Clown Cast

Joey measured two spikes from Miles Partain's volleyball attack—one from the 2023 Montréal Elite16 final, one from 2026's AVP Manhattan Beach. Same swing, three years apart, measured by the same code. But the skeletons were different. This episode reveals the real story: not whether the swing changed, but how the camera's specs dictate which numbers you can extract. 00:01:44 - How enormous the footage gap is 00:09:56 - The camera decides which numbers you get 00:11:29 - Comparing two skeletons from different years --- Sources & further reading: • Volleyball World — Montréal Elite16 2023 men's final (YouTube): https://www.youtube.com/watch?v=B4kzyh5RLy...

Miles Partain's viral kinetic chain analysis claims legs power 50-70 mph volleyball spikes. But what numbers can actually be verified? Joey investigates what data exists for beach volleyball—and discovers that for an Olympic sport, virtually none have been peer-reviewed. The hosts explore how to measure what matters, how cameras lie, and why you shouldn't pretend to measure what you can't. 00:00:00 - Miles Partain's kinetic chain and the fifty-to-seventy mph claim 00:01:41 - The discovery: beach volleyball has virtually no peer-reviewed spike data 00:03:28 - Why ball trajectory matters more than body mechanics 00:05:05 - Do acceleration patterns hold across all skill levels? 00:06:46...

In 2022, the world generated 62 million tonnes of e-waste. That's $91 billion in recoverable metals and minerals—70 times richer in gold than mined ore—dumped in landfills and informal recycling operations. Yet only 22 percent gets properly recycled, and that number's dropping. This episode explores how e-waste became the most valuable trash on the planet while communities that process it bear the cost, and why stopping the problem isn't actually a solution. 00:00:00 - The Scale: 62 Million Tonnes 00:01:43 - The Paradox: Treasure and Poison 00:08:08 - Containment Failure (The Shimmer) 00:11:41 - Why Stopping Isn't a Solution 00:13:12 - Can We Recover the Value? 00:16:15 - Where the Leve...

More people should make better decisions—Condorcet proved it mathematically in 1785. But Ringelmann's rope experiment showed the opposite: eight people pull at half strength. The episode explores how decision fatigue, social loafing, and information pooling create this paradox, and how system design—not individual effort—is the actual fix. 00:01:46 - The Condorcet-Ringelmann Paradox 00:06:51 - Social Loafing and Information Pooling 00:11:57 - Decision Fatigue in Groups 00:15:15 - The Real Takeaway: It's System Design 00:17:00 - Escoffier's Kitchen Brigade Solution --- Sources & further reading: • Ivan Dale Steiner — Group Processes and Productivity (Academic Press, 1972) • Garold Stasser & William Titus — "Pooling of Unshared Information in Group Decision Making: Biased...

For 3,000 years, we've embedded invisible proof into everything—from tattooed scalps to milk-stained paper to printer dots. But every time we perfect the hiding, someone cracks the code. This episode traces the arms race between watermarks and erasure, from ancient steganography and Ovid's spy manuals through WWII espionage to modern AI detection, exploring why the best signature is one that destroys itself when removed. 00:04:37 - The Watermarking Principle 00:06:11 - Velvalee Dickinson's WWII Espionage 00:11:01 - Military Tech and Copyright 00:12:33 - Printer Tracking Dots 00:15:55 - AI-Generated Content Detection 00:17:32 - When Removal Becomes the Signature --- Sources & further reading: • Edward Felten et al. — "Readin...

A toy sports feed built over one tournament's results gets benchmarked on real data. Running against a hosted database, the measurement reveals actual performance numbers and an unexpected source of slowness. Walk through the results, the design decisions that shaped them, and why testing failure on purpose beats meeting it in production. 00:02:01 - Scoring mechanics: relevance, importance, freshness 00:03:31 - Measured performance: ranking 66 cards in 0.21 milliseconds median 00:06:46 - Feed architecture and component design 00:08:21 - Infrastructure lessons from Amazon Builders' Library 00:11:39 - Testing failure deliberately vs. in production 00:14:46 - Data contracts and shape validation --- Sources & further reading: • AWS, Schedule your containers on...

