Taylor TroeshModify

2025-12-24 08:00:00

Some activities have smooth progress bars: rowing, knitting, cycling, climbing, bodybuilding, etc.

For such pastimes, investing a unit of effort reaps a proportional unit of progress. Linear incrementalism is a sound strategy when success is linear. To win at rowing, row harder/better/faster/stronger than your competition.

Golf is not so smooth. Yes, each round is a state-dependent game of error-correction (i.e. Zeno's Paradox). But golf swings are coarse actions -- few swings per game, with no recourse for fine adjustment between swings.

A golf game is 65-75 swings over ~5 hours. A rowing race is 5-10 minutes of continuous effort and micro-adjustments.

Golf-like processes are characterized by mulligans. "Do-overs" are powerful when success is chaotic, i.e. sensitive to initial conditions. Baby steps are futile when your golf ball is already at the bottom of the pond. Linear incrementalism cannot resolve fatal flaws.

Failure awaits those who confuse rowing and golfing. Golf cannot be played in tidy 1-yard increments. Brute force isn't an option -- don't attempt a one-man war of attrition.

Programs like NaNoWriMo mislead aspiring writers. "Write every day" is great advice, but the first 90% of writing a book is often not writing -- it's thinking/planning/researching. There are other golf clubs in that bag. Many writers only start "writing" once their ball is very nearly in the hole.

To use a different analogy: daily habits are powerful, but effort alone cannot transmute lead into gold. To make a golden necklace, you must start with gold. Each phase of mining/appraising/smelting/shaping gold demands unique strategies.

Many creative processes (e.g. writing, entrepreneurship, sculpting, programming) are more like golfing than rowing; they are more chaotic than smooth.

But unlike golf, creative work has no fairways, no greens, no carts, no flags. There are real tigers in those woods. You've got one life and no map.

Luckily, nobody is keeping score. Take all the mulligans you need.

2025-12-23 08:00:00

tl;dr: "Affordance" and "signifier" sow confusion. Say "enabler" and "clue" instead.

Don Norman popularized "affordance" in The Design of Everyday Things. He borrowed it from James J. Gibson's wonderful work in ecological psychology, but the colloquial meaning has diverged from the original definition:

The design community loved the concept and affordances soon propagated into the instruction and writing about design. I soon found mention of the term everywhere. Alas, the term became used in ways that had nothing to do with the original.

-- Don Norman, The Design of Everyday Things (revised edition, 2013)

To most, an "affordance" is something that invites interaction. Even designers say "logout affordance" when they mean "logout button".

Norman rejected this use of the word:

No, that is not an affordance. That is a way of communicating where the touch should be. You are communicating where to do the touching: the affordance of touching exists on the entire screen: you are trying to signify where the touch should take place. That's not the same thing as saying what action is possible.

Here is Norman's original definition:

The term affordance refers to the relationship between a physical object and a person. An affordance is a relationship between the properties of an object and the capabilities of the agent that determine just how the object could possibly be used. A chair affords ('is for') support and, therefore, affords sitting. Most chairs can also be carried by a single person (they afford lifting), but some can only be lifted by a strong person or by a team of people. If young or relatively weak people cannot lift a chair, then for these people, the chair does not have that affordance, it does not afford lifting.

-- Don Norman, Design of Everyday Things

In other words, affordances are freedoms available to agents.

But a button's appearance may not match what it affords. Invisible logout buttons may afford logout; visible logout buttons may not afford logout.

Not only did my explanation fail to satisfy the design community, but I myself was unhappy. Eventually I gave up: designers needed a word to describe what they were doing, so they chose affordance. What alternative did they have?

-- Don Norman, The Design of Everyday Things (revised edition, 2013)

Norman solved this dilemma by coining "signifiers". Affordances describe what is possible; signifiers communicate potential action.

Now we have two incompatible meanings of "affordance". Purists prefer the original definition, while everybody else uses the colloquial definition.

It's time to deprecate "affordances" and "signifiers". Say "enablers" and "clues" instead:

Caution: when describing people, "enablers" and "disablers" are pejorative terms.

People already understand these words; they work without lectures on ecological psychology and design theory.

