*“Yesterday, I Went to Mars ♡”*

This episode looks at the fast-moving smart glasses space, prompted by Warby Parker's announcement of a collaboration with Google and Samsung — a development that arrived just one day after a previous post on the same topic. It walks through how the category is beginning to split along AI lines: Meta with Ray-Ban and Oakley, Google and Samsung with Warby Parker and Gentle Monster, and Even Rea...

This episode looks at two recent developments in smart glasses: Subvocal, a device that converts silent mouth movements into text using radio waves, and Ray-Ban Meta Audio, Meta's camera-free version of its smart glasses line. The Subvocal section walks through how the technology actually works — not thought-reading, but mouth movement detection, with no microphone and no camera. It's an i...

This episode looks at a news story about Google disabling AI features by default for student accounts, and the quiet but interesting detail buried inside that announcement. It touches on a feature called Guided Learning — where instead of simply giving a child the answer to a math problem, the AI asks questions back and helps them work toward it themselves. A tool built to answer anything, deliber...

This episode sits with a small but recurring question: whether to finally close down a dormant company registered in Hong Kong, or to keep holding on to something that hasn't moved in years. It touches on the odd timing — having written the day before about letting go of what you don't need, and then flying to Hong Kong the very next day to file paperwork for exactly that kind of thing. Th...

This episode looks at a piece of SpaceX news — the company quietly stepping back from offering new launch services for small lunar landers — and what that decision might signal about where SpaceX is actually headed. It touches on the contrast between that pullback and another figure: a stated vision of up to ten thousand rocket launches per year, roughly twenty-seven a day. The tension between tho...

This episode reflects on a shift that's quietly happening right now — the move from "I wonder if there's a service for this" to "let me just build it myself." The catalyst was a small personal frustration, a quick conversation with an AI, and something usable running thirty minutes later. It draws a comparison to the Google era, when knowing how to search gave people an...

This episode looks at two ways of relating to AI and automated systems — being "in" the loop versus "on" it — and what that distinction quietly reveals about how work actually gets done. It touches on the shift from checking tasks as they come to stepping back and seeing the whole picture: not absorbed in the process, but above it, stepping down only when a real decision is nee...

This episode looks at a technology called PRIMA — a system pairing specialized glasses with a tiny chip implanted beneath the retina, designed to restore some central vision lost to advanced macular degeneration. It walks through how the system works: a camera in the glasses captures a scene, a processor refines it, and near-infrared light projects that image into the eye — carrying both the visua...

This episode looks at a day spent almost entirely working with AI — testing how much of the daily work of running an eyewear shop could be handed off, from photo editing to writing. The tiredness at the end of that day is real, but it came from something unexpected: not from doing things, but from constantly deciding whether what the AI produced was right, and then explaining why it wasn't. Wi...

This episode looks at Amazon's resistance to Meta's AI shopping agent Muse — and what it reveals about who controls the entrance to retail. It traces how Google once took hold of that front door, how stores learned to compete for search rankings and ad placement, and how AI agents may now be taking hold of the same door in a more consequential way: not just surfacing options, but selecting...

This episode looks at the sudden arrival of agentic AI tools — Muse, Dots, and Grok Bot — and the honest experience of using all three without yet knowing which one to reach for, or when. It covers how each works: Muse opening browsers and booking travel while pausing to confirm before it acts; Dots running continuously on a dedicated cloud computer, connected to over 4,000 apps, learning preferen...

This episode looks at the new beta version of Siri arriving on Apple Watch, and what it's like to actually try it the same day it becomes available. The honest impression here is that Gemini on the Pixel Watch still feels a bit more natural — better at holding a conversation, a little sharper at understanding what you're asking. Siri has improved, but the comparison is still there. It also...

This episode follows a purchasing decision that kept shifting — starting with a plan to buy a Mac mini, moving toward a Mac Studio, and then arriving at unexpected uncertainty just as the new model was finally announced. The arrival of OpenAI's Dots is what changed things. It's an AI that works on your behalf, continuously, on its own cloud computer — even when your own machine is powered...

