The Ten-Dollar Wall

The Fires of History Newsletter, No. 5

In 2001, a twenty-five-year-old web developer named Richard “Lowtax” Kyanka faced a problem that would have been familiar to any medieval guildmaster: the barbarians were at the gate, and they were not paying for the privilege.

Kyanka ran Something Awful, a humor website with a growing forum community. Internet advertisers were deadbeats who did not pay their invoices. The site was hemorrhaging money. His solution was not a paywall, not a subscription model, not a freemium tier with premium features. His solution was ten dollars.

$9.95, to be precise. A single, one-time payment for the right to post on a message board that looked – and still looks – like it was designed during the Clinton administration by someone who had recently learned what HTML was and decided that was enough.

That ten dollars built a civilization.

What Ten Dollars Did

To understand the significance of Something Awful’s registration fee, you need to understand what free forums looked like in 2001. The first wave of web forums – phpBB, vBulletin, Invision Power Board – cost nothing to join. Nothing to join meant nothing to lose. Any fourteen-year-old with a disposable email address could register, post something catastrophically stupid, get banned, and create a new account thirty seconds later. The ban was theater. The account was disposable. The incentive to be anything other than whatever you felt like being at that particular moment – which, if you were fourteen and it was 2001, was usually terrible – was zero.

Lowtax’s ten dollars changed the calculus. Not human nature. The people who paid ten dollars to post on Something Awful were not inherently smarter, kinder, or more creative than people who posted for free elsewhere. But when they got banned, they did not simply create a new account. They had to pay another ten dollars. And another ten dollars after that. The ban was not primarily a punishment. It was an invoice. Multiple bans became a financial record of poor judgment, and the community treated repeated bans as evidence not of rebelliousness but of stupidity.

The mechanism was so simple it sounds like a joke: when creating a new account costs money, each account has nonzero value to its owner. Nonzero value creates an incentive to protect it. That incentive produces self-moderation. Not perfectly. Not consistently. But measurably more than free accounts produce. The ten dollars did not eliminate bad behavior. Something Awful had plenty of bad behavior. What it eliminated was disposable bad behavior – the drive-by, zero-effort, zero-stakes variety that constitutes the vast majority of antisocial behavior on free platforms.

Not machine learning. Not a Trust and Safety department with five hundred employees. Not an algorithm. Ten dollars and a ban button.

The Evidence

The empirical case is embarrassing for every platform that came after.

Something Awful’s forums – populated by users who called themselves “goons” with the aggressive irony that characterized everything about the site – produced cultural artifacts that now generate billions of dollars in annual economic activity.

Let’s Play. The entire genre of video game commentary that undergirds YouTube Gaming, Twitch streaming, and the careers of creators from PewDiePie to Markiplier was invented on Something Awful. In January 2007, a forum user named Michael Sawyer, posting as “slowbeef,” started recording himself playing games while talking over the footage, inspired by DVD director’s commentary tracks. The format seems obvious in retrospect. All inventions do. SA’s Let’s Play subforum spent years refining the conventions – episode length, commentary balance, editing techniques – that became the template for an industry generating an estimated ten to fifteen billion dollars in revenue by 2024.

Weird Twitter. The absurdist comedy tradition that produced @dril and reshaped how humor works on social media. The connection is direct and documented. @dril – Paul Dochney, the most influential comedy account in Twitter’s history, whose tweets have been quoted by congressmembers and cited in academic papers – was a poster on SA’s FYAD subforum who went by “gigantic drill.” The voice he brought to Twitter was the voice he had been developing on FYAD for years. “i will face god and walk backwards into hell” is a FYAD post that happened to be posted on Twitter.

Slender Man. Born in a 2009 SA Photoshop contest. Feature film. Multiple video games. An entire subgenre of internet horror.

Image macros. The text-on-image format that became the atomic unit of meme culture was coined as a term on Something Awful in 2004.

Count the artifacts. Count the innovations. Count the communities that trace their DNA to SA’s forums. Then compare that output to any equivalently sized segment of any free platform over any equivalent time period. SA wins by an embarrassing margin, and it did it on ten dollars and a twenty-year-old version of vBulletin.

The Controlled Experiment

In 2007, South Korea ran the closest thing to a national-scale controlled experiment on online behavior. The government implemented real-name verification: national ID number required before posting on any website with more than 100,000 daily visitors. The goal was to reduce malicious comments. The Korean Communications Commission studied the results.