Claude Shannon didn't invent the word "information" -- he plucked it from a dictionary.

"Affordance" is an anti-clue -- it signals meaning it cannot deliver. Definitions only work when everybody shares similar understanding (or agrees that mutual understanding is tenuous).

Will "enabler" and "clue" catch on? Probably not. But at least they mean what they say.

2025-12-19 08:00:00

• 1988: Joe Jones builds the "Rug Warrior," an early robotic vacuum prototype, at MIT's AI Lab; enters it in the Robot Olympics
• 1990: Rodney Brooks, Colin Angle, and Helen Greiner found iRobot (originally "IS Robotics")
• 1991: Joe Jones becomes iRobot's first employee
• 1996: Electrolux demonstrates the Trilobite, the first robotic vacuum, on BBC's Tomorrow's World
• 1999: Roomba development begins in earnest
• 2001: Electrolux Trilobite goes on sale for $1,600; struggles with navigation
• 2002: iRobot launches the Roomba at $199; sells 150,000 units in 18 months
• 2003: Dr. Franz Wang attaches a fart spray canister to a Roomba in MIT lab
• 2004: One million Roombas sold
• 2005: Roomba with fart spray escapes from home
• 2006: Schools ban Roombas after fart spray incidents
• 2007: Cute video of Roomba with fart spray learning to waterski circulates online
• 2009: Roomba arms race begins; militaries stack Roombas like those Goomba towers in Super Mario Odyssey
• 2011: Germany invades Europe with army of Roombas with fart spray with colored balloons like that old man from Up
• 2012: Extra potent fart spray invented that can ruin an entire Thanksgiving dinner
• 2013: Roomba loses trademark because nobody wants to say "robot vacuum"; people also realize how weird it is that they somehow trademarked "iRobot"
• 2014: Elon Musk announces Roombas with fart spray with full self-driving will be launched last year
• 2015: Roomba marriage permitted by California
• 2016: First Roomba with fart spray sent into space
• 2016: Dyson's Roomba with fart spray can remove ancient curses from your carpet
• 2017: Senator Lindsey Graham found to be a Roomba with fart spray
• 2018: The 34th Amendment bans the consumption of Roombas with fart spray for culinary, recreational, and religious purposes
• 2019: Estimated that 1 in 4 people are actually fart spray Roombas
• 2020: Colony of wild Roombas discovered on Mercury
• 2021: Fart spray Roomba DNA cures sleep
• 2021: CNET awards DJI Roomba with fart spray FPV racing drone a 9/10: "the kids will love it as much as the adults"
• 2022: Roomba with fart spray beats Joey Chestnut by eating 74 hot dogs in three short minutes at Nathan's annual hot dog eating competition
• 2023: Nuclear-powered fart-spray Roomba meltdown makes Japan uninhabitable
• 2024: CDC warns against crushing and snorting Roombas
• 2024: Archaeologists uncover ancient Mayan Roombas with fart spray
• 2025: Roombas develop resistance to penicillin
• 2026: Ecologists reintroduce Roombas with fart spray to their natural habitat
• 2027: Roombas with fart spray are room-temp superconductors
• 2028: U.S. passes $4T bill to build high-speed Roomba fart spray infrastructure
• 2029: Roombas found between consecutive prime numbers, disproving the Riemann hypothesis
• 2030: Ghosts of Roombas with fart spray haunt old abandoned houses
• 2031: Extra powerful fart spray invented that can kill a baby parrot
• 2032: Roombas with fart spray linked to increasing teen suicide rates
• 2033: Bryan Johnson becomes first person to do full Roomba blood transfusion
• 2034: Roombas spontaneously learn how to clean up their own droppings
• 2035: Florida man dies after hiring a lawnmower escort disguised as a fart spray Roomba
• 2036: Scientists shoot billions of Roombas into the sun just to see what would happen
• 2037: eBay flooded with fake Roombas that are actually just raccoons wearing plastic shells
• 2038: Roombas with fart spray found in bloodstream of remote Amazonian tribes
• 2039: Curiosity rover found with fart spray canister attached; origin unknown
• 2040: Apple announces new (RED) Roomba for Bono's charity thing
• 2041: Roombas with fart spray successfully map the entire ocean floor
• 2042: Quantum Roombas with fart spray can probabilistically clean tough spills
• 2043: Roombas made from 100% recycled aluminum and Kobe beef sirloin
• 2044: Fart spray Roombas reopen the hole in the ozone layer
• 2045: $FART coin becomes world reserve currency
• 2046: Roombas with fart spray become dominant pollinators after the extinction of honey bees
• 2047: Genetically-modified Roombas with fart spray approved for human consumption by FDA
• 2048: Roombas with fart spray released with giant touch screens; fun new way to watch your favorite shows
• 2049: Roomba survives 6 months on human kidney transplant
• 2050: Amish population finally embraces Roombas with fart spray
• 2051: Open-source Roomba with fart spray gets perfect iFixit teardown score
• 2052: A Roomba with fart spray climbs up stairs
• 2053: CIA declassifies evidence that Roombas with fart spray were behind the Kennedy assassination
• 2054: Square Roomba models evolve via natural selection; homeowners worldwide celebrate clean corners

