Hyperlegible 003: Julian Lehr

After almost two years, Julian Lehr is BACK to writing. He wrote a piece called The case against conversational interfaces, arguing that we're not going to be talking to our computers instead of using graphical user interfaces. GUIs work pretty well! Instead, he thinks that conversational interfaces are going to be a complement to existing workflows. We'll talk to our AI while doing what we do now, to do things like tell other apps to start doing things while we stay in flow.Julian shares his writing process -- chat through a draft with AI, write ~60% of it by hand, and then pull it together in Figma, or sometimes, Google Docs. He said that like some people need a change of scenery to write, he needs a change of tools. So why did he come back after two years in the wilderness? Simply: too many people were too consistently wrong on the internet. After seeing one too many "we're all going to be chatting with our computers" takes, he had to write the other side. And he delivered. We cover a lot, from why he keeps coming back to Kevin Kwok’s Arc of Collaboration to how he uses his "thanks to" section to status signal. For this essay, he thanked Blake Robbins, Chris Paik, Jackson Dahl, Johannes Shickling, Jordan Singer, and Signulll -- an absurdly high signal roster. Conversations like this one - where I get to nerd out with the people I've read for so long - is exactly why I'm doing Hyperlegible. I hope you enjoy it as much as I did.