VBA written by AI usually works.The problem isn't when it doesn't work.It's when it works, but you can't fix it. I read an ...

The short version: scrypt leans on memory, forcing an attacker to buy expensive RAM-heavy hardware instead of cheap GPU ...

LibreOffice patches CVE-2026-63277, a Calc flaw that runs remote Java code from a crafted spreadsheet; OpenOffice ...

An open-source Android app called Buddy can run Claude Code, Codex, or OpenCode directly on a smartphone and let the AI ...

Last time, I wrote about how I started building a web app that I could use on my smartphone because it was a hassle to carry a laptop just to enter Mahjong scores.This time, I will introduce how I ...

Take me to your reader.

AI agents are transforming legacy application modernization, but they still fail at three things. A grounded 2026 guide to ...

BLAKE3 beats SHA3-256 by up to 12x on throughput benchmarks, but SHA-3 keeps its FIPS 202 edge. Full specs, pricing, and ...

I gave an AI agent one prompt and scored the codebase after every commit. Health 50 to 61, 10,000 lines gone, and a hard stop ...

Why did the test fail: the product, the test, the data or the environment? An evidence-based triage rulebook with rules and 21 labelled cases.

Project Zenith is proof that Microsoft knows how to make a less annoying version of Windows 11. So, why is it making it ...

"If it works, don't touch it" is a well-known meme in the programming community. I'm glad that it's just that. A meme. Bad code didn't make me a worse developer. It made me faster, because I learned ...