This week came and went in a blur. I hadn’t realized how much i’d missed being on the creative side of things. Don’t get me wrong — i love managing my team of designers and planning our overall strategy. But i missed being in the thick of it, thinking deep and hard about the tiniest technicality, and being absolutely knackered by 3 PM. I’m thrilled that i still get to do it… and that i still got it!

Speaking of the creative side of things: in my People and Blogs interview, i mentioned Architypes, my little side project documenting old-school French storefronts. In the few weeks since, it’s garnered more interest than in the 10+ years i’ve been talking about it here in France. Huge thanks to Cassidy, Matt, Marilyn, Nicolas, Chad and the others for shining a light on this tiny project of mine.

Books

We Solve Murders by Richard Osman. Richard Osman sure knows how to write a thoroughly enjoyable book, but i don’t know, this one felt a bit trite. Most of the characters seem to have been dreamed up by the publisher’s marketing team to hit all the “key demos” and the hops around the world seem to have no other justification than to offer regular changes of scenery for the eventual film adaptation. Even the seemingly obligatory mention of ChatGPT feels uninspired at best. I really hope that it’s not a sign of things to come with the next instalment in the Thursday Murder Club series. (That being said, Trouble is a great name for a cat.)

Links

“The next evolution of The Verge’s homepage is here” by William Joel. They’re living my dream of five years ago. Contrary to my last employer, they understand the difference between traffic and audience, and are doing everything they can to transform fleeting traffic into a long-lasting audience. I wish them well in their effort to build their own community on their own (eventually federated) platform. That’s where the money is, and actual tech journalism deserves all the money it can get.

“MacBook Neo and How the iPad Could Be” by Craig Mod. In order to justify its existence between the iPhone and the MacBook, the iPad had to be, to quote Steve Jobs, “far better at doing some really important things […] than a laptop or a smartphone.” Alas, Apple quickly lost that ambition and revelled in making the iPad far worse at doing a lot of really important things than a laptop or a smartphone. I agree with Craig: “the iPad should be a highly-focused touch playground. Weird as hell, one-of-a-kind apps. And MacBooks should be for multitasking, moving information and data around, building evermore powerful tools (tools within tools within tools), all bounded by a keyboard-first universe.” It’s high time that Apple, if not the whole of the computer industry, understood that “consistency” doesn’t mean “convergence”. Things can be consistent in principle and wildly different in appearance. People will figure it out, if only you trust them to be cleverer than most of Silicon Valley thinks they are.

“At Machine Speed” by Matthias Ott. We’re in the midst of what i called a “LLM-powered cultural DDoS attack”, but Matthias describes something more nefarious entirely, an actual LLM-powered DDoS attack. “Language models can find zero-days and write working exploits faster than we can patch them”, which is putting intense pressure on solo-run and open-source projects. I’m sure that OpenAI and Anthropic will tell you that the solution is simple: let LLMs patch your code automatically, and if they make mistakes, more powerful LLMs will patch them in the future. In short: remove humans from the equation entirely and let the machines talk to the machines. It’s going to cost us dearly, and not only in euros and cents.