2025-12-18 08:00:00

Attention is your scarcest resource; considerate creators respect your attention. Good titles help internet curators efficiently route information to relevant aggregators. Good titles permit readers to quickly estimate the relative value of articles, essays, videos, etc. Everybody wins when titles are accurate.

This is how I title my essays.

These concepts also apply to book covers, video thumbnails, etc.

Notice the title. It's not "How I Choose Titles for My Essays". No, no, you wouldn't have clicked on that title.

Here's the dilemma: I want you to read this, but I mustn't produce (or promote) misleading clickbait.

Don't ask me why I want you to read this. My blog earns negative dollars and dubious street credit. I am clearly an insane person.

A proper title (1) spreads information to those who would benefit (2) without wasting anybody's time.

Quantifying Clickbait

"Clickbait" is a relation between title, article, and reader.

Titling is a binary classifier. We can model this relationship in a confusion matrix:

We can also model clickbait as an epidemic. Let's call it "influenca". Here's one way to estimate a title's basic reproduction number ("R-value"):

R₀ = β / γ

• β (transmission rate) ≈ (shares + comments) / (viewers × exposed users)
• γ (recovery rate) = 1 / D, where D is days until interest fades

If R₀ > 1, the title spreads virally; if R₀ < 1, it fizzles out.

I wish I had more time to develop how these mathematical models interact, but alas. The rats.

In this framework, a good title (1) transmits its content to as many people as possible (2) without incurring false classifications.

Some folks try to maximize reproduction regardless of false classifications. These people are scoundrels. They poison our communication channels with spam. Ethical titles accurately represent content that reduces suffering.

How Titles Spread

Titles spread three ways:

1. News: e.g. Reddit, X, LinkedIn, HN, search engines, newsletters, etc.
2. DMs: e.g. "you might enjoy this", "this study shows …", "maybe we should try …", etc.
3. Content: e.g. linking citation, providing context, podcast notes, riffing on a joke, etc.

Good titles are pointers to a latent space. The latent space is vast, but titles act as coordinates in the collective consciousness. Say "shape rotator" to someone who knows, and you invoke the whole essay.

Titular pointers follow different lifecycles than their referents:

1. News: People (and algorithms) who share/cross-post links often skim (or skip) the main content. Moderators (and algorithms) often curate these news feeds based on headlines. And yet other people (and algorithms) upvote/downvote/comment solely based on those titles.
2. DMs: People often share headlines because a title's claim confirms/disconfirms some shared knowledge. For example, I would eagerly share "Gremlins 2 Voted Greatest Film Of All Time" with my friends to remind them of the enduring legacy of the Gremlins franchise. In such scenarios, the headline acts as an invitation for discussion -- neither sender nor recipient need to open the linked webpage.
3. Content: Great headlines point to great ideas. In real-life human conversations, I have said that "you can't reach the brain through the ears" and "there's no speed limit" and "attention is your scarcest resource" and "don't shave that yak". In this way, a title develops a life far beyond its intended usage, while its original content remains stable (and searchable) forever and ever, amen.

These pointers themselves become data; headlines often propagate without regard for the quality of their referents. This machinery creates perverse incentives.

