My experience with moving from Mac to Linux for professional UX and UI design work and how to make the switch yourself.

macOS Sequoia is introducing a small change to the way certain apps can be opened, bringing new protection but also potentially annoyance too.

A new study found the amount of pesticides used on farms was strongly associated with the incidence of cancer. It comes on the heels of substantial lobbying by the pesticide industry to limit its liability from lawsuits over their products’ health impacts.

If you happened across the Olympic Air Pistol event for the first time, you might've been puzzled by the competitors' wearables: So those are mechanical irises. Shooting competitors fine-tune them to bring their gunsights into sharp focus. Here's an example…

How hard could it to be to build a two-stroke internal combustion engine (ICE) from scratch? This is a challenge that [Camden Bowen] gladly set for himself, while foregoing such obvious wastes of t…

Python's memory model is a performance bottleneck. Gauge is solving the monolith/microservices dilemma. We’re building tools to untangle codebases through incremental modularization. Our open-source toolkit supports defining and enforcing rules for interfaces…

A federal judge ordered Ripple to pay $125 million in civil penalties and imposed an injunction against future securities law violations on Wednesday.

The documents show inappropriate police surveillance of social media posts about pizza nights and study groups.

After the atomic bombings of Hiroshima and Nagasaki, the US Army Air Force had plans to drop a third bomb on August 19, 1945—with more to come if necessary. Fat Man bomb casing at the USAF Museum

Development of Gazette and related projects. Gazette has 2 repositories available. Follow their code on GitHub.

Elias Canetti described “The Book Against Death” as “the only book that I was born to write;” it was also a book he could not finish, or even properly start.

With macOS Sequoia this fall, using apps that need access to screen recording permissions will become a little bit more...

# RLHF is just barely RL

Reinforcement Learning from Human Feedback (RLHF) is the third (and last) major stage of training an LLM, after pretraining and supervised finetuning (SFT). My rant on RLHF is that it is just barely RL, in a way that I think is not…

Some conversations are difficult to have with members of older generations who grew up with different social norms. In particular, when you’re talking to people who grew up with PDP-11s with their completely flat memory, or Lisp machines or Burroughs Large…

Stare into the eyes of a stranger