Last week, I updated the layout of my digital garden and pulled over a ton of stuff that was behind the members-only paywall. I’ll be doing some pruning and tending to the garden over the coming weeks. The sections on fetch(), flexbox, and CSS grid could all use some cleaning up to make them more clear and simple. I’ve got guides around web components, linting, and JS testing that I haven’t added yet, because they need a complete rewrite based on how I work today.

I have a confession: for most of my career, I’ve hated testing. But I recently learned that’s because I’ve been doing it wrong, and maybe you have, too. In this video, I show you… The “right way” to approach testing. What I got wrong about TDD all these years. How to make testing easier, simpler, and more useful. If you’re not someone who likes watching videos to learn (👋 hello, my fellow millennials and older!

My toxic trait is that I believe access to the things you literally need to survive (food, clothes, shelter, healthcare) should be free. I know, I know… If people didn’t have to work to survive, why would anyone go to work? Congratulations, you’ve just unlocked the hidden door to the capitalism engine room! Capitalism needs poverty Capitalism depends on the looming threat of poverty, suffering, and death to work.

“Artificial intelligence is the future,” they tell me. The thing is, futures are tricky.

A few months ago, I wrote about how Safari is holding back the web. Today, I wanted to share another example of a Safari bug that’s apparently done on purpose that I caught when testing a new web component I’m building for Kelp UI. Let’s dig in! The behavior I’m building a <kelp-disclosure> component. A disclosure component is one where clicking or pressing a button displays some hidden content, and clicking or pressing it again hides that content.

One of the more surprising things to me about the AI era is the number of thought leaders in our space who have historically been against tech that causes massive harm to humans but regularly advocate for AI use. They almost always treat AI as if its inevitable. Take, for example, this thread on Bluesky from Anil Dash… So there’s this article atop Hacker News right now that says that “nobody wants AI in Firefox”, trying to make the argument that Mozilla shouldn’t go down that path…