After my first day in Israel, two things are immediately clear:

1) I’m too fat to live here. Or, more specifically, I don’t have the body you need for the beaches of Tel Aviv, where the scene is more SoCal than SoCal. Men do push-ups. Models play beach volleyball. Two women jog by on a running path, and a guy on a Segway follows behind with a camera, apparently filming a commercial for sports bras.

2) I’m too poor to live here. Or, more specifically, I can’t hang with the deep pockets of Israel’s exploding tech and startup scene, Silicon Wadi, which includes players like Google, Facebook, Intel, SAP, and basically the entire opening credits of Silicon Valley.

In most places “startup scene” also means “blockchain scene,” and Israel is no exception. To the relief of everyone on the beach, I put on my shirt and take a short walk to the Tel Aviv Stock Exchange, and across the street I find the crypto-enthusiast’s answer to the Wailing Wall: a place called the Bitcoin Embassy.

The Embassy is ridiculous and glorious. The walls are festooned with bitcoin flags, bitcoin t-shirts, and shelves with bitcoin wine—(a bottle called “Hodl Reserve”)—and bitcoin bumper stickers with slogans like “Mine your own business,” “Bitcoin, an idea too big to fail,” and, for the MC Hammer fans, “Can’t block this.” An old-school TV flashes crypto prices. The Embassy is a part of Israel’s crypto eco-system that includes the companies Sirin Labs, Orbs, Game Protocol, Zen Protocol, Stox, Spacemesh, and arguably the biggest blockchain show in town, Bancor.

A visit to Bancor is the reason I’m in Israel. (Along with, you know, millennia of history and culture and Jerusalem and stuff.) In 2017, or a decade ago in crypto-years, Bancor raised a then-record $153 million through an ICO, with a promise of doing something new and different. They would be a crypto exchange. Yet the world has plenty of exchanges—theirs would be decentralized, one which never holds the assets of buyers and sellers, and thus immune to the kind of hacks that befall traditional marketplaces. (See: Mt. Gox.)

Yet Bancor planned to be more than that, even—they would be a marketplace that is not really a marketplace: one that doesn’t need to connect buyers and sellers at the same time, but instead, through the magic of their algorithm, creates a kind of Crypto Vending Machine that converts your Coin A into Coin B… even if it’s 4 a.m. and no one wants to trade. Even by the standards of blockchain, the mission of Bancor is a little weird and wonky and divisive and tough to explain, but, in the end, sort of fascinating.

The Long Tail of Culture

Nearly fifteen years ago, Chris Anderson, then the editor-in-chief of Wired, wrote an essay called The Long Tail. With spooky prescience, Anderson suggested that as the internet grew and as the barriers to entry crumbled, we would find surprising value not from the Top 40 of songs and books and movies, but from the oddball 1,000th or 100,000th title. There’s a market for a broader range than just the mainstream hits—additional revenue, or value, can come from niche entries. “What’s really amazing about the Long Tail is the sheer size of it. Combine enough non-hits on the Long Tail and you’ve got a market bigger than the hits,” Anderson argued. “Take books: The average Barnes & Noble carries 130,000 titles. Yet more than half of Amazon’s book sales come from outside its top 130,000 titles. Consider the implication: If the Amazon statistics are any guide, the market for books that are not even sold in the average bookstore is larger than the market for those that are.”

This long-tail concept is baked into Bancor’s DNA. Co-founder Galia Benartzi calls it the “long-tail of culture,” framing it as something of a fundamental axiom. “The culture of humanity is built by those who participate in culture creation,” Galia tells me.

We’re sitting in a conference room in Bancor Headquarters, on the 21st floor of a skyscraper. A cute little dog scampers back and forth between us, nuzzling my leg. The dog (a mutt, adopted by Galia) is the color of a desert, and is named Negev, after the nearby Negev Desert. (Galia calls her the “Chief Happiness Officer.”) At the moment, though, I don’t think I understand how the “culture of humanity” fits into a crypto exchange any more than Negev does, and Galia sees me frown in confusion. “Think of video culture,” she explains. Historically, the people who could create video culture—TV shows and movies—were the NBCs and the Warner Bros of the world. A small group of elites. “And way before that, it was the owners of the printing press. And way before that, it was the king or the scribes,” she continues. “If you were a peasant, you couldn’t create culture.” Fast forward several thousand years. “There are two billion users who do whatever they want with video, and some of these things go viral. The engine of culture creation is now open to so many more people.”

