FeedCity

Today's links Shopping isn't politics: The personal isn't political. Hey look at this: Delights to delectate. Object permanence: Neither arphid nor RFID; Gor novel sex slave cult; Violent economist sex criminals; Vade et caca in pilleum et ipse traheatur super aures tuo; "We ...

I’ve been slowly listening to Poor Charlie’s Almanack: The Essential Wit and Wisdom of Charles T. Munger. I like his practicality. He’s never trying to be overly academic, as if he needs to prove how smart he is. He says Berkshire’s success doesn’t come from them solving har...

Plex sends a message to the self-hosting community with a massive upcharge targeted at the very people who hate monthly fees.

Intended status: Best Current Practice.

A major shift in the housing market in the last several years is the rapidly increasing popularity of “build-to-rent” homes — single-family homes that are built specifically for the purpose of being rented out.

De zorgen over digitale autonomie nemen toe, en steeds vaker krijg ik van (grote) organisaties de vraag: wat kunnen we NU doen?
Nou heb ik wel een lijstje, maar voor we beginnen, besef dat we 15 jaar lang alles naar Amerika hebben geschoffeld. Of, we hebben het uitbesteed, en DIE mensen stuurden het naar Amerika. 15 jaar is een lange tijd, en we zijn niet 1 2 3 weer “in vorm” om de zaken weer anders aan te pakken.

In April last year, Kelsey Piper discovered that OpenAI’s o3 model was surprisingly good at figuring out where a photo was taken from. Like human “geoguessr” pros, o3 could sometimes take a nondescript photo of a beach and tell you exactly where it is. Here’s the example Kel...

The standard answer to managing dotfiles on NixOS is home-manager. I've never used it, due to two aesthetic and one practical objection:

We’re working on an adventure game, Leonardo’s Moon Ship. The player character, Leo, has the usual pile of animations: idle left and right, walk in four directions, a couple of stair anims. Since implementing the basics of the character, Leo has had this slightly cursed behaviour where he’d grow about 3% taller the moment he started walking, then shrink back when he stopped. His feet would also lift off the ground by up to 27 pixels depending on which way he was facing. Not enough to be obviously broken in a screenshot. Exactly enough to make the game feel wrong.

We have the ability to use compute resources to support our proprietary AI applications (such as Grok 5, which is currently being trained at COLOSSUS II), while also providing access to select compute capacity to third-party customers. For example, in May 2026, we entered into Cloud Services Agreements with Anthropic PBC (“Anthropic”), an AI research and development public benefit corporation, with respect to access to compute capacity across COLOSSUS and COLOSSUS II. Pursuant to these agreements, the customer has agreed to pay us $1.25 billion per month through May 2029, with capacity ramping in May and June 2026 at a reduced fee. The agreements may be terminated by either party upon 90 days’ notice.

— SpaceX S-1, highlights mine

Tags: anthropic, grok, generative-ai, ai, llms

In the hours following the release of CVE-2026-45584 for the project Microsoft Windows, site reliability workers and systems administrators scrambled to desperately rebuild and patch all their systems to fix a memory safety vulnerability resulting in arbitrary code ex...

I just finished reading Privacy's Defender: My Thirty-Year Fight Against Digital Surveillance, by my friend Cindy Cohn. It's excellent and you should buy a copy.

Cindy is the Executive Director of the Electronic Frontier Foundation, and this memoir/legal history goes over three major legal

People assume all sorts of wild stock market returns when they make their financial calculations. Here are some numbers that show up on web searches:

6 %

8.4 %

10 %

10.1 %

11.3 %

11.5 %

13.6 %

16 %

These are all correctly computed under their respective assumptions, but they are very misleading because whatever those assumptions were, they’re not relevant for most calculations. You should assume 4 % in your calculations. Here’s one way to arrive at that:

(Continue reading the full article on the web.)

How fast is 10 tokens per second really?

Neat little HTML app by Mike Veerman (source code here) which simulates LLM token output speeds from 5/second to 800/second.

Useful if you see a model advertised as "30 tokens/second" and want to get a feel for what that actually looks like.

Via Hacker News

Tags: ai, generative-ai, llms

It's hard to find much to write about Google I/O this year because I have a policy of not writing about anything that I can't try out myself, and a lot of the big announcements are "coming soon". I actually prefer to write about things that are in general availability, becau...

I thought about posting this paper but rebranding it as the Claude Mythos technical report. As far as I can tell, there’s no secret tricks the US frontier labs have, and that basically describes how Mythos was trained. What’s in that paper just works, and for verifiable domains, it’s only a matter of fixing bugs and scaling up. That’s why Anthropic is so desperate for regulatory capture, AI has no moat.

Release: llm-gemini 0.32

• New model gemini-3.5-flash for Gemini 3.5 Flash.

See also my notes on Gemini 3.5 Flash, and the pelican I drew using this upgrade to the plugin.

Tags: gemini, llm

Release: datasette-llm-accountant 0.1a4

• Fixed bug tracking chains of responses. Refs datasette-llm#7

Tags: llm, datasette

Release: llm-gemini 0.32a0

• Compatible with llm>=0.32a0 alpha - adds the ability to stream reasoning tokens.

Tags: gemini, llm