Apurva Shukla/writing

11-29-25: Internet goodies I read last week

  1. 2025 Things

    • A designer’s not-a-gift-guide: the objects that truly delighted in 2025. Coffee: Origami dripper for even extractions; AeroPress Go for camp. Making: Bambu Lab X1C 3D printer for low-fuss, high-output parts and molds. Carry: Peak Design Outdoor Backpack (lighter than the trusty but heavy GR1), works with Camera Cubes and sling; Heroclip (M) always-on hook. Tools/EDC: Festool Track Saw for small garages; Snow Peak titanium Aurora Bottle; Casio G‑Shock GW5000HS‑1. Wear: Altra Lone Peaks for trails. Food: Jack’s Creek Wagyu from Grand Western; Hasami plates (original texture); POJ Kyoto noren; Monograno Felicetti spaghetti. Home/tech: Starlink Roam (tiny, backpackable) with Owl Vans mount; Home Assistant Yellow hub plus ratgdo for garage freedom. Nice extras: Wildr Goods Hitchhiker Wallet; ZSA Voyager split keyboard for RSI relief; leather dog leashes age beautifully. Quotes: “We are eerily close to Star Trek Replicator territory.”

  2. How I Lost 100 Pounds Using TDT — LessWrong

    • A LessWrong contributor explains how a Timeless Decision Theory mindset helped him finally control overeating. Instead of one-off willpower or elaborate precommitments, he chose as if his current decision would be the same decision in every identical future moment—turning each choice into a policy. This reframing beat hyperbolic discounting and made rule-following automatic. Practical steps: - Cut from three meals to two; no snacks except water. - Weigh daily; ratchet the “baseline” down when below it and tighten rules when above. - Define per‑meal quantity limits and never exceed them. - Deprioritize exercise for weight loss (it improved fitness, not the scale). Results stuck, though excess skin required surgery. He later generalized the method to other domains. Caveat: success leaned on strong willpower, rule affinity, belief, tolerance for temporary hunger, and treating changes as permanent—but some of these can be learned. Quotes: “Would I rather enjoy today the full effect of never having had the knots, in exchange for not having any?”

  3. Sigil on X: “The most capable individuals are learning and shipping 1000x faster using AI. I sat down with Gabriel Petersson (@GabrielPeterss4) - a high school dropout who self-taught Math & ML with AI and is now an AI Researcher at OpenAI. We deep dive into the most important traits to https://t.co/j99nXOBOu4” / X

    • A candid deep dive with Gabriel Petersson—Swedish high-school dropout turned OpenAI researcher—on using AI to learn and ship at breakneck speed. He shares scrappy origins (door-knocking sales, instant A/B tests, couch-hopping while learning to code) and a playbook for mastering hard skills: top-down project goals paired with bottom-up fundamentals, recursive ChatGPT loops (ELI5 → intuition → gap-filling), and turning prompting into a daily reflex. He outlines a research workflow (“shortcuts to foundations”), obsession with feedback and elite code reviews (with AI as reviewer), and practical paths to opportunity: ship public demos, bypass recruiters with proof of work and risk-free trials, and navigate the O-1 visa. Petersson argues America—especially SF—remains the launchpad for ambitious builders. Quote: “Curiosity + Agency + AI will take you further than any legacy credential.”

  4. How to figure out what to do with your life

    • A simple afternoon exercise led the author to stop chasing startups and double down on writing. The method: list and rank your values (Knowledge, Adventure, Fame, Power, Money, Exercising Talent, Human Connection), then mark which projects best satisfy your top ones. Writing won by delivering Human Connection and the skills he most enjoys; Money fell in priority given hedonic adaptation. Use the 80-year-old test (regret minimization): which choice would you most regret not pursuing? Guardrails: prioritize growth (Knowledge) and challenge (Exercising Talent) to sustain long-term motivation; de-emphasize Money once basics are covered; beware social groupthink. Action: run this self-audit regularly and act on who you are today, not yesterday. Quote: "Becoming very rich is appealing to people who've never actually achieved fulfillment."

