal copy?

Long-standing publishing lore is, there are roughly four readers per book sold. This includes library loans, second-hand sales, and friends borrowing books. This is normal.

I don't get mad at warez sites (most downloads go unread, most readers wouldn't have bought one), but I *DO* get enraged by pirates who re-list and sell my ebooks on amazon (they're stealing income from authors).

“Epic Games goes full vibe code: MAXIMUM AI ACCELERATION – Pivot to AI”

https://pivot-to-ai.com/2026/10/15/epic-games-goes-full-vibe-code-maximum-ai-acceleration/

“Georgia’s Building Boom for Data Centers Comes With a Cost for Schools: $300M – The 74”

https://www.the74million.org/article/georgias-building-boom-for-data-centers-comes-with-a-cost-for-schools-300m/?ref=2ndbreakfast.audreywatters.com

> The organization estimates that the losses in some of the state’s school districts would range from $3.6 million in Newton County to $35.5 million in Gwinnett County, the state’s largest district.

Turns out AI can cause more problems than it solves. Cleaning up AI slop is one issue. 36% of AI execs at companies with $1B+ in annual rev reported an AI incident that led to data loss, financial damage, brand damage or operational disruptions. Major companies are now limiting employee AI usage. https://www.theguardian.com/technology/2026/oct/13/ai-companies-rushed-workers-fallout

“Make Shift”

https://2ndbreakfast.audreywatters.com/mis-shift/

> What ed-tech has forgotten to prepare students for: their present lives. Their humanity.

Forget a billionaire's algorithms, use RSS. Most sites have them.

Both the Thimbleweed Park Blog and Grumpy Gamer do. Full text feeds.

I've been using and serving RSS since 2005.

All the cool kids are doing it.

https://toolbox.rss.here.now/

@energisch_ I agree. This is why I hesitate to speak of a "bubble" in the first place for reasons @alineblankertz has explained as well:

https://www.structural-integrity.eu/crashing-hard-why-talking-about-bubbles-obscures-the-real-social-cost-of-overinvesting-into-artificial-intelligence/

Trump and Hegseth would be an embarrassment if they didn't have control of weapons of mass destruction. As it is, they're war criminals and need to be held to account. These women are heroes and are entirely legitimate and correct. https://www.youtube.com/shorts/COvmkDsRdUA

One of the ecology postgrads working with me at the moment, Dee Patel, has spent this year doing lots of data curation work, and now stats analysis, on my 23-years of ongoing high frequency bird counts between Ōtautahi-Christchurch city and Lincoln town in Canterbury, NZ. That's where I repeatedly map and count birds while biking and running along 172 km of standard routes about the landscape.

This week Dee got the first preliminary results for all birds and one clear story is worth sharing: of the ~21 species of native and endemic birds that were seen often enough to be able to analyse a trend, *ALL* have increased in numbers over the past two decades.

All of them!!

Many have increased by large amounts, and on average, the NZ endemics are up more than the natives we share with other countries.

Over the same period, the introduced birds (mostly English birds) have mostly stayed about the same, with just a few declines (eg Muscovy duck are down a bit).

It's a strikingly different result from what's been reported in the media from some other parts of the world. I expect it's due to the big increases here in regeneration and planting of native vegetation and mammalian predator control, leading to a resurgence of native species in areas where they'd declined last century. We've also had the agricultural intensification and urban expansion and warming climate seen in most of the world, but those negative effects on birds seem to (so far) have been more than offset by the return of native vegetation and reduction in predators.

Dee's due to complete and submit her thesis early next year so there's still plenty of work to do before we get the details ironed out and figure out all the nuances in the story. Still, the overall result is looking really promising for the city's birds.

#birds #Ōtautahi #Christchurch #nz #ecology #LincolnUniversityNZ

Fun discovery: Firefox can't tolerate two bookmarks with the same URL but different keywords.

If you try to do it, the keyword gets silently removed from one, or your change actually affects the other bookmark.

RE: https://hachyderm.io/@rain/117446083340236126

This is why, when people say things like "omg the LLM can generate an embedded kernel for me!!!" I can't help but think

"Yeah, it was trained on _at least_ three of mine, plus immeasurable training load on my webserver serving the docs."

By claiming ownership of the result, or even just claiming to have managed the machine while it spit it out, you're basically just stripping all attribution for the actual work that went into its inputs. You might as well just clone my repo, remove the license, and remove my name. So no, I am not impressed.

Anyway, Anthropic just ripped off a bunch of Rain's work with no attribution, as they do.

dear audio experts of fedi : how can u irl-suppress knocking sounds (of eg people on doors, hammers, etc) ? asking for trauma trigger reasons
boosts appreciated to help get answer

ChatGPT now lets you target ads based on specific prompts or any phone number or email address. This is absolutely 100% going to be used for harassment.

https://www.404media.co/how-chatgpt-targets-ads-to-you-based-on-your-prompts-phone-number-and-email/

It's also very frustrating to constantly have my opposition to AI be taken as an opposition to abstraction, or to automation, and a desire to do every little thing by hand. But it's the opposite. For a whole bunch of reasons, AI anti-abstracts. It adversarially creates oceans of duplication to process. It creates a situation where the AI user gets to set the terms of some ecosystem with synthetic productivity but then everyone else needs an AI to figure out what's going on in that new ecosystem.

#Forgejo 17.0.0 was just released!

Repository funding links are supported. Repositories can be migrated from Bitbucket Data Center. LFS support is revamped. Job summaries from Forgejo Actions jobs are displayed.

Check out the blog post at https://forgejo.org/2026-10-release-v17-0/.

#forgejorelease

🔵 This week's Tom the Dancing Bug
-
DONALD AND JOHN: A Calvinesque and Hobbesian look at the midterms
-
PLEASE JOIN THE INNER HIVE TO SUPPORT TOM THE DANCING BUG; tomthedancingbug.com
-
READ IT RIGHT HERE👇

I'm sitting in one of the fullest European Parliament hearings I've ever seen: it's about the infamous proposed Cloud and #AI Development Act #CADA and it's a huge "we need to build build build #datacentres", #gigafactories and farring #unicorns. And I'm like 😭

Let's see if I get speaking time. :)