When a billion people wave around the cameras on their smartphones, we get a whole bunch of crap but we also get some viral gems, like a seal slapping a kayaker in the face with an octopus, or a double rainbow. (“Whoaaa… Double Rainbow…What does it mean?”) We get the long tail of culture with Twitter. Same with coding. Same with apps.

And we will see the same tail with crypto, argues Bancor. As of publication, there are 2,003 coins listed on CoinMarketCap—something of a long tail. Is there value added by that 2,003rd coin? (Which at the moment happens to be “EliteShipperToken,” whose website goes to a 404 Error page.) 2,003 coins, according to the logic of Bancor, is a number that is much, much too low.

Eventually there will be more. Millions more. “As technical barriers to entry are removed, we are on the precipice of even millions of user-generated currencies, of all shapes and sizes,” Galia once wrote in a Coindesk op-ed. “This is similar to inflection points in User-Generated-Content we saw with the rise of WordPress for blogs and YouTube for video.”

Hmmm. A skeptic (raises hand) might ask, Okay, so I see how billions of Tweets might produce some winners, but how does this apply to money? Why can’t one universal coin work for everyone?

As an example of how niche currencies can matter, Bancor points to an earlier experiment, called “Hearts,” that the founding team launched in Israel. In 2013, they created a digital currency that was targeted at Israeli mothers. It was basically Monopoly money—conjured from scratch. The Hearts couldn’t be used to pay taxes or buy onions or shop on Amazon, but Galia says that when the moms were given this new currency, they invented creative ways to use it. “So let’s say I’ve got clothes that don’t fit the kid anymore. I’m selling them for Hearts,” she says. “Or I really love to make super fancy birthday cakes. Who’s got a birthday party coming up? I’ll make the cake. Or I can watch your kids. Or I can dog-sit while you’re away.”

Their lightbulb moment: The new currency created new value. The currency was more than simply a way to lubricate the engine of commerce—it created commerce. Twenty-thousand moms did an estimated $20 million worth of economic activity, according to her team’s estimate, which they say is purely incremental. (Without the new incentive of Hearts, theoretically, the mom wouldn’t have baked that super fancy cake.) Even if the numbers are puffed, it’s still an intriguing possibility… do currencies create value? If so, can you scale this?

“There are a lot of reasons why, but the bottom line is that there are many, many people—we’re talking billions—across the world who do not have money in their hands. So they’re locked out of financial collaboration,” says Galia. “This long-tail is the hundreds of millions of communities all around the world that want to collaborate, but don’t have the tools.” In a more recent proof-of-concept, Bancor is digitizing a new local currency in Kenya, letting farmers trade tomatoes on the blockchain. (Their Phd whiz-kid in charge of the project, Will Ruddick, gives an I-dare-you-to-explain-it-to-me visual simulation of how Bancor connects local currencies in a video that looks like an Atari video game.)

Yet there’s a reason you never heard of Hearts…and Bancor is counting on that reason.

Shot Across the Bow

Israeli moms could bake all the cakes and knit all the sweaters they wanted, but at some point, they’d want to cash out their Hearts for shekels. They couldn’t. There was no liquidity and no market. The project sputtered and died.

“Hearts failed because it wasn’t tradable to anyone outside of the community,” says Nate Hindman, Bancor’s head of communications. “This formed the mission for Bancor.”

As a few employees sit down for a lunch of hummus and pita, Beatles music plays softly in the background. (Much of Tel Aviv is like this—a blend of Middle Eastern and American pop culture.) Nate guides me through a large room of maybe 30 programmers, all hunched over their keyboards and clacking away. There are some whispers in Hebrew. “No one in this room speaks English,” Nate tells me. “You can say whatever you want, and they won’t hear you.”

“Somehow I think you’re lying.”

A few chuckles from the Israeli programmers.

“In Bancor there are no trading fees,” Nate explains, back to business. “Everything is done on-chain. And there’s no ‘market maker.’” This last one—the lack of an intermediary—is arguably Bancor’s raison d’être.

For thousands of years, fundamentally, markets have worked in more or less the same way: buyers meet sellers, and then they shake hands and agree upon a price. Whether this transaction occurs in goats, gold, paper dollars, digital dollars, electronic credit, or even bitcoin, the concept was basically the same—the price is “whatever the market decides.” As every student of Adam Smith knows, this is Economics 101.

