Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
lcamtuf argues that electronically controlled working memory—not Babbage—was the real bottleneck unlocked on the path to modern computers, tracing early registers, delay lines, and RAM.
8 min · 1,775 words
Psychologist Adam Mastroianni argues against the claim that ideas are getting harder to find: discoveries beget and replace old ones, many old findings were wrong, new tools reopen old questions, and falling per-researcher output reflects how science is organized rather than exhausted ideas.
23 min · 5,230 words
Gregory Gundersen reconstructs backpropagation from first principles, explaining why the algorithm needs a backward pass to compute neural network gradients efficiently.
6 min · 1,490 words