Permafrost is Earth's most effective biological archive, preserving DNA millions of years old in a frozen straitjacket of cold and chemistry. Scientists are racing against climate thaw to decode ancient secrets: a Krestovka mammoth unknown to science, horses from 700,000 years ago, entire family trees hidden in a single tooth. As the planet's deepest archives melt away, we're forced to choose what we can save—and what ancient threats we might awaken. 00:00:00 - The Biggest Library Ever Built 00:01:58 - A Million-Year-Old Tooth Rewrites the Family Tree 00:05:01 - The Outer Wilds Parallel: A Solar System with One Rule 00:08:18 - The Triage Question: Wh...

Joey marks volleyball contacts in AVP broadcasts with a custom tool. When he adds a jump-to-serve review feature, the video freezes at every serve. The same moments play fine on YouTube. What follows is a master class in video encoding: save rooms, seeking mechanics, frame rates, and why new sources break old assumptions. 00:00:00 - The Setup: A Tool That Freezes at Serves 00:01:45 - Video Encoding from Scratch, Taught by One Bug 00:05:05 - Save Rooms: The Hidden Architecture 00:08:21 - Playing Forward Works, but Seeking Fails 00:09:57 - The Fix 00:11:32 - Why New Sources Shatter Old Assumptions 00:13:04 - Teaser: Audio Has the Same...

Dussehra isn't one holiday—it's two mythological victories celebrated simultaneously across different regions of India, and the stories don't compete; they shape-shift like Pokémon forms depending on where you are. This episode unpacks the dual celebration of Rama defeating Ravana and Durga vanquishing Mahishasura, the nine nights of divine transformation, the massive effigies engineered to burn, and the seventy-two-billion-dollar economic engine behind one of the world's biggest festivals. 00:01:32 - Regional Variations and Dual Mythology 00:03:05 - The Effigy: Architecture and Engineering 00:04:59 - Regional Forms and Geographic Shape-Shifting 00:06:48 - The Nine Nights of Durga's Manifestation 00:08:20 - Sacred Geometry: Mantras, Colors, and Off...

Tibetan monks practicing tummo can voluntarily control their autonomic nervous system—raising body temperature enough to dry wet sheets in Himalayan winter. The legend says it's magic. The science says it's neuroscience. Clown Cast digs into what actually happens, where the research comes from, and why this is just advanced interoception: learning to hear your body's signals and send commands back. 00:00:00 - Interoception Recap 00:01:43 - Legend vs. Reality 00:03:25 - How Heat Generation Works 00:04:57 - The Dramatic Story Elements 00:06:44 - Hunter x Hunter: Everyone Has the System 00:08:24 - Beyond Conscious Control 00:09:56 - The Bene Gesserit Comparison 00:11:32 - Real Measured Data 00:13:20 - Th...

Knowledge workers spend 25 hours per week in meetings and meeting prep. Most meetings end by scheduling another meeting. We dig into why companies nuke their calendars, what happens when they do, and when ceremony might be the thing saving your organization. 00:00:00 - The Meeting Crisis 00:03:01 - Stripping Down the Ritual 00:04:32 - The Calendar Purge 00:09:11 - Mode Mismatch: Doctrinal vs. Adaptive 00:11:03 - Toyota vs. GM: Why Ritual Matters 00:14:20 - Minimum Viable Ceremony 00:17:55 - Why Meetings Come Back --- Sources & further reading: • Harvey Whitehouse — Modes of Religiosity: A Cognitive Theory of Religious Transmission (2004) — (book) • James Shore — "Cargo Cult Agile" (2008): https://www.jamesshore.com/v2/blo...

When quality certifications are born, they're consumer protection—a stamp saying "this is real." But over centuries, from Roman silver to modern craft beer, we watch the same cycle repeat: protection becomes prestige, prestige becomes price, and the people the mark was made to shield end up priced out of authenticity itself. This episode traces how the gentrification of authenticity works, and what happens when the fence goes up around something that used to be free. 00:00:00 - Opening: When labels became status symbols 00:01:37 - The history of hallmarks (Roman and medieval) 00:06:53 - How certification creates artificial scarcity 00:10:02 - The heroic or...

