flâneur

Mathurah Ravigulan

144 followers · 54 following · 5203 views

on the atlas — 181

highlights — 271

  • So I hope that this history of the Sino-Indian divergence conveys a simple lesson: if you want your country to go from poor to rich, the most important thing is investing in your people.
    Why China got rich and India didn't - David Oks
  • between 1949 and 1976, the Communist Party totally transformed Chinese life.
    Why China got rich and India didn't - David Oks
  • China invested in its people; India did not.
    Why China got rich and India didn't - David Oks
  • Indian state never managed to develop its people’s human capital as China had
    Why China got rich and India didn't - David Oks
  • Become more visible, because this makes it easier to make friends.
    The Old World Is Dying: Advice for 2026 graduates
  • AI makes the world more nepo.
    The Old World Is Dying: Advice for 2026 graduates
  • But god, don’t go to the Notes app and post it. Write about it. Make art about this. Don’t respond to trolls in your comments. That’s not what we want from you. We want your art.
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • I can’t tell a person how I feel, so I’ll write it in this song
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • Let’s make it a music conversation, rather than just ganging up on the female artists. And I think the more male artists that are messy, or emotionally complex, or confessional, or upset, the happier I am.
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • rant bridges where it’s basically stream-of-consciousness — endless pouring-out of emotion, intrusive thoughts, blended with metaphor, with discussion, with shouting.
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • I don’t have social media on my phone. It looks like I’m endlessly scrolling, but I’m scrolling through words, the words in my file. If we’re in the middle of writing a song, I’m searching for a perfect line that I thought of four years ago at 3 in the morning.
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • And this is why you need to discipline your kids — because they might write songs. That go No. 1.
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • I have this very strong opinion that when you’re young, you feel things on such an intense and detailed level. There’s an attention to detail when you are 17 to 22 years old, and you’re longing, or you’re reaching and grasping but never holding — someone’s attention, or someone’s love, or someone’s dedication — and you just can’t understand why you spend all day thinking about it. You notice everything. You notice candle ash on the cuff of the shirt, and the button, and it’s everything that makes the mythology of those intense feelings that you have. And I’ve always tried to — without being a …
    How Taylor Swift Writes a Song, In Her Own Words - The New York Times
  • If you're worried that people will think you're cringe for posting too much, that's also fine. Most people won't see it, and even the people that see it probably won't remember because our memories are so rotted from scrolling. If someone remembers your work enough to criticize it or you, you've already won
    You should probably post more | aadil pickle
  • The problem is if you create in a vacuum, you'll have no way to improve.
    You should probably post more | aadil pickle
  • Yet, to make anything good, you have to cross the chasm of cringe
    You should probably post more | aadil pickle
  • Well-enforced export controls11 are the only thing that can prevent China from getting millions of chips, and are therefore the most important determinant of whether we end up in a unipolar or bipolar world.
    Dario Amodei — On DeepSeek and Export Controls
  • All of this is to say that DeepSeek-V3 is not a unique breakthrough or something that fundamentally changes the economics of LLM’s; it’s an expected point on an ongoing cost reduction curve. What’s different this time is that the company that was first to demonstrate the expected cost reductions was Chinese
    Dario Amodei — On DeepSeek and Export Controls
  • DeepSeek produced a model close to the performance of US models 7-10 months older, for a good deal less cost (but not anywhere near the ratios people have suggested)".
    Dario Amodei — On DeepSeek and Export Controls
  • People are naturally attracted to the idea that "first something is expensive, then it gets cheaper"
    Dario Amodei — On DeepSeek and Export Controls
  • “Modern loneliness masks itself as hyper connectivity. And so people have easily 1000 virtual friends, but no one they can ask to feed their cat. That loneliness, which is really a depletion of the social capital, is extremely powerful. […]
    The friendship problem - by Rosie Spinks
  • I paid attention to things I liked to do, and found ways to do more of that. I made it easy for interesting people to find me, and then I hung out with them. We did projects together.
    She asked me what I wished for on my wishlist - by Aadil
  • “I’m just going to pay attention to what happens when I hang out with various people and iterate toward something that feels alive”
    She asked me what I wished for on my wishlist - by Aadil
  • Because at a certain point that was me, and still can be. And I’m so grateful for the people in my life that didn’t see me as unworthy of love. They just see me for who I am.
    She asked me what I wished for on my wishlist - by Aadil
  • It just means giving myself some grace when I don’t easily get along with people rather than beating myself up or changing myself to be likeable.
    She asked me what I wished for on my wishlist - by Aadil
  • "You have to pay your own electric bill. You have to be kind. You have to give it all you got. You have to find people who love you truly and love them back with the same truth. But that’s all."
    gracenguyen.ca/montreal
  • There are decisions that won’t change your life in the long run but will feel good and agentic, and for that reason alone you should do it. I think youth so far is just about running experiments, one after the other, learning from it and being diligent in the repetition of that pursuit.
    gracenguyen.ca/montreal
  • I now tell younger friends that a job—not a shit and stifling one, but one where you can really learn and grow—is actually a great place to spend your early twenties.