This episode looks at a piece of research news: a technology called Scentronic, developed at Ohio State University, that may one day allow smells to be transmitted over the internet — detected by an electronic nose, identified by AI, and recreated on the receiving end using pre-prepared fragrances. Early experiments worked with coffee, orange, cake, beer, and red wine. From there, it becomes a qui...

This episode sits with a question that opens quietly but goes deep: rather than simply hoping a child meets a transformative coach someday, what if a parent could become that presence themselves? It traces what research and reflection suggest a great coach actually does — not only building technical skill or stacking up praise, but helping a child feel that they belong, that they have a role, that...

This episode draws an unexpected line between a 0.15mm sizing error in a lens edger and the error-correction problem at the heart of quantum computing. It touches on a specific frustration: lenses coming out fractionally too small — a difference invisible to most people, but consequential in eyeglass work where a lens cut too small can't be salvaged the way one cut too large sometimes can. The...

This episode looks at NASA's CHAPEA mission — four people living inside a 3D-printed habitat in Houston for 378 days, simulating life on Mars, with delayed communications, assigned work, and no stepping outside. It draws an unexpected parallel to a memory from years ago: climbing Denali as part of a team, where one person's limits could stop everyone, and sitting honestly with how that sit...

This episode looks at a Forbes JAPAN article about verbal abuse directed at AI — and Anthropic's reported move to add a clause to its terms of service prohibiting sustained, unnecessary abuse of Claude. It touches on something quietly counterintuitive: because generative AI learns from human language, directing insults at it can pull responses in a rougher direction — not because the AI is hur...

This episode looks at something familiar but rarely named: the way curiosity, not work, is often what keeps us awake. It traces a pattern — an AI conversation raises a new question, an overseas glasses brand leads to another, a news story about space opens into the technology behind it — and how each thread feels less like distraction than research, which makes it harder to stop. There's a ref...

This episode looks at Elon Musk's latest Tesla Roadster announcement — a car that uses SpaceX-derived thruster technology to lift off the ground, reorient itself in the air, and land back on its tires. It touches on the specific hardware involved: carbon-fiber-wrapped pressure vessels, fans, and diffusers pulled straight from the space industry, with mass production reportedly targeted for 202...

This episode looks at an unexpected encounter with the future — a paper flyer in the mailbox announcing that a nearby building would be inspected by drone, operated by Rakuten Drone Co., Ltd. It touches on the gap between knowing something exists and actually having it arrive at your doorstep. Drone inspections, AI-analyzed footage, autonomous systems — all familiar from the news. But a single she...

This episode revisits Queenstown, New Zealand — a place visited just two months ago — through a completely different lens, after coming across a newsletter describing the area as a favored doomsday retreat for the ultra-wealthy. It looks at Peter Thiel's acquisition of New Zealand citizenship after just twelve days in the country, and his purchase of roughly 193 hectares of land near Lake Wāna...

This episode looks at a phrase that seems harmless but carries a quiet weight: "that's not like you" — and what it might actually be doing when someone says it. It touches on an idea from a newsletter by Takashiro Go: that the phrase might function as a kind of preemptive defense, a way of resisting change in someone else because it would require rethinking a familiar image of them....

This episode takes Hikaru Utada's phrase "human activities" — the term she used to describe her 2010 hiatus from music — as a starting point for thinking about the gap that can quietly open between a social role and the self living inside it. It touches on how Utada described feeling that her true self and others' image of her had grown further and further apart, to the point where she had lost a...

This episode looks at what came up at a daughter's parent-teacher meeting — specifically, the part about sleep that ended up feeling more personal than parental. The detail that lands hardest: by midnight, after waking at seven, cognitive function has dropped to the equivalent of a 0.1 percent blood alcohol level. Legally drunk. And it doesn't even require an all-nighter — six hours a nigh...