  5. Memorized Rules: How to give your life direction

    • Adults hoard advice but rarely act on it. Treat advice not as trivia to bookmark, but as behavioral frameworks you drill until automatic. The fix: keep only six “Memorized Rules”—the few principles you can recall and use daily (Miller’s Law). Write them on a sticky note, consult them during tough decisions, and iterate as new rules demonstrably change your behavior. Rules stick when they align both emotionally (fit your lived experience) and logically (make sense from first principles); varied experiences create the “scar tissue” that cements them. Adopting new rules reshapes identity, reduces friction, and turns content consumption into a game of extracting usable principles—so you act today, not someday. Quotes - “Most hard work is a form of laziness. The real hard work is in finding a way to make it easy.” — James Pierce

  6. Vanity metrics: How to pick better life goals

    • We waste huge amounts of time chasing vanity metrics—easy, socially rewarded measures that don’t create the outcomes we want. Use three checks: The Goal Path (optimize for downstream outcomes, not early steps like visits or hours; only fix bottlenecks and 80/20 the rest), The Ability Spectrum (prefer hard-to-game signals like profit or real engagement over credentials, awards, or follower counts), and The Virtue Spectrum (favor high-effort actions that produce change over low-effort posturing). Examples: read to build capability, not to tally books; value friends who show up at 3 am, not a big contact list. The takeaway: define your own metrics, prioritize value creation and meaningful impact, and let output—not optics—do the signaling. Quotes: "Time and resources are always vanity metrics."

  7. Intro | Growth Guide by Demand Curve

    • This guide reframes “growth” as the cross-disciplinary intersection of marketing, sales, and product. It’s not just getting more eyeballs—it’s attracting the right ones, helping sales close them, and shaping product to retain them. The goal: remove blockers to the next stage and build systems that make growth seamless. The authors are practitioners who’ve led growth at companies like Grammarly and Webflow, and through Demand Curve’s agency have worked with Microsoft, TechCrunch, Ancestry, and more. With an average decade of experience and seven founders on the team, they focus on no-fluff, actionable tactics. You’ll learn how to: audit competitors, write high-converting landing pages, create and scale social ads, plug onboarding leaks, run paid efficiently, grow via content, and systematize A/B testing—then decide what’s next. A clear starting point for building a durable growth engine.

  8. Manish Basargekar — I design and build websites. Frontend Developer & Designer

    • A year into running a personal site, the author reflects on what worked, what didn’t, and what’s next. Highlights: ~2,000 global views with a high bounce rate; 279 commits across 21 branches signaling lots of experimentation. Key lesson: prioritize content, simplicity, and usability. A visually fancy, layout-shifting homepage (built with react-grid-layout) hurt maintainability, navigation, and mobile performance—CSS grid would’ve sufficed. Don’t reinvent fundamentals like navbars. The short domain (mnsh.me) is hard to pronounce and fails the “radio test.” Showcasing projects is tricky; a timeline alone doesn’t communicate quality or context. What’s changing: ditching react-grid-layout and the 900px max-width, improving mobile performance, rethinking project presentation, and possibly choosing a new domain. Upcoming: TypeScript, dark mode, a redesigned /projects page, more interactive blog elements, and a stronger design system—evolving the same codebase into “MNSH 3.0.”

  9. bootcamp-monorepo/advice/career at main · fractal-bootcamp/bootcamp-monorepo

    • The Fractal Job Hunt Strategy argues that “spray-and-pray” applications fail because online channels can’t carry enough signal about early-career engineers. Instead, create high-context introductions: warm referrals, in-person networking, and “spearfishing” (build a targeted demo for a specific company). These intros buy attention, not trust; convert it with a crisp story, aligned portfolio projects, a concise portfolio site, and a resume that corroborates—not carries—your narrative. Done well, interviews become due diligence rather than a lottery. Treat the search like a sales pipeline: prospect (targeted companies), create contact (high-context outreach), drive serious interest (interviews), and close (offers); diagnose and fix the stage where drop-off occurs. Quote: “You are not a victim of a mysterious, random market. You are running a pipeline.”