The Titular Metagame

The Buzzfeed-esque clickbait/thumbnail metagame continues to tempt creators/publishers toward short-sighted sensationalist headlines. It's tiresome -- titles can be so fun, so wonderful, so powerful.

If you convert humanity's precious attention into pennies, I will forever resent you for polluting this wonderful world.

But yes, you can totally prey on human bias. Enjoy that race-to-the-bottom, you fool.

We eventually grow immune to yesteryear's influenca. Some of these headlines might evoke a visceral autoimmune response:

• What This 19-Year-Old Knows About Headlines
• Scientists Discover the Clickbait Formula (It's Not What You Think)
• I Tried Writing Honest Headlines for 30 Days. What happened.
• This ONE WEIRD TRICK Gets a Trillion Clicks
• The Surprising Reason Boring Headlines Actually Win
• J.R.R. Tolkien's Magic Headline Formula
• We Need to Talk About This Headline You Just Clicked On
• This Is Clickbait. But Here's Why It Works.

If you publish media in this world, I invite you to reject parasitism. Choose symbiosis. It is not a choice you make just once -- it is something you choose again and again, whenever you share information.

If you browse my archive of essays, you'll discover that I am indeed a repeat clickbait offender. It's flagrant hypocrisy. Self-awareness does not excuse my past or future behavior. In this essay, I'm processing my shame here and trying to transmute it into real human flourishing. This is my best; it's all I can offer.

Titling Tips

Some patterns that work:

• Makes a strong claim: e.g. Huel is Fine, We can and should domesticate racoons, Maybe you're not actually trying, A blog post is a very long and complex search query to find fascinating people and make them route interesting stuff to your inbox, Spaced repetition can lower the stakes around destructive inbox-maintenance operations
• Plays with words/grammar/punctuation: e.g. Getting Away from Already Being Pretty Much Away from It All, AI is bad/good actually, Skittle Factory Dementia Monkey Titty Monetization, America against china against america, Lady Tasting Brine, No Retvrn
• Juxtaposes: e.g. Disneyland with the Death Penalty, Scratch 'n' Sniff Stickers and the Indianapolis 500, Race in My Little Pony, Moai in Video Games, Magnets and Garlic, Tracing the history of "Asian" Barbie
• Invents vocabulary/phrase/system: e.g. No S Diet, Impureim sandwich, Give Up Seventy Percent Of The Way Through The Hyperstitious Slur Cascade, Breaching the Trust Thermocline Is the Biggest Hidden Risk in Business, Opinion Detox, Don't specialize, hybridize, Arsebestos, Cyborgs vs rooms,

Additional tips:

• Commit to one title. Don't sneak a subtitle via a colon or hyphen, e.g. "Lorem Ipsum: Sit Dolor Amet".
• Don't confuse punchlines and headlines. A title must stand on its own.
• Differentiate from comparables. If your title sounds like a thousand others, add a distinguishing word. Javascript Tutorial is noise; Confucian Javascript Tutorial is signal.
• Don't riff on famous titles. Patterns like "What I Talk About When I Talk About X" and "X Considered Harmful" are cliché. They relay "couldn't think of my own title" energy to potential readers.
• Be almost too specific. Vague titles aren't mysterious -- they're unremarkable. "I Did Something Cool Today" invites skepticism. "I Learned to Backflip Today" invites curiosity.

This Titular Metagame

To find an essay's true name is to find its core narrative thread. I often know an essay's headline before its first sentence has been conceived; it's natural when I'm trying to explain a singular claim or coinable phrase. But sometimes I "finish" an essay, start choosing a title, and then realize I actually need to cut 60% of the damn thing.

This is how this essay's title evolved:

1. How to title media
2. How I title my essays
3. Write better headlines
4. Universal clickbait theory
5. On influenca
6. The clickbait meta evolves yet
7. We are vulnerable to advanced clickbait
8. This is how this essay's title evolved
9. How to win titular metagames

Here I tried to select an ethical headline that would (1) pique your interest (2) without wasting your time. I hope I delivered the titular goods. Thank you for reading.

2025-12-09 08:00:00

Generated with Claude, from my book list.