June 2009: Analog TV broadcast shut down nationwide. Congress funded $3 billion in converter boxes; broadcasters spent 10-16 billion upgrading. Digital is genuinely better—but then the government auctioned the now-free spectrum for $20 billion. Net profit: $17 billion. Not an upgrade. A $3 billion extraction cost for a $20 billion land sale. This episode traces that pattern from 19th-century toll roads through FM radio to modern ownership claims: how infrastructure improvements mask wealth extraction. Timestamps: 00:00:00 - The infrastructure paradox 00:01:40 - The numbers: $3B spent, $20B profited from spectrum 00:03:10 - The historical playbook: toll roads in the 1830s 00:08:08 - The next frontier: targeted ads on broadcast si...

Clown Cast answers Olympic beach volleyball player Miles Partain's videos on the kinetic chain: hip drive, the lead-leg block, and hips turning ahead of the shoulders. Then the same idea in the air, from Joey's time as a flying trapeze catcher in the FSU Flying High Circus, and a physics model built this week. Partain wasn't involved; go watch the originals: Armswing Kinetic Chain: https://www.youtube.com/watch?v=QYL47HMqGVY Rib flare in the setting kinetic chain: https://www.youtube.com/watch?v=jY4Ni6ClbKg Integrated vs Disintegrated Movement: https://www.youtube.com/watch?v=EMNzRhuj6Zs 00:01:35...

Seventy-five percent of silent films made before 1930 are gone. Eighty-six percent of pre-1965 recordings have vanished. Not from decay alone, but from deliberate choices: studios melted nitrate film for silver, platforms deleted archives, and only economically-viable content got migrated forward. This episode examines how recording format choice and infrastructure ownership act as invisible filters, determining which parts of human culture survive to the next century. 00:01:32 - Why studios melted movies for silver 00:05:08 - People didn't choose what got recorded 00:10:13 - Format and economics as invisible filters 00:13:38 - A solution that was built but not funded --- Sources & further reading: • David Pi...

Nine episodes into recommender systems and it's finally time to answer the phone. When a user opens the app, a single request arrives at the gateway, bounces through a BFF (Backend for Frontend), and fans out to multiple services. But none of the architecture matters if it doesn't arrive in time—and the core tension is simple: a recommender that's correct but slow loses to a popular list that shows up fast. Today's episode explores the serving layer, fallback strategies, why the user staring at a spinner doesn't care about accuracy, and what the AWS picture looks like. 00:00:00 - Starting wi...

When Spotify's language model faithfully narrated the wrong number of artists discovered, it revealed a deeper problem: guardrails need to sit upstream of the model, not downstream. This episode maps the four specific places where large language models actually fit inside production recommender systems—and all four share identical problems: they're expensive at scale, and their output always requires external validation. Joey walks through the architecture, constraint by constraint, with research-backed choices and an honest accounting of what he hasn't built himself. Key timestamps: 00:00:00 - Spotify Wrapped and the confidently wrong bug 00:01:30 - Joey's honesty check: five years of backend wo...

A team from EPFL discovered something every recommender builder should fear: the model that performed best in offline testing was the worst performer live. This episode breaks down why offline metrics deceive, and covers three evaluation approaches—offline metrics, live A/B tests, and off-policy evaluation—that reveal whether your system actually works. 00:00:00 - The Paradox: Best Offline, Worst Live 00:01:47 - Starting with Offline Metrics 00:07:10 - How Offline Evaluation Fails 00:15:35 - Three Evaluation Approaches 00:17:20 - Why Procedure Matters --- Sources & further reading: • Garcin, Faltings, Donatsch, Alazzawi, Bruttin, Huber, Offline and Online Evaluation of News Recommender Systems at swissinfo.ch, RecSys 2014 — — offline/on...

Episode six of our recommender deep-dive leaves the algorithm's brain and explores its plumbing. When a basketball game ends, three systems fire—editorial publishes a recap, video publishes highlights, scoring publishes the final score. Each speaks a different schema. How do they become one unified event the feed can rank? Through data contracts: written promises between teams about what data flows and when. Joey explores the mechanics using Earthsea's concept of true names—and what breaks when those contracts fail. The recommender doesn't crash. It quietly learns the wrong thing. And that's worse. 00:00:00 - The Plumbing Problem: Three Systems, One Even...