The problem with a marketplace is that you need to have enough buyers and sellers engaging with each other at the same time. In Econ-nerd talk, you need to have a “double coincidence of wants.” If you bring your cow to a marketplace of vegans, then you don’t have a trading partner and you’re out of luck. The invention of currency helped to solve this problem, but even then, you still need buyers and sellers to show up at the same time.

So what happens when you have a very “illiquid” currency, where there are not many buyers and sellers, in a currency like, say, Hearts for Shekels? What if you want to trade a just-invented BREAKER Coin, but you don’t see many takers? What happens if Bancor’s vision for the Long Tail of Currency is realized, and we have millions of coins?

In the Bancor model, there’s a built-in “market maker” that connects any Token A with Token B, no matter how niche or tiny the market. Using a clever interlocking system of smart contacts (initially built on Ethereum, and now also using EOS), Bancor is able to swap a Token A to Token B using a “connecting” token, which adjusts the price with each and every transaction. As more of Token A is purchased, the demand increases, and so does the price. The price slides up and down the supply curve, according to the Bancor pricing algorithm. There are no Order Books. There is never a spread between the buy and selling price (they’re one and the same), theoretically cutting fat from the system. With this sprawling list of coins that can be traded for different coins (as opposed to just bitcoin or ether), Bancor has 7,750 token pairs and counting.

Why does any of this matter? Some say it doesn’t. From its inception, Bancor has been accused of lacking a real purpose. Why do we need this magic box that swaps Token A for Token B? “There already exists a common currency through which we can trade. It’s called ether, and we can use it no matter which token pairs we want to trade, because those very tokens are, by definition, implemented on top of Ethereum and were purchased with ether in the first place,” argue Emin Gün Sirer and Phil Daian in a blistering blog post titled ‘Bancor is Flawed.’ “Using BNT [Bancor] tokens is like stepping into a kid’s swimming pool, placed in an ocean.”

Shots fired. “Should all token development stop now that we have ETH?” Responds another co-founder, Eyal Hertzog, in a measured, collegial, line-by-line rebuttal. (The third co-founder is Guy Benartzi, Galia’s brother.) “Should ETH never have been created because we had BTC? Should BTC never have been created when the USD was already a liquid medium of exchange?” (There are many, many weedy points and counterpoints—almost as many arguments as there are coin pairings. The full rebuttal runs 12,000 words, or four times the length of this article.)

And then there’s the hack. On July 9 of this year, someone hacked into a Bancor wallet and swiped 24,984 ether, 229,356,645 of a coin called Pundi X, and 3,200,000 BNT (Bancor’s native token)—worth a combined $23.5 million at the time. The Bancor brain-trust then smashed an emergency button, using a break-the-glass override to freeze wallets and prevent the loss of $10 million, capping the downside at $13.5 million. (None of the lost $10 million, Bancor took pains to emphasize, came from private wallets—it all came from Bancor’s own piggy bank, or the “reserve.”)

That rescued $10 million came at a qualitative cost, as it tarnished the company’s “decentralized” street cred. Decentralized is to blockchain what “mindfulness” is to yoga instructors. “An exchange is not decentralized if it can lose customer funds OR if it can freeze customer funds,” blasted Litecoin founder Charlie Lee on Twitter. “Bancor can do BOTH. It’s a false sense of decentralization.”

In a humble, sober blog post called “Decentralization is a Journey, Not a Destination,” Eyal Hertzog (who always seems to write these things) patiently points out that the emergency override was a transparent feature that they disclosed from the jump, that every user owns their own wallets, that Bancor cannot even access those private wallets, that Bancor never takes possession of anyone’s tokens (unlike most exchanges), and besides, if people still aren’t happy, they could always just (and I’m paraphrasing here) go fork themselves—fork the protocol and create their own faux-Bancor.

“If you want to build your own Bancor, go for it. All the code is there. We make it easy for you,” says Galia, back in the conference room. She acknowledges that, yes, Bancor does not pass a Decentralization Purity Test, but how much is that purity really worth?