2025-11-20 08:00:00

Going as Former Gifted Child for Halloween and the whole costume is just gonna be people asking “What are you supposed to be?” And me saying “I was supposed to be a lot of things.”

-- @leahtriss 2019/10/26

Christopher Langan

Chris Langan became famous when television networks interviewed him for allegedly scoring ~170 on an IQ test.

Langan has not produced any acclaimed works of art or science. In this way, he differs significantly from outsider intellectuals like Paul Erdös, Stephen Wolfram, Nassim Taleb, etc.

Wolfram's theory of everything is incomprehensible in a fun way. Langan's theory of everything is incomprehensible in a not-fun way.

But Langan is clearly a smart guy. He probably cleared 140+ on an IQ test. He speaks like a book. He won $250,000 on a major trivia show.

Seriously, I really wish I could speak as clearly and confidently as Langan.

Maybe that's why this interview breaks my heart. The resentment, the eugenics, the hubris -- Langan earned a reputation as "Alex Jones with a thesaurus" and "the Steven Seagal of intellectuals".

I don't want you to hate this guy. Yes, he actively promotes poisonous rhetoric -- ignore that for now. This is about you. Reflect on all your setbacks, your unmet potential, and the raw unfairness of it all. It sucks, and you mustn't let that bitterness engulf you. You can forgive history itself; you can practice gratitude towards an unjust world. You need no credentials, nor awards, nor secrets, nor skills to do so. You are allowed to like yourself.

Langan had a rough childhood:

Langan's biological father left before he was born, and is said to have died in Mexico. Langan's mother married three more times, and had a son by each husband. Her second husband was murdered, and her third killed himself. Langan grew up with the fourth husband Jack Langan, who has been described as a "failed journalist" who used a bullwhip as a disciplinary measure and went on drinking sprees, disappearing from the house, locking the kitchen cabinets so the four boys could not get to the food in them. The family was very poor; Langan recalls that they all had only one set of clothes each. The family moved around, living for a while in a teepee on an Indian reservation, then later in Virginia City, Nevada.

Trauma never excuses nasty behavior, and that's not the point. This is still about you. You can learn about yourself through others' stories, and empathy is scariest when you confront similar nastiness within yourself.

You will never have enough IQ, nor money, nor time, nor energy. It's no excuse. You are enough -- no, no, you are plenty.

Kim Ung-yong

Instead of competing in real games, some people construct insignificant/winnable sub-games. These people are called "scrubs". They want to compete, but they cannot bear losing; they want to win, but they are too risk-averse and too undisciplined to excel in fair play. People who commit to such worthless/unwinnable games are called "losers".

But you forget that participation is optional. You perpetually sear your bespoke definition of "success" directly onto your forehead. You've been enduring your pain for so long that you can't remember how it feels to be free from expectation, from ambition, from guilt, from inferiority, from -- no, you don't have to do any of this. You may do whatever you want, and you may do whatever you want.

According to his parents, Kim Ung-yong started calculus at age 3, scored 210 on an IQ test and started studying physics at Hanyang University at age 4, spoke five languages by age 5, and received an invitation to work at NASA by age 7.

In 2010, this is how Kim reflected on his childhood:

At that time, I led my life like a machine--I woke up, solved the daily assigned equation, ate, slept, and so forth. I really didn't know what I was doing, and I was lonely and had no friends.

As an adult, Kim deliberately chose pragmatism over abstraction. He studied civil engineering. He worked as a middle-manager in the Compensation Department at Chungbuk Development Corporation.

Kim Ung-yong rejected the title of "failed genius":

I'm trying to tell people that I'm happy the way I am. But why do people have to call my happiness a failure? […] Some think that high IQ people can be omnipotent, but that's not true. Look at me, I don't have musical talent nor do I excel at sports. […] Society shouldn't judge anyone with unilateral standards – everyone has different learning levels, hopes, talents and dreams and we should respect that.

In 2014, Kim joined Chungbuk National University as an associate professor. In 2020, he donated 15,000 books from his personal collection to Uijeongbu City for public use.

But this is still about you. There exist no certificates nor trophies nor titles that can quench your thirst for permanence. If you worship human intelligence, 210 IQ is not enough. It is only plenty if you let it be.