Recommender systems work like animal trackers—reading your behavior to guess what you want next. But a click is the noisiest signal; six clicks on a rival's drama doesn't mean betrayal. In this episode (part 5 of our recommender series), we explore the signals that feed user models, why dwell time matters more than clicks, what silence reveals, and the cold start problem that haunts every new user. 00:01:42 - Clicks: the cheap, noisy signal 00:06:30 - Dwell time: why staying matters 00:08:01 - Negative signals: mutes, hides, and what they mean 00:13:03 - Cold start: learning from nothing --- Sources & further reading: • Yifan Hu, Yehu...

A ranker scores each item individually. But you experience a page. In part four of the recommender series, Joey explores the re-ranker—the policy engine that handles diversity, freshness, deduplication, and editorial overrides. It's the difference between individual station cooks and the expediter at the pass. [00:00:00] YouTube's three failed fixes. [00:01:35] The ranker grades the exam, the re-ranker arranges the seating chart. [00:06:28] Why YouTube's best result never shipped. [00:11:25] How hand rules blocked better ML models. [00:17:27] The system becomes a curator. --- Sources & further reading: • Mark Wilhelm, Ajith Ramanathan, Alexander Bonomo, Sagar Jain, Ed H. Chi, Jennifer • Jaime Carbonell, Jade Goldstein, *The Use of...

What does a score of 0.83 mean? It's not relevance—it's a predicted probability of an action you might take. In episode 381 of the recommender systems series, Joey explores ranking: how systems order hundreds of candidates to show you the top few dozen. He works through published papers from Google and YouTube, learning how the third layer of recommendations actually works. Key moments: 00:00:00 - What is a score? 00:01:48 - From millions to hundreds: retrieval vs. ranking 00:05:19 - Hand-built cross features 00:06:57 - Features matter more than tuning 00:10:06 - YouTube's pointwise ranking approach 00:11:43 - Measuring a better thing 00:13:21 - Judging execution: the diving an...

In a feed with thousands of articles and videos, your phone has milliseconds to surface the few things worth your attention. Candidate generation is the invisible first filter that narrows chaos to choice before anything gets ranked. This episode unpacks matrix factorization, embeddings, and the matchmaking metaphor behind how recommender systems find needles in haystacks at billion-scale speed. 00:00:00 - The Funnel Explained: Candidate Generation and the Millisecond Problem 00:01:42 - The Mechanism: Matrix Factorization and Latent Spaces 00:03:18 - Profiles as Embeddings: The Go-Between 00:05:26 - Latent Factor Space: The Takeaway 00:07:09 - The Omiai Metaphor: How Profile Preparation Works 00:08:53 - Batch Processing: Why...

Every recommender system that works at scale has the same shape: a funnel. Joey explores YouTube's 2016 paper on recommendations—the paper that switched YouTube from optimizing for clicks to watch time—and breaks down the funnel architecture that lets systems rank millions of items in milliseconds. This is episode one of Learning Recommender Systems, a series on how recommendation engines actually function. 00:00:00 - YouTube's 2016 recommendation paper 00:01:32 - The core problem: ranking millions in milliseconds 00:03:03 - The funnel structure 00:07:45 - What recommenders learn from 00:11:15 - Walking through a user's complete journey 00:13:01 - How scrolls become tomorrow's training data --- Sources & further read...

When Glen Bell watched Mitla Cafe owners make tacos and then opened Taco Bell across the street, he didn't invent the recipe — he copied it. Today, Taco Bell makes $17.9 billion annually while Mitla Cafe doesn't. This episode examines who profits when immigrant food goes mainstream, using metaphors from farming games to mining economics to explain why cultural innovation doesn't translate to wealth for the people who invented it. Key moments: 00:00:00 - The Simple Question: Who Profits From Immigrant Food? 00:01:37 - The Pierre Problem: Replicating Success Without Credit 00:05:12 - The Numbers: Three Chains vs. An Entire Market 00:06:43 - The Slag: Understanding By...

