From the archive
Sunday, 18 October
Folded from 3 sources that day.
Sorted by quality and by how many sources covered it, the same for everyone.
-
My obituary for programming
A long-time programmer who resisted coding agents describes trying them and finding they fix bugs and refactor code quickly. He reflects on what that means for a craft he enjoys.
-
The Problem with C++ Safety: Why "It Works" Isn't Good Enough
The first post of Rainer Grimm's blog after his death, by a new group of C++ writers, argues that C++ is not memory safe. It cites NSA and EU Cyber Resilience Act guidance favouring memory-safe languages.
-
The Brain Sandwich
Proposes a workflow for AI coding: think through the problem first, delegate the build to an agent, then review the diff and PR description before sharing it. It aims to keep engineering skills from atrophying.
-
Notes on the complex roots of unity
Builds up the roots of unity from real to complex numbers, using polar coordinates to compute the n-th roots and showing how they rotate on the unit circle.