The complicated case of Threes, 2048, and the giants that ripped everyone off in the end

Illustration by Claudia Chinyere Akole

Imitation is the sincerest form of flattery — and the most profitable way to make a mobile game

Gabriele Cirulli was 19 years old when he made 2048. At the time, in early 2014, he didn’t realize two things: first, that something as simple as a web game could go so intensely viral for reasons that would elude him. And second, that making something so extraordinarily popular, even unintentionally, would change the course of his life.

After all, 2048 was supposed to be a weekend project — one of many he’d started — intended to help Cirulli practice his JavaScript skills. In fact, another version of 2048 already existed. It was made by someone who hung out on a forum Cirulli was familiar with. But that guy’s version had no animations, and a color scheme that Cirulli didn’t like. So, like many things on the internet, he took the idea and built his own version. Personalized it. Remixed it.

Cirulli had just finished high school in northern Italy, and was deciding whether he wanted to try finding work as an engineer or go to college. So he re-created 2048 from scratch, to better learn how a developer might structure a project like this. But the game was more or less the same as the one he’d based it on: the player is given a 4×4 grid, slides tiles to combine them, and chases bigger and bigger scores.

He published his 2048 on GitHub, and dropped a link on a design website, hoping for some feedback. The web game ended up on Hacker News, and spread from there. Suddenly, Cirulli was receiving congratulations from