    exit interview - by Jasmine Sun - half-baked futures
  • Delusion and faith are two sides of the same coin. You need a little bit of both to dream.
    the weight of dreams - by Spencer Chang
  • A cell of your own making that constantly asks for more from you—more courage, more endurance, more delusion.
    the weight of dreams - by Spencer Chang
  • I don't feel able to let myself settle for less—perhaps, because for too long before this, I was constantly compromising and giving up what I wanted. I've tasted my dreams, and it's become my lifeline, for better or for worse.
    the weight of dreams - by Spencer Chang
  • This sense of destiny and something bigger than yourself is important. Mastery requires a purpose — a solid foundation of your sense of the good life, some degree of felt alignment with it, and the driving belief that the challenge you’re tackling will bring you closer to the ideal you hold. It’s a hunger for the catharsis of fixing something a person feels to be wrong (either inside or outside of themself) coupled with a specific vision for what it would look like for things to be right and the tenacity to do whatever it takes to make that vision real.
    seeing the court clearly - by Molly Mielke - Moth Fund
  • The best way to drive product adoption is through an authentic recommendation from a trusted person.
    Why YourStack failed
  • I resent and reject the idea that everyone in this field needs to move to California to do good work. We need to diversify the kinds of people who make software, the kinds of problems they're solving, and the philosophies and worldviews that govern how they build. Putting every software creator inside a tiny, privileged bubble that's awash in VC cash and obsessed with hierarchical status games isn't going to give us novel, “innovative”, or diverse software that serves the needs of everyone.
    Leaving Elicit
  • “design is a reliable process for identifying and solving problems, then clearly communicating that process to others.”
    Leaving Elicit
  • In the future, Falcon is a world-class IDE (integrated design environment)
    Accelerate design
  • Design once, test everywhere” is the ethos of design computing.
    Accelerate design
  • It was then that I knew that the answer lay in dev tools. I knew code and command lines would have a critical role to play in the future of design tools. With all the clues in hand, the stage had been set. All I had to do now was to put all the pieces together.
    Accelerate design
  • coding in IDEs that lint your code for syntax and errors; hot reload so you can see your designs in real-time; running commands in the terminal to automate tasks or pull packages. Having an idea, being able to build the idea and seeing the idea play out in real time was euphoric.
    Accelerate design
  • The tool has no idea what I’m currently doing or what I’m trying to do. I wish design tools knew that for example, I was designing an iOS app so it could offer me guidance on best practices or correct me on mistakes that I’m making with design conventions. Devtools for example, know which programming language you’re using, and what variables you’re using and as a result will help you correct your mistakes. IDEs have some level of intelligence on what you’re trying to achieve.
    Accelerate design
  • The second issue with design tools is that making changes remains a tedious and rote task, especially when designing at scale. It feels like very little has changed in terms of productivity, since the time I designed the KFC app in Photoshop. Even in Figma, simple changes can take hours.
    Accelerate design
  • The mismatch represents a deep epistemological problem with design tools. The rift isn’t simply a gap between design and code, it’s the gap between the concepts that exist within the design tools and the concepts that exist in real-world software and products.
    Accelerate design
  • When you design there is data, business logic, products, features, states and more. In the tools today, there is no conceptualization of these concepts. Instead, we get projects, pages and layers which don’t map correctly to the primitives of how products and software work and do not map to how engineers think about their code.
    Accelerate design
  • Figma’s AI snafu is a good example of shiny stuff that doesn’t really help with day-to-day work. Many of the issues Figma’s AI was attempting to solve were non-problems for most designers. Their AI seemed to lean more towards replacing the creative aspect of a designer’s work, rather than replacing the mundane aspect of a designer’s work. Had they focused on the latter, Figma AI would have been met with much fanfare.
    Accelerate design
  • what’s going to make my life easier day-to-day? What will provide a leapfrog change in my design outcomes? I want changes that will multiply my creativity and productivity, not incremental changes.
    Accelerate design
  • Figma’s multiplayer feature became a flagship feature and allowed the company to differentiate, achieve product market fit and it grow quickly.
    Accelerate design
  • Photoshop popularized the concepts of editing images through layers and direct manipulation
    Accelerate design
  • it’s Adobe Photoshop that marks the inflection point of what would become a Cambrian explosion of creative digital works.
    Accelerate design
  • we should speak up and demand tools that support thoughtful creation and support the next generation of designers, and not applaud efforts that mindlessly reduce for a dubious promise of convenience.
    Design deserves better than Figma AI
  • In a future where creative tools support creators through AI, we should benefit from model training as an end user more than the company pushing it; my tool should learn from my personal style and process, and help me expand, learn, and take away my annoyances and redundant work. That should be part of my creative output and I should own it as much as the designs I craft. AI should be my production designer, my assistant and helper; not my creative director. Or, worse: an understudy of all, ready to copy your work and process for anyone to use without any of your thoughtfulness and soul.
    Design deserves better than Figma AI