  10. Nasreen Mohamedi - Wikipedia

    • Nasreen Mohamedi (1937–1990) was a pioneering Indian modernist celebrated for austere, luminous line-based drawings, as well as photography and collage. Born in Karachi and educated in London and Paris, she taught at MSU Baroda from 1972 until her death, traveling widely through the Middle East and Asia—absorbing deserts, Islamic architecture, and Zen aesthetics. Mentored by V.S. Gaitonde yet distinct from contemporaries, she forged a disciplined, grid-and-tilt vocabulary often compared to Minimalism and informed by Constructivist/Suprematist thought, Sufi lyricism, and Islamic design. Despite Huntington’s chorea, she sustained extraordinary precision. Long underrecognized, her posthumous acclaim includes solo shows at the Met Breuer, Reina SofĂ­a, Tate Liverpool, and inclusion in documenta; her works reside in MoMA, the Met, the Art Institute of Chicago, and KNMA. Quote: “Again I am reassured by Kandinsky – the need to take from an outer environment and bring it an inner necessity.”

  11. CARI | the Consumer Aesthetics Research Institute

    • CARI (Consumer Aesthetics Research Institute) is an online community working to build a visual lexicon of consumer ephemera spanning from the 1970s to today. Its goal is to document how mass-market design looks and feels across decades—packaging, ads, objects, interfaces—and to develop a shared language that makes these shifts easier to study and discuss. CARI invites anyone interested in design history to participate in researching and refining this collaborative archive, framing the effort as a new medium for cataloging visual culture. The project is community-driven and welcomes support to help sustain ongoing research. If you’re curious about how everyday artifacts capture the aesthetics of their time, CARI offers a focused, collective space to observe, compare, and name the styles that define eras.

  12. The Eight Books That Built My Taste

    • A venture investor shares eight books that shaped her “taste”—discernment, not consumption—and why she pairs philosophy with urban fantasy. Sontag: experience before interpretation. Zafarnama: dignified, exacting resistance. Emerson: the integrity of one’s own mind. Seneca: feel fully, choose wisely. Cicero: decorum links beauty and ethics. Agamben: taste as pre-argument knowledge. Bourdain: work as love and the faith of repetition. Brillat-Savarin: pleasure as intelligence. The throughline: read for texture, not facts; trust strangeness; don’t mistake data for truth. Taste is built by contrast—rigor beside escapism—so you see from new angles. Sure, read business books, but also something useless and difficult; it will change how you see your work. Quotes: “Taste is the gap between what you can explain and what you know.”

  13. Weekly Reading List: 22/11/2025

    • This week’s reading list asks: who gets to tinker with bodies/knowledge/sky; whose futures count as serious history; what happens when the sidelined lead. Pieces span anti–cosmetic surgery, leeches laundered into “science,” midlife neuroscience, and humane productivity—treating bodies as negotiation sites. Zoom out: India’s aviation gap, per-capita myths, predatory journals’ structural roots, and Rokeya’s Sultana’s Dream restoring Indian SF’s lineage. Planet-scale stakes include phosphorus’s fragile cycle, Indian doctors’ wartime solidarity, and a VC-backed bid to dim the sun. From bacteria that “remember” antibiotics to collapsing institutions, the motif is memory: cells, genres, nations, and the atmosphere remember harm while experimentation goes private and consequences stay public. Who’s rewriting the rules—and who remains a footnote? Quotes: - "Meanwhile, the systems we built to forget remember us just fine."

  14. Data-Driven SEO

    • SEO can be a compounding growth engine—unlike paid ads—when you build content around real search intent. This playbook shows how: generate 500–1,000 seed keywords (brainstorm, Google Search Console, competitor analysis, Google Autosuggest), then prioritize by monthly search volume and intent (valuable vs. informational). For each high-opportunity term, inspect the SERP to pick the right content type (product page or educational piece). Draft with data: mine top results for related terms, structure, and media; include the keyword in compelling titles; and ensure completeness. To lift existing assets, use Search Console to find pages with high impressions but ranks 5–15, identify missing topics, update, and republish. Track with Google Analytics, Search Console, and a rank tracker; re-optimize as needed. Results: Optimizely saw +28% impressions, +52% clicks/sessions, and 29% average rank improvement; Springboard grew a post 42% MoM and won a featured snippet.

