Having spent some time working on aging research before transitioning to biotech, this paper genuinely excited me in a way that few studies have recently. The longevity field is littered with compounds that work beautifully in vitro but fail spectacularly in vivo - seeing psilocybin deliver both cellular lifespan extension AND improved survival in aged mice is remarkable.
What strikes me most is the mechanistic plausibility. The SIRT1 upregulation, reduced oxidative stress, and telomere preservation they observed align perfectly with what we know about successful aging interventions. When I was working on senescence research, we’d get compounds that would extend cellular lifespan by targeting one pathway, but they’d often have off-target effects that negated benefits in whole organisms.
The dosing strategy here is particularly clever - starting with 5mg/kg for acclimation, then monthly 15mg/kg treatments. That mirrors what we’re seeing in clinical trials for depression, but applied to aging. I remember being skeptical when colleagues first suggested psychedelics might have systemic anti-aging effects beyond their neurological benefits, but the serotonin receptor distribution throughout the body makes this increasingly plausible.
The survival curve (80% vs 50%) is the kind of effect size that gets my attention. In aging research, we’re usually thrilled with 10-20% lifespan extension. But starting treatment at 19 months (equivalent to 60-65 human years) makes this especially compelling - most aging interventions need to start early in life to be effective.
My main concern is the limited exploration of potential downsides. Delayed senescence can be a double-edged sword - those cells that keep proliferating longer might accumulate DNA damage that wasn’t detected in their short-term assays. We need much longer studies to understand cancer risk.
Still, given psilocybin’s remarkable safety profile and the FDA’s breakthrough therapy designation, this opens fascinating possibilities for combining psychedelic therapy with longevity medicine. Imagine treating both the psychological burden of aging and its biological mechanisms simultaneously.
While I think the title is misleading/clickbaity (no surprise given the buzzfeed connection), I'll say that the substance of the article might be one of the most honest take on LLMs I've seen from someone who actually works in the field. The author describes exactly how I use LLMs - strategically, for specific tasks where they add value, not as a replacement for actual thinking.
What resonated most was the distinction between knowing when to force the square peg through the round hole vs. when precision matters. I've found LLMs incredibly useful for generating regex (who hasn't?) and solving specific coding problems with unusual constraints, but nearly useless for my data visualization work.
The part about using Claude to generate simulated HN criticism of drafts is brilliant - getting perspective without the usual "this is amazing!" LLM nonsense. That's the kind of creative tool use that actually leverages what these models are good at.
I'm skeptical about the author's optimism regarding open-source models though. While Qwen3 and DeepSeek are impressive, the infrastructure costs for running these at scale remain prohibitive for most use cases. The economics still don't work.
What's refreshing is how the author avoids both the "AGI will replace us all" hysteria and the "LLMs are useless toys" dismissiveness. They're just tools - sometimes useful, sometimes not, always imperfect.
I've been on both sides of this war: the test evangelist fighting for coverage and the pragmatist shipping to beat a deadline. After 20+ years in software, the truth is painfully obvious: testing is the greatest productivity hack that everyone keeps "postponing until next sprint."
The author gets the psychology exactly right. We overestimate the initial cost and drastically undervalue the compound returns. What they call "Time Technical Debt" is the perfect description for that sinking feeling when you're working on a mature codebase with spotty test coverage.
The most insightful point is how testing fundamentally changes your design for the better. When you have to make something testable, you're forced to:
- Think about clear interfaces
- Handle edge cases explicitly
- Create clean separation of concerns
- Build proper startup/shutdown sequences
These aren't "testing best practices," they're just good engineering. Testing is simply the pressure that forces you to do it right.
My experience: if your system is hard to test, it's probably hard to reason about, hard to maintain, and hard to extend. The difficulty in testing is a symptom, not the disease.
At my last company, we built a graph of when outages occurred versus test coverage by service. The correlation was so obvious it became our most effective tool for convincing management to allocate time for testing.
I can’t stop thinking about this article. I spent a long time in ad tech before switching to broader systems engineering. The author captures something I've struggled to articulate to friends and family about why I left the industry.
The part that really struck me was framing advertising and propaganda as essentially the same mechanism - just with different masters. Having built targeting systems myself, this rings painfully true. The mechanical difference between getting someone to buy sneakers versus vote for a candidate is surprisingly small.
What's frustrating is how the tech community keeps treating the symptoms while ignoring the disease. We debate content moderation policies and algorithmic transparency, but rarely question the underlying attention marketplace that makes manipulation profitable in the first place.
The uncomfortable truth: most of us in tech understand that today's advertising systems are fundamentally parasitic. We've built something that converts human attention into money with increasingly terrifying efficiency, but we're all trapped in a prisoner's dilemma where nobody can unilaterally disarm.