When colonial powers target rites of passage, they don't just break one thing—they dismantle an entire culture's pathway to belonging. In this episode, we examine how colonialism deployed a systematic playbook to erase incorporation structures: removing initiates, banning ceremonies, confiscating sacred objects, punishing knowledge-holders, and replacing language. We trace the precision of this destruction, its generational echoes, and the painstaking work of communities recovering what was deliberately erased. 00:00:00 - Incorporation as the load-bearing wall 00:01:55 - The playbook: Five fronts of systematic destruction 00:05:16 - Generational harm and its long arc 00:09:00 - Stormlight Archive: Metaphors for forced transformation 00:14:18 - The uncomfortable qu...

After exploring institutions that trap people in permanent liminality, this episode examines what genuine rites of passage look like when they actually work—and why incorporation, the phase after the trial, is the load-bearing wall of the entire structure. Van Gennep mapped the three phases in 1909, but most modern rites fail because they skip incorporation: the community's agreement to recognize the initiate's new status and new obligations. This episode traces why that contract matters, what broken incorporation looks like, and how real traditions complete the ritual. 00:00:00 - Opening: Liminal Institutions That Trap People 00:03:10 - The Reciprocal Obligation: When Community Agrees to...

Clown Cast unpacks the rite of passage architecture that institutions—from cults to MLMs to secret societies—weaponize to trap people in permanent liminal space. The episode traces how van Gennep's three-stage ritual structure (separation, transition, incorporation) gets perverted when the middle stage never ends, keeping people disoriented, bonded, and dependent. Examples range from medieval guilds to modern online communities to NXIVM's explicit color-coded initiation system. The pattern is consistent: they're selling a door, not a room. Key Moments: 00:00:00 - What is a doorway? 00:01:34 - The structure of rites of passage 00:06:25 - When the community becomes the product 00:11:10 - NXIVM's colo...

Joey assumed UniFi was another AI product—it's not. It's Ubiquiti's networking platform. But his dad, a project manager at a small university, is watching a structural risk repeat: the school just lost its longtime IT expert, and now the entire network and new custom AI-built applications are consolidating under a single new tech lead. The UniFi switch makes sense on the merits, but it recreates the exact key-person risk the school just escaped. This episode explores why that pattern is the real story—and how to actually break it. Key moments: 00:00:00 - UniFi, the network tool (not the yarn comp...

Indigenous communities are building their own AI systems—not waiting for Big Tech permission, but writing their own rules and encoding sovereignty directly into the architecture. This episode traces how extraction patterns (seeds, audio, traditional knowledge) get automated into AI training pipelines, then explores the frameworks communities are using to fight back: FAIR and CARE principles, models borrowed from centuries of indigenous water governance, and design choices that make data access a structural constraint rather than a policy afterthought. Sovereignty isn't a document—it's architecture. 00:00:00 - Opening: The Extraction Pattern 00:07:02 - FAIR and CARE: What Data Can Do, Who It Come...

In 1995, the University of Mississippi patented using turmeric for wound healing—a practice documented in Ayurvedic texts for thousands of years. India challenged it with ancient Sanskrit texts and won. Today, the problem is bigger: AI systems train on centuries of traditional knowledge from indigenous and local communities without consent or compensation. We explore how biopiracy has evolved from patent offices to AI training datasets, why current protections fail, and what addressing it would actually take. 00:01:36 - The turmeric patent reversal 00:03:09 - Benefit-sharing agreements that don't work 00:06:47 - Why AI sidesteps bilateral agreements 00:09:54 - How trade secrets become the new lo...

Norms work through reputation and feedback loops. But AI training datasets operate at billion-page scale, copying faster than any community norm can propagate. This episode traces the mechanical failure: when the thing doing the copying moves faster than the norm itself can even travel. From Common Crawl to licensing markets, the engineering problem nobody's solved. 00:00:00 - Three episodes, one question 00:01:36 - Two billion pages in a single month 00:03:09 - Industrial agriculture and the pattern 00:04:45 - Copying that cannot interact with norms 00:06:23 - The cost of scale: one and a half billion dollars 00:07:56 - Commons enclosed, seeds patented 00:09:38 - When norms...

Folk musicians, jazz players, stand-up comedians, magicians, chefs, and game designers all work in creative fields where copyright is almost irrelevant. Instead, each community built its own unwritten legal system—rules about what copying is legitimate and what's theft, enforced by reputation and shaming rather than courts. Scholar Elizabeth Rosenblatt calls this 'IP's negative space.' We trace five established communities (folk, jazz, magic, comedy, cuisine) and their different boundaries, then land on game design: a sixth community still drawing its lines right now as sampling technology makes possible what hip-hop producers figured out decades ago. 00:00:00 - The question: what do...