  15. Michelle Lim on X: “If you’re an engineer starting a company, here’s how I’d recommend you set up your Growth stack (data, SEO, email, etc.) 👇 (1/10)” / X

    • On X, Michelle Lim (@michlimlim) posted the opener to a 10-part thread aimed at engineers launching startups, outlining how to set up a pragmatic growth stack—covering data, SEO, email, and more. Shared on June 15, 2023, the post has accrued 54.3K views, signaling strong interest in no‑fluff tactics for early-stage growth. If you’re wearing both builder and marketer hats, this thread promises a clear starting map; follow the link to read the full series. The rest of the captured page is standard X chrome (login prompts and trending topics), but the takeaway is the resource itself: a concise, engineer‑friendly roadmap for growth foundations that helps technical founders get their analytics, discoverability, and outreach basics in place without overcomplicating the stack.

  16. Alter Magazine | Pitch Us

    • Alter Magazine is inviting pitches at the intersection of science, technology, and progress. If you’d like to write for them, email [email protected] with samples of past work and a short note on why you’re the right person for your proposed piece. They’re especially keen on stories that fit Joel Mokyr’s “useful knowledge” framework—work that advances Propositional knowledge (why things work; scientific understanding) and Prescriptive knowledge (how to do things; engineering practices and recipes). For tone and thematic inspiration, they admire publications like Works in Progress, Asimov Press, Asterisk, Noema, Roads & Kingdoms, Phenomenal World, Inflection Points, Colossus Magazine, the erstwhile Civil Lines, Fifty Two, and essays in the London Review of Books. If you’re exploring progress and practical insight—with clear explanations or actionable know-how—they want to hear from you.

  17. I Became the Rich Friend

    • The author revisits 2022’s “Rich Friend, Poor Friend” from a mid‑career vantage. With more resources, favors feel lighter (spare bedrooms, fronting group costs), friendships deepen—and social circles quietly sort by class, which stings. A newcomer’s example showed how deliberate vulnerability and clear bids for connection can accelerate closeness. Lesson learned: you can’t optimize for “ride‑or‑die” up front; grow good relationships and resilience follows. The uncomfortable twist: becoming the “rich friend” likely improved friendships by reducing neediness and lowering the stakes of openness. Material conditions create slack that makes generosity and intimacy easier—wonderful yet faintly cursed. The piece ends with a hope to extend this “easy mode” to everyone. Quote: “Wealth is playing on easy mode. Wealth removed so many barriers. Wealth feels like cheating.”

  18. after my dad died, we found the love letters

    • After her father’s death, the author finds love letters and meets Edward, the man her closeted father secretly lived with in Canada. Forced into a mismatched marriage and often absent, he had been closeted for decades, yet with Edward he glowed—touring open houses, buying small luxuries, planning to come out and divorce. Edward’s hidden grief contrasts with the family’s restrained rituals: his sprawling shrine, daily songs, and a final goodbye to the cherry-wood urn in the author’s living room. The piece holds two truths at once: compassion for a man denied authenticity and anger at his cowardice and affairs that prolonged her mother’s suffering. It becomes a meditation on filial love, cultural pressure, and the brief flare of late happiness against a lifetime of claustrophobia. Quote: “i call him dad here but i didn't lose a dad, i lost someone who was abstractly a father to me.”

  19. GitHub - mattt/iMCP: A macOS app that provides an MCP server to your Messages, Contacts, Reminders and more

    • iMCP is a macOS (15.3+) menu bar app that bridges your personal data and AI via the Model Context Protocol, with first-class support for Claude Desktop. It exposes local tools for Calendar, Contacts, Location, Maps (search/directions/ETAs/maps), Messages (history), Reminders, and Weather. Setup is simple: install (or brew cask), toggle services and grant per-capability macOS permissions; iMCP itself stores nothing, though your MCP client may transmit data off-device. One click writes the Claude config to launch the bundled imcp-server, after which tools appear in Claude for personalized tasks (e.g., “How’s the weather where I am?”). Under the hood: a Swift app + CLI using stdio transport and Bonjour discovery, returning JSON-LD (Schema.org) objects; Messages access is achieved via sandbox-approved chat.db reads and a custom typedstream decoder (Madrid). Debug with MCP Inspector or the Companion app. Open source (MIT).