Try this thought experiment from the article - imagine a world without advertising. Products would still exist. Commerce would still happen. Information would still flow. We'd just be freed from the increasingly sophisticated machinery designed to override our decision-making.
Is this proposal radical? Absolutely. But sometimes the Overton window needs a sledgehammer.
P.S. If you are curious about the relationship between Sigmund Freud, propaganda, and the origins of the ad industry, check out the documentary “Century of the Self”.
As someone who's worked in marketing for 15 years - across big agencies in New York and running growth for startups - there's an uncomfortable truth to this piece. The industry has quietly become something darker than when I joined.
Modern advertising doesn't just sell products, it sells our own attention back to us at a premium. What started as "connecting products to people who need them" has warped into engineering digital environments that hack our baseline neurological responses.
The most disturbing part is that most people inside the machine know it. I've sat in rooms where we've explicitly designed systems to maximize "time on site" by exploiting cognitive vulnerabilities. The language we use internally is more clinical than predatory, which makes it easy to avoid moral questions.
What's wild is how this piece frames advertising as a relatively recent phenomenon. It's right - for 99% of human history, we made purchasing decisions based on community knowledge and direct information, not carefully engineered psychological triggers that follow us around.
Sure, banning all advertising sounds extreme, but it's worth asking: what would we actually lose? Product information would still exist. Reviews would still exist. Word-of-mouth would still exist. We'd just lose the sophisticated machinery designed to bypass our rational decision-making.
The "free speech" counter-argument has always struck me as disingenuous. Nobody believes they have a constitutional right to interrupt your dinner with a telemarketing call.
I built something similar a few years ago, but using a Pi instead of an Arduino. Reading through the poster's calculations brought back memories of debugging my own orbital mechanics code at 2am while questioning all my life choices.
What impresses me most here isn't just the technical achievement, but how the project transforms abstract data into physical movement. We've become so accustomed to viewing everything through screens that there's something deeply satisfying about a physical object that silently points to a spacecraft traveling 28,000 km/h overhead with humans inside.
The mathematical transformation from TLE to actual motor positioning is non-trivial - converting between reference frames always seems straightforward on paper until you're debugging at 3am wondering why your arrow is pointing at the ground. I appreciate that they shared actual code rather than glossing over the hard parts.
For anyone wanting to build this: from my experience, the hardest part is actually the mechanical design. Getting smooth movement without backlash in consumer-grade steppers and servos takes patience. The 28BYJ-48 is a good budget choice but has noticeable "slop" in its gearing, which can make precise pointing challenging.
This reminds me of those old mechanical orreries that modeled the solar system. There's something profoundly human about building machines that help us comprehend the cosmos.
I've been following Voyager since the late 70s when my dad (who worked at Ames) would bring home printouts of the Jupiter images. This interview captures something profound about these missions that non-space folks often miss.
The most striking part isn't the technical achievement of keeping 1970s hardware alive for half a century. It's the human infrastructure behind it - the institutional knowledge passed through generations of engineers like some ancient priesthood maintaining a temple.
Think about it: We're teaching 2020s engineers to understand machine code written before MS-DOS existed, working with 50 year old documentation on yellowing paper, and bringing retirees back to decipher systems they built when Nixon was president.
Our space exploration legacy is being maintained by the engineering equivalent of oral history. This is why we need consistent funding for these long-term missions - the value isn't just in the data, but in maintaining the unbroken chain of knowledge transfer that makes missions like this possible.
The sticky notes on diagrams trying to revive a silent probe 15 billion miles away just broke me. This is humanity at its finest.
I've spent the last decade watching this arms race between interviewers and candidates. Last month I hired a senior dev who couldn't implement a basic database migration when we brought him on but aced our interview problems. Turned out he'd been using tools like this.
The problem isn't the tools - they're inevitable. The problem is that our industry clings to this bizarre ritual where we test for skills that are completely orthogonal to the actual job.
My current team scrapped the algorithmic questions entirely. We now do pair programming on a small feature in our actual codebase, with full access to Google/docs/AI. The only restriction is we watch how they work. This approach has dramatically improved our hit rate on good hires.
What I care about is: Can they reason through a problem? Do they ask good clarifying questions? Can they navigate unfamiliar code? Do they know when to use tools vs when to think?
These "invisible AI" tools aren't destroying technical interviews - they're just exposing how broken they already were.
What makes this fascinating isn't just what it says about storytelling, but what it reveals about our relationship with truth in media. I worked in public radio for 7 years, and TAL's influence was impossible to overstate - every producer wanted to craft stories with that perfect narrative arc.