We now have the answer: The purity is worth less than $10 million. “You can make a claim that says, I would rather see a thief get away with $10 million dollars than see any inkling of centralization, because I don’t trust you, Bancor. And that’s a valid claim. Those are the Decentralization Purists. And they can create their own systems that are completely decentralized when they get hacked. And when the money is lost, they’ll be okay with it, because that’s part of the price that they’re willing to pay for decentralization… We don’t think that people want to pay $10 million in the name of decentralization on Day One.”

Negev, the dog, looks up at me and wags her tail, wanting some attention. She’s getting restless and clearly wants to leave the room, and I sense that our time is running up. But something’s bugging me. I’m starting to get the sense that what Bancor is doing—replacing a real-time buyer-vs.-seller marketplace with an algorithm-built price model—is more fundamentally different than most people realize. Is Bancor’s model a shot across the bow to the classic, Adam Smith-style economics of free markets? I ask Galia if that’s how she sees it.

She thinks for a moment. “Those guys,” meaning 18th century economists, “They didn’t have computers. They didn’t have the internet. They didn’t have data. They didn’t have blockchain. They didn’t have the ability to make a system that could somehow automatically balance the trades that are coming in. They did the best they could. They couldn’t have anticipated all of the loopholes.”

These are the loopholes of whales and bots and super-traders. “Free markets would work fine if they’re really free. But they’re not free,” says Galia. “We don’t have the same information. If I’m a trader at Bloomberg with my 10 monitors and my proprietary software—software, by the way, that costs money to get—and you’re a person at home trying to buy or sell a stock, we don’t have symmetrical information.” She cites the problem of “spoofing,” where a devious trader can stack up Buy Orders in the queue to indicate interest, entice others to buy, and then sell when the price is higher. Or they could just buy and sell to themselves as a way to juke. (Because there are no Order Books on Bancor, theoretically that can’t happen.)

“In the Bancor system, the price is known,” she says, pointing to a classic supply and demand curve. “With a digital asset, you already know the supply curve,” (Such as the case of bitcoin, which will eventually have 21 million coins, expanding on a perfectly predictable schedule.) “And the demand changes all the time. So in the Bancor system, the price-curve is always known.”

“So yes,” she continues, “It is a shot across the bow. This is one of the reasons blockchain is so fundamentally powerful. It’s a new tool that lets us do things in a new way. Sure, you can do new, shitty things with this tool. But you can also fix some very apparent problems. I don’t think anyone would tell you, ‘Ooh, the markets are so stable and so great.’ I don’t think anyone would tell you, ‘International trade is so great.’”

From the floor-to-ceiling windows on the 21st floor, you can see much of the Tel Aviv skyline. Galia nods toward a window. “If you look at a city skyline, every single tall building in every city is pretty much either a bank, an insurance company, a media company, or a real-estate company,” she says. “The holders of capital continue to amass more and more capital.” If the middle-men are cut out… if decentralized tools like Bancor can replace the rent-seekers… Galia finishes the thought. “There would be trillions of dollars of profit that could no longer be made by traditional profit-makers.”

Yet even in the most outlandish and crypto-bullish scenarios, we’re still a long, long way from the topple-the-banks ending of Fight Club. According to Galia’s own back-of-the-envelope, Bancor makes up about 50 percent of the action on decentralized exchanges. Decentralized exchanges, in turn, represent one percent of all the trading in crypto. And crypto is maybe one percent of the global economy, she estimates. (That last guess is tricky to pin down, but one December 2017 estimate had the global market cap of every stock on the planet at $80 trillion; at the time, the market cap of crypto was around $700 billion, or a little less than one percent. Since then, of course, it has dropped.) Even with the most generous estimates, Bancor, at present, is about .005 percent of the world’s trading activity.

Galia is called away to another meeting, the programmers go back to programming, Negev goes back to looking really cute, and I walk back towards the Bitcoin Embassy and to the beach. The tanned fitness models have been replaced with visions of supply and demand curves. I’m still skeptical that the world needs 2 million or even 2,000 currencies. Do we really need a BREAKER coin? A Jeff coin? But on the walk home, as I stroll past the banks and the tall buildings, I think about that long tail of culture, and how, even when Chris Anderson wrote that original essay, we couldn’t have predicted what, exactly, would come from the funky corners of the internet—only that it would surprise us.

Jeff Wilser is the author of The Book of Joe: The Life, Wit, and (Sometimes Accidental) Wisdom of Joe Biden. Follow him on Twitter. Photos of Galia by Maayan Benartzi. All other photos by Jeff Wilser.