While machine learning research has made incredible theoretical advances, the day-to-day tools most researchers use are… poorly optimized, to say the least. And much knowledge is locked up in people’s private .bashrc files or wikis. This post aims to shed light on some very useful tools for beginning researchers.

A summary and rephrasal of three plausible scenarios for AI timelines. Most of this post is unoriginal. Cross-posted to LessWrong, and may contain more jargon than usual.

Covid is probably going to get much worse before it gets better. How do we protect ourselves and those around us? I propose a novel approach.

It’s amazing how tall the tower of computer abstraction is. Let’s explore a few of those layers here.

Yubikeys are great for security, but their benefits decrease somewhat when you leave them in your computer unattended. I also have login passwords that are way too long. How to solve these problems? Use a Yubikey to unlock your computer!

A look into call forwarding and MMI codes. On my cell provider, you can make calls automatically go to voicemail by dialing a single phone number. But how does this all work, anyway?

Don’t ignore webserver logs! Even a cursory glance can reveal some really interesting things about the internet.

An automatic mesh network that uses WireGuard under the hood. Why it’s good, and what I use it for.

Exposure notifications, or contact tracing apps, have failed. It’s been months since I wrote my previous post on contact tracing, and yet here’s what I see when I try to download an app.

Google and Apple’s exposure notification technology privacy-preserving, safe, and necessary. We need to embrace it.

A young upstart named TeaVM attempts to challenge bloated JavaScript web frameworks. Can it succeed?