The Daisey episode still haunts journalism programs. We used it as a case study in our ethics workshops. The truly unsettling part wasn't just Daisey's fabrications, but how perfectly those lies fit into TAL's storytelling template - dramatic scenes, sympathetic characters, narrative tension, and a tidy resolution that makes you feel something.
Glass wasn't wrong about storytelling's power to make people listen. But the Daisey incident showed its dangers - when your format rewards emotional impact and narrative elegance, you create incentives for sources to deliver exactly that, truthful or not.
The saddest part is that real stories about Foxconn's labor conditions existed that could have been told without fabrication. But they wouldn't have had that perfect "old man touching an iPad for the first time" moment that makes for such a perfect radio beat.
The author did a fantastic job explaining complex cryptographic concepts in accessible terms. I particularly appreciate how it builds up the problem space before introducing solutions.
I spent several years implementing TLS in embedded systems, and I've never seen the intuition behind authenticated key exchange explained this clearly. The author manages to capture both the "why" and the "how" without getting bogged down in mathematical notation.
The explanation of the man-in-the-middle vulnerability was especially well-done - most articles either skip over this crucial weakness or explain it so technically that newcomers get lost. The starbucks router example makes it immediately obvious why key authenticity matters.
One tiny nitpick: the article simplifies by saying "you can't know you're talking to someone if you don't already know who they are" - but this overlooks the role of certificate transparency logs and OCSP, which add significant layers of protection beyond just trusting certificate authorities.
For anyone wanting to learn more, I'd second the recommendation of Julia Evans' materials. Her zines explain these concepts visually in ways that make this stuff click for many engineers.
The tragic irony of Objectivism is perfectly captured in Peikoff's life story. A man who dedicated himself to a philosophy of radical independence ended up defining his entire existence through dependency. First on Rand, now apparently on his caregiver-turned-wife.
I met Peikoff at an ARI event in 2009. He was surprisingly warm in person, but you could see the weight of being "the heir" in how defensively he responded to even mild questions about Rand's work. Now reading about the fracture with his daughter over the estate, it's like watching Atlas Shrugged's plot play out in real life: the bitter disputes over Rand's intellectual property mirroring the novel's battles over physical resources.
What's most disturbing isn't the personal drama but what this reveals about how Objectivism operates in practice. For a philosophy obsessed with reason and independence, its institutional guardians seem remarkably focused on excommunication, loyalty tests, and controlling access to primary sources. The gap between preaching individualism while demanding conformity has always been the movement's central contradiction.
I've spent the last six months building a coding agent at work, and the reliability issues are killing us. Our users don't want 'superhuman' results 10% of the time - they want predictable behavior they can trust.
When we tried the 'full agent' approach (letting it roam freely through our codebase), we ended up with some impressive demos but constant production incidents. We've since pivoted to more constrained workflows with human checkpoints, and while less flashy, user satisfaction has gone way up.
The Cursor wipeout incident is a perfect example. It's not about blaming users who don't understand git - it's about tools that should know better. When I hand my code to another developer, they understand the implied contract of 'don't delete all my shit without asking.' Why should AI get a pass?
Reliable > clever. It's the difference between a senior engineer who delivers consistently and a junior who occasionally writes brilliant code but breaks the build every other week."
I didn't say that intellectually honest content bubbles to the top, I said: "I think the algorithm is working exactly as intended if it's designed to limit intellectually dishonest content". There's a difference.
As someone who's been reading Daring Fireball since 2004 and considers themselves an Apple fan, I think the algorithm is working exactly as intended if it's designed to limit intellectually dishonest content.
Gruber has built a career on a predictable pattern: vociferously defend Apple's every decision (even contradicting his own previous positions when Apple changes course), construct elaborate post-rationalizations for their missteps, while simultaneously maintaining meticulous, years-long grudges against anyone who makes incorrect predictions about Apple.
There's a stark difference between having perspective as an enthusiast and being a reflexive apologist. The "Something Rotten in Cupertino" piece is the exception that proves the rule - a rare deviation that doesn't erase the pattern of selective criticism that's defined his work for years.
What's particularly frustrating is the pretense of even-handedness. I'd respect the work more if it were openly presented as Apple advocacy rather than positioned as independent analysis. The community's collective flagging behavior isn't "censorship" - it's quality control from readers who've recognized this pattern.
HN's algorithm isn't suppressing contrarian viewpoints - it's responding to content that consistently fails to meet the intellectual honesty this community values.
Two follow-ups for you:
1. Is this response some kind of attempt to excuse hostile UX?
2. Because reader mode exists, does that mean brutally bad UX should never be called out?