Link Notes

I really like the petal mechanism, how each compartment is able to open without blocking the others.

I like stickers. The site is like all stickers! Fun reference for when you don't know what to doodle in your sketchbook.

I ask this question every few months, so I'm just saving this link here.

But don’t spend your life waiting for permission. Start the thing

--

​What you pay attention to shapes what you think about​. What you think about shapes what you do. What you do shapes who you become. So be careful.

Anytime you need to do something more than twice, spend the minimal time possible asking Copilot CLI to automate it. Don't write out a spec or give details or provide examples. Let it try to figure it out. Maybe it will fail, oh well, it is only a tiny bit of time spent, and then you can check where it failed and ask it to correct it.

--

Through this process, I also learned that automation allowed me to change how I worked, not just speed up my work.

I've been finding this out a lot lately, when you lean into AI, you find new ways of working; instead of thinking about how to implement, you just start focusing on the ideas you had. Instead of thinking, that'll be too much work to change, or procrastinating implementing a feature to test out and idea, you just use AI and do it, and start using the new feature.

"Copilot, read my Copilot session logs from the last 7 days and propose opportunities for automation. Draft a reusable skill for each."

I've always wanted something to watch my terminal session and do this for me. How can I vim better? how can I shell better? what tools are out there I don't know about that'll speed up my workflows. This would be a great (maybe a tad creepy) AI app. It just watches you work and then suggests ways you could improve your workflows.

What I’ve come to understand - and never really articulated to myself before - is that research creates its own momentum. The more time I invest in a decision, the more I start to feel like I should buy something, if only to justify all the effort I’ve already put into researching it.

I feel this way about ideas, you scribble down an idea, draw something in your sketchbook, and then you brain is like "YOU MUST NOW MAKE THIS THING". Calm down brain, I'm just exploring ideas here. Sometimes I don't explore ideas because of this; feeling like I have to build something if I spend the time researching/designing it.

Back to the actual article. My other problem with buying new things is Enshittification. I just don't want to deal with the lower quality version they made this year. I don't like buying stuff anymore, it usually just turns into work, frustration, and disappointment. The last time I found shoes I liked I bought 4 pairs, because I realized each year they would make these shoes cheaper and cheaper. At least Enshittification makes it easier to save money.

All I need is for the screen to pop up when I'm backing up for the backup camera, and then go away until I backup again.

I'm not sure if this is the message the article was trying to convey, but it seems to come off as: Vibe Coders are the people that have ideas, they build up their ideas with prototypes and experiments built via AI agents, while the Software Engineer is not the idea person, they are just there to implement code in a clean way. An analogy would be the architect that designs a beautiful building as opposed to the workers that take that design and then follow building codes to frame the house, or wire the electrical.

At the end of the article they do say the same person should be able to switch back and forth, being both the prototype person and then implementer, and maybe going forward that is the skill set you want to have to stay relevant.

A Library of video game / Chiptune music with a built in player for the web browser.

a cartulary was a medieval collection of important documents, charters, letters, and records that institutions preserved for future generations. A historical record.

Maybe all of these robots they are building to clean your house and answer your door can be repurposed to scan in documents; I can hand the robot my sketchbook, and it'll page through scanning and tagging each page in a digital format somewhere; a glorified scanner.

At first look, this table seems kind of pretentious, but then when you think about it, the idea is pretty ingenious. I think the image at the brewery was the selling point for me, since when you are in a place like that you are standing and there isn't a lot of places to put your drink or food, so if the brewery gave you trays that clip on, and your drinks sit in the middle part, you have a great way to hold your food without having to try and juggle it standing.

These might be kind of nice outside of food trucks, or at outdoor events.

"Radical simplicity is at the heart of our approach," Frascella says. "We achieve clarity by reducing everything to the essential. We live in a world that is often shrill, fast-paced, and overloaded. Almost everything is overdone. The danger of losing your way is greater than ever. Our responsibility is to be better and do what really matters. And the outcome always has to be an emotion."

I like the Idea of Radical simplicity in car design. I remember back when I was a kid, cars had no screens in them; that was nice, I kind of wish we could get back to that kind of radical simplicity.

I think my main problem with displays in cars are they are slow to start, and are just used to advertise crap; why am I forced to have a Pandora icon on my display, are they even still in business? Maybe a Radically simply display in a car is one with only the features you have, and goes away or turns off when not in use.

mechanical buttons in cars

A post about prototyping out functionality as opposed to just trying to build something from ideas / consensus / meetings... etc. Just build something and use it and see where that goes; make it easy to fix and change; add and remove as you find use cases through the use of the product.

When building something I always want to jump to the pretty end state, but really the best way to design a product it to build a janky prototype and use it every day; adding and removing features with actual use, not just ideas that something might or might not be good.

This article shows how the push went from just a pile of parts, to the final product you see today.