Malicious comments decreased by 0.9 percentage points. Less than one percent.

The most aggressive identity-verification regime ever attempted on the internet, backed by the full apparatus of a modern state, accomplished almost nothing. Repealed in 2012 as ineffective.

Something Awful’s ten-dollar fee accomplished more. Not because ten dollars is a better verification system than a national ID. Because ten dollars creates stake. South Korea’s system created accountability – “we know who you are” – but accountability without stake produces nothing. People will be cruel under their real names with full confidence. Cruelty backed by social consensus is one of humanity’s oldest traditions. What people will not do is be recklessly cruel when recklessness costs them ten dollars every time they get caught.

Facebook proved the inverse from the other direction. Facebook’s real-name policy is the most widely deployed identity-verification system in internet history. It did not produce more civil discourse. It produced the same amount of cruelty, often more, because people are perfectly willing to be terrible under their own names when the cost is zero. Facebook is free to join, free to use, and free to ruin.

SA charged ten dollars and got Let’s Play. Facebook charged nothing and got the 2016 election. The lesson was right there.

What the Fork Proved

On October 1, 2003, a fifteen-year-old Something Awful user named Christopher Poole launched an English-language imageboard called 4chan.org. He modeled it on Futaba Channel, the Japanese imageboard, and recruited his initial users from SA’s anime subforum. The early users were goons. The early moderators were goons. The culture was SA’s culture – the transgressive humor, the ironic detachment, the weaponized absurdity, the instinct for collaborative creative projects – transplanted into new infrastructure.

The new infrastructure had one critical difference: 4chan was free. Free and anonymous by default. Free, anonymous, and ephemeral.

Same creative DNA, different selection pressure. SA produced Let’s Play and Weird Twitter. 4chan produced Anonymous and meme culture. The ten dollars was the only variable that changed. Everything that 4chan inherited from SA – the humor, the creative ambitions, the contempt for earnestness – it inherited without the consequences. Without the ten dollars. Without the persistent identity. Without the moderation that stung.

Every conversation about 4chan’s contribution to internet culture is implicitly a conversation about what happens when you take the culture that ten dollars built and remove the ten dollars. The answer is: extraordinary creative output and extraordinary toxicity, in roughly equal measure, and for exactly the same reason.

What Was Lost

Something Awful’s decline was not caused by a failure of quality. It was caused by the economics of attention. SA charged ten dollars. Facebook was free. Twitter was free. Reddit was free. Each offered some version of what SA provided – community, discussion, humor – without the registration fee. For most users, “free” was sufficient. The quality difference was real but not large enough to overcome the price differential for the majority of potential users.

This is the market logic that kills every paid community. A paid community can produce higher-quality discourse, more creative output, and a more cohesive culture. But “better” competes against “free,” and when the consumer has zero switching costs and unlimited alternatives, “free” almost always wins. The paid community does not get worse. The attention market simply reallocates eyeballs toward the lowest-friction option.

When SA shrank from the center of internet culture to a historical footnote, the internet lost a working proof of concept. Proof that friction is generative. Proof that a trivial economic barrier produces more lasting cultural output per capita than any free platform in history. Proof that the subforum model – one site, many civilizations, each with its own norms and moderators – can solve the problem that modern platforms claim is unsolvable: different communities with different standards of behavior within a shared space.

The entire subsequent history of the internet has been a sustained experiment in removing friction. Free accounts. Free posting. Free everything. Subsidized by venture capital, optimized for growth at the expense of community. Three billion users generating an infinite volume of content, almost none of which will be remembered in twenty years.

Something Awful shaped the internet for the price of two coffees. The lesson was right there the whole time. Nobody learned it, because learning it would have meant admitting that the answer to the internet’s moderation crisis was not an algorithm, not artificial intelligence, not a billion-dollar Trust and Safety division. It was ten dollars and a human being with a ban button.

Silicon Valley will spend another decade and another hundred billion dollars before it arrives at the same conclusion. Or it will not arrive at it at all, which is the more likely outcome, because arriving at it would require charging money, and charging money would reduce growth, and reducing growth is the one thing the model cannot survive.

The wall was ten dollars high. It was enough.

This essay draws from Lurk More, coming fall 2026.

Next week: The man who defined modern physics spent more time on alchemy than on calculus. Newton was an alchemist, and what that means for the story we tell about science is worse than you think.

Source URLs