sieste

I thought the context would be obvious, but you're right. I've changed it to the full title.

you can only do things that suck when attendance isn't optional.

The late Jake Seliger and his widow Bess blogged at length about their experience navigating the medical system after his squamous cell cancer diagnosis. That was in 2023-25 though.

jakeseliger.com

bessstillman.substack.com

Ready Player One, together with my 9yo boy.

For people working with small local models like Qwen 35b what's even more important us that the LLM is good at coding in the language in which the agent harness is written so that it can write skills. I really like pi but had a few instances now where Qwen wasn't able to extend it because it isn't very good at writing typescript. Does that sound familiar to anyone, and did you find any fixes?

The couch to 5k programme is now routinely recommended by GPs in the UK (and elsewhere), a very positive development in health care in my opinion.

https://www.nhs.uk/better-health/get-active/get-running-with...

> I wanted to avoid teaching syntax. While Python makes sense as pseudocode [...] it is still nearly incomprehensible to new learners [...] I decided I wanted to either teach Forth or Haskell.

Sorry but this reads like a joke. If you find python incomprehensible, Forth and Haskell won't even register as languages.

What is a "credit" and how does it translate to tokens for the different models?

I treat these nutjob articles as a subgenre of science fiction and when they're somewhat well written find them very entertaining. The author cosplaying investigative journalist, one part of the audience acting as if it's a real discovery and another part playing debunking and outrage, is all part of the performance and it's hilarious. Check out the flat earth movement, it's great entertainment all around!

The zine enthusiasts in this thread might enjoy this little tool I made to print out and zine-fold Project Gutenberg books: https://github.com/sieste/pocketbook

German football goalkeeper here :)

I'd love for European countries to embrace their priorities around people and environment and train a model on public domain data that can run efficiently on cheap hardware. Redefine benchmarks around these ideals and optimise towards them, rather than trying to enter a race that you can't possibly win and didn't want to run in the first place.

The "middle powers" (cf Carney) should invest in local models, rather than relying on US and China allowing them to rent their AI models. It takes a single executive order to cut the rest of the world off of American AI tools. "I'm happy to pay whatever to rent frontier models from hyperscalers" makes sense if you're citizen of a superpower, but it's risky, naive, bordering on irresponsible to adopt this mindset otherwise, especially when your business or career depend on the tool.

Imagine how accurate you could be if you could circumvent the coding agent and just type the source code directly into an editor all by yourself o_O Like a write_file skill but for humans!!!

That's almost exactly my setup and I'm very happy with its performance.

I noticed recently that I started to prefer my local Qwen3.6 35B A3B and pi agent over Claude Code.

Both fail at different tasks, and Qwen more so than Claude.

But the way Qwen fails is much more straightforward. In writing tasks Qwens hallucinations and bullshitting are much easier to spot because it doesn't have the sleek vocabulary and wordsmithing skills to disguise its ignorance.

In coding tasks that Qwen can't solve it often just goes into a tool calling doom loop that the pi harness can catch, whereas Claude attempts ever more convoluted and creative things just making more and more mess that takes forever to clean up.

I think part of the story is that the tasks for which I use AI are fairly simple and maybe don't need a frontier model. But I wonder if "proper" developers had similar experience?

I have a different approach. Just treat all LLM queries as what they are, instructions to a computer program to generate a desired output. Neither niceties nor insults make a qualitative difference, so you might as well just skip them altogether.

It's a bit as if shell commands added im/politeness arguments that do nothing other than making you feel better about the interaction, like

    git pull --please

or

    ls --forthemillionthtime

I wouldn't use those either.

This is very cruel towards people who want to have children but can't.

Good point, it's like having an instruction "Never fucking output a token just because it's the one most likely to occur next!!1!"

Maybe a superhuman AI? /s

I was expecting to find a link to a github pages site where I can see the rendered examples, but only found a link to the html sources in examples/out. Am I missing something?

2021 just called and want their comment back.

Wow.

> The bill makes it a Class A felony (15-25 years imprisonment) to “knowingly train artificial intelligence” to do ANY of the following:

• Provide emotional support, including through open-ended conversations with a user

• Develop an emotional relationship with, or otherwise act as a companion to, an individual

• Simulate a human being, including in appearance, voice, or other mannerisms

• Act as a sentient human or mirror interactions that a human user might have with another human user, such that an individual would feel that the individual could develop a friendship or other relationship with the artificial intelligence