Posts

When most people think of summer, they talk about beaches, picnics, and pools. Not me. My best summer memories are at night, when dusk goes on forever and it’s ten o’clock before the stars come out. We used to hang out on my friend Mary’s second-floor porch, laughing, playing Cosmic Wimpout, and spinning old 78s.

The sounds of a city at night enveloped us. A chopper brumming on the next street over. A party across the way. Crickets mixed with car radios, echoes bouncing off buildings, warped and Doppler-bent. We talked until dawn some days. How was there ever that much to say?

Something shifted for me recently. This week, for the first time, I could actually see myself walking away from the flagship corporate AI products. Not from AI itself — from the companies.

This week I did a talk and published a companion website for nonprofit fundraisers on how to get started with AI, including tips, pitfalls, and some of the tools I recommend. It went well, and sitting with all of it afterward, I realized I’d never written a general version of that advice for this site. Most of what I write here assumes you’re already using these tools and looking to go deeper. This post is for those who might be a bit earlier in the journey.

When we got our first house, it also came with our first lawn. Right about this time of year, the dandelions would start sprouting everywhere, invasive weeds ruining the manicured green that kept us in good standing on our block. Wanting to be the perfect homeowner, I bought an amazing contraption that let me yoink them out, one by one, without even bending down. It was so satisfying to keep those forces of entropy away … until a couple of days later, when the lawn filled up with them again. Come on!

A few weeks ago, Robin Sloan published a newsletter asking a fascinating question: what is it like to be a language model? His answer — that “the model” is the forward pass, a flash of computation lasting milliseconds before dissolving into nothing — sent me sideways into a different question entirely.

Anthropic (the company behind Claude, the AI assistant I use daily) recently released a new model called Claude Mythos Preview. If you’ve heard of it, you’ve probably noticed that it’s not available to us. That’s intentional, and the reason why has me all fired up.

This week we finally got our first real hint of spring. The kind where you can open the windows and let the outside world back in after months of keeping it at bay.

I didn’t decide to take a break from AI. It just kind of happened.

We’ve had a rough winter. Penetrating cold that clings to your bones. Ice dams, pipe freeze protocols, shutting down the upstairs bathroom. The kind of winter where the days blur together and you realize somewhere along the way that you’ve been going through motions rather than actually living inside them. You suspect you need something. You’re just not sure what, or how much.

I’ve been thinking about the gap between the list of things I want to get done and the things I actually get done. For a long time, my attention went almost entirely to what wasn’t finished, and kept getting pushed back. Recently I’ve been trying to flip that around and look at what did happen. Turns out it’s usually plenty. Writing. Teaching. Following through on things that matter. The list of what’s actually getting done is longer than I give it credit for.

The first time I heard about what would become the Khmer Magic Music Bus, I wasn’t even in Cambodia. I was listening to my colleague Steve describe a rest stop on a long drive through the countryside.

There’s a moment every computer person eventually has, usually more than once, where something gets deleted that can’t come back. Working in the command line for years, I’ve been there. It teaches you, more than anything, that the things you can’t take back deserve a different kind of attention than the things you can.