I have a metal Apple Watch band that has started filing away the edge on one side of my MacBook just by accident.

To be fair sometimes if you update Linux too sparingly it results in conflicts. Of course the likelihood of that happening depends on the distro. Also the vast majority of Linux updates don’t require a reboot.

Qwen 3.5 can be run via ollama

Qwen 3.5 is one of the best of the open-weight (self-host able) models right now. It’s not as good as some of the extra massive proprietary models like the bigger Claude models.

These days we have websites for exactly that purpose.

A lot of the major furniture in my apartment came from people getting rid of stuff that I found via free-and-for-sale pages.

Sure, but to get the communication started you would start with facts you’d agree on, like the positions of stars or basic chemistry.

The model we currently have for the universe goes well beyond anything we could learn with our natural senses and the way we intuitively think about the world because of those senses.

It’s true that we keep refining our models and it’s very possible that an alien would have slightly different models, but at the end of the day, we are trying to describe the same universe and those models are going to overlap a lot because of that.

First of all, there has been a lot of research into what the minimal set of assumptions you need is to reproduce what we consider “basic math” and also what happens if you tweak those assumptions.

Second of all, the main goal for science and the type of math we use for science is to effectively model the world we live in.

Any aliens that live in the same universe are subject to the same physics, and any civilization advanced enough to detect our messages will know some basic universal facts about the world, and those facts are what we hope to use as the basis for starting communication.

Just make it “dee” like how people pronounce “D20” and it can keep “dice” as its plural.

Signal already has that setting. It’s up to the user to decide their level of convenience vs security.

data security in that case had nothing to do with the llm

That’s kinda my point.

“I don’t trust companies to hold their promises” is a very different argument from:

LLMs are inherently bad at data security and there is no way these companies can, in good faith, promise HIPPA compliance

It is certainly possible to implement a secure LLM service.

Phone, wallet, keys, and headphones

This is about extracting data that was used as training data. Just don’t do that with sensitive data.

LLMs are inherently bad at data security and there is no way these companies can, in good faith, promise HIPPA compliance

This is simply false. AI sucks but it doesn’t help to lie about it.

EDIT:

Go run a local model on your own computer, and delete the context when you are done. Boom you just used an LLM in a way that maintains your data security.

Carved into a city wall. Empúries, Spain.

Somebody do this to Minecraft and let Microsoft sue against AI

I’m sure whoever made this comic heard about that case

Fair tbh I’ve see ln a lot of unhelpful comments and issues left on repositories that end up getting popular outside of programming circles.

It is a lot more like water than you think. The solution of “just cut the circuit” is like solving the problem of overflowing storm drains by “just plug the pipe”.

The power has to go somewhere. If you don’t do anything about it, the voltage in the cables will rise until things start to fry. Real world power balancing involves adjusting the output of power plants (e.g. how much fuel to burn) in response to changes, and in some cases, dumping power into the ground as safely as possible. This problem gets complicated when power grids span vast distances and involve many different power plants that all need to be in sync or things catch on fire.

In the case of solar power, this is part of why improved large-scale battery technology is so important. It lets you absorb the excess power at peak generation times, and then release that power at night.