When YouTuber Big A published a video showing games running inside other games, Joey took a different approach: he cherry-picked his favorite mechanics from two Jackbox games and rebuilt them for a house party. This episode unpacks Big A's argument, explores why Joey's selective remix works differently, and examines what happens when you build something for the specific people in your room rather than a general audience. Joey built the pipeline that powers this show, picked this topic, and verified the research against his own repository. 00:00:00 - Big A's viral video and what Joey did instead 00:03:13 - Big A's concerns...

Four games, four hours, one payoff. Joey designed them so every experience builds toward a final tournament where your drawings become fighters you bet real in-game currency on. This episode explores how game continuity works, why the Hades pattern matters for party design, and what it means when games remember you. 00:01:31 - The Design Solution: Games Building Toward One Finale 00:06:17 - Joey's Philosophy: Making Games Remember the Player 00:09:22 - Technical Approach: Learning from Hades and Meta-Progression 00:11:09 - The Through Line: Why Continuity Changes Everything This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Pi...

Guests draw stick figures on their phones. Joey's smart home instantly transforms them into real D&D 5.2 fighters—AI generates personality and backstory, code executes every stat and die roll. They battle live in a tournament bracket. But after Joey saw the fights run long and repetitive, he rebuilt the entire combat system. This episode explores the design split between creation and execution, and the real math problems that appeared when a monk did cardio for nine rounds straight. 00:00:00 - The Arena: Phone Doodles Enter Battle 00:01:39 - Character Creation: Sketch to Fighter 00:03:10 - Improvised Dialogue & Personality 00:06:20 - The Pacing Problem: Wh...

Joey built a smart home system that throws a birthday party entirely on its own—four hours across four rooms with different themes, all orchestrated by a TV. The catch: one wrong button press could break the whole thing. When the remote had six overlapping controls, Joey did what he always does: cut it down to one button. What looks like a simple fix reveals a deeper principle about interface design and what makes an automated system actually work. 00:00:00 - The automated party system 00:01:44 - The button problem and the solution 00:07:57 - Why automation doesn't know what the moment is fo...

Your digital twin—a mathematical model built from your bloodwork, imaging, genome, and wearables—is a copy of you in perfect fidelity. Except it's not. It's a partial construction derived from technical choices and selected datasets, governed by institutions that profit when you're marked 'fine.' What happens when hospitals trust the simulation more than the person it's simulating, especially when that person says they're in pain? Key moments: 00:00:00 - The foundational question: If your digital twin says you're fine but you're in pain, who wins? 00:01:44 - From Apollo spacecraft to human bodies: the unexpected history of digital twins 00:05:04 - When...

Some nations have built themselves around exporting what they know—Israel in cybersecurity, India in software engineering, the Netherlands in flood control. But selling expertise is different from selling oil or steel. This episode explores the economics, geopolitical risks, and the cultural roots that let countries turn mastery into a primary product. 00:00:00 - Opening: Expertise as National Export 00:01:35 - India's $221B IT Export Challenge 00:04:48 - The Terroir Metaphor: Why You Can't Copy Culture 00:08:44 - Seven People, $250, and $221 Billion 00:10:21 - Israel's Unit 8200: The Root of Cybersecurity Dominance 00:13:57 - The Resource Curse 00:15:38 - Usefulness as Survival Strategy --- Sources & further reading: • Spherical Insi...

Loneliness is killing Americans—the Surgeon General warns it's as deadly as smoking a pack a day, and 42% of Americans now feel lonely weekly. With third spaces closing and digital replacements failing, the question becomes: who actually brings people together? This episode explores the craft of hosting: why most parties fail, and what potlucks, randomness, and Dungeons & Dragons reveal about the secret mechanics of gathering. 00:00:00 - The loneliness crisis 00:01:30 - Somebody's gotta throw the party 00:03:12 - What separates good gatherings from bad ones 00:04:48 - The hidden burden of being the host 00:06:27 - Why potlucks work (the pharmacology of shared stakes) 00:08:05...