Good introduction to how a common encryption algorithm works, described by Ryo Nakao.

Researchers at Google applied GANs to crude sketches of imaginary creatures and managed to render them in rich detail. From their post on the Google AI Blog: Today, we present Chimera Painter, a trained machine learning (ML) model that automatically creates a fully fleshed

There have been many accounts about what it was like to work at Bell Labs, here's another short one by speech-processing researcher Man Mohan Sondhi. Two notable quotes, emphasis mine: The department that I joined was ostensibly concerned with various aspects of speech signal

At some point I'll get around to reading the paper, but meanwhile Jay Alammar has managed to explain how GPT-3 works in a short Twitter thread. It doesn't seem all that different from a typical language model. It's trained to predict the next word

A bit of light fun: 1SecondPainting uses a neural network to generate abstract paintings. Some of them are quite good: Generated by 1SecondPaintingGenerated by 1SecondPaintingGenerated by 1SecondPainting

Using neural networks to generate realistic, animated faces from portraits. Not all of them are compelling but the Mona Lisa one is very believable.

Probably the world's most interesting operating system. Some quotes from the CodersNotes blog on the OS: TempleOS is somewhat of a legend in the operating system community. Its sole author, Terry A. Davis, has spent the past 12 years attempting to create a new

'Dying since 1995', that's apparently the unofficial slogan for PHP. This year it turned 25, and it's more popular than ever. The JetBrains folks (developers of the PhpStorm IDE), have put together a timeline of the language's evolution since 1995. I wish PHP a

Rodney Brooks on the state of the peer review process for scientific papers: My three most cited papers were either rejected under peer review or accepted with no peer review.  So I am not exactly a poster child for peer reviewed papers.If a

Searching for root causes can often be misguided, because complex systems rarely have a single root cause. This article discussed some of the problems with the 'single root cause' idea: Again, this tendency to look for a single root cause for fundamentally surprising (and

This YouTube channel has summaries and demos for computer science papers, usually from artificial intelligence and computer graphics. Here's one on making Mona Lisa speak.

TIL that many large banks, airlines, insurance companies, and government agencies are still using IBM mainframes, a ~50 year old computing platform. In fact mainframes still seem to be in active development: IBM announced its new z15 model last year. Why are these ancient

A short 1-page overview of TCP for developers.

This 1991 paper discusses the author's experience with building two early time-sharing systems, and his learnings on managing reliability of such large software projects. To be honest, I paid less attention to the reliability parts of the paper than I should have. What really

An explanation of the IEEE 754 floating point standard that appeals to the intuition.

A set of interviews with senior engineers at tech companies.

Some solid advice on how to debug and in general make sense of complex distributed systems. Deductive reasoning is unlikely to work; we need to rely on empirical methods instead.

Vanevar Bush's vision of machines that would augment human knowledge.

This article discusses the life and work of Douglas Engelbart, who played a major role in modern computing – he invented the mouse, hyperlinks, and text and graphics editing, among other things. He demonstrated all of these inventions in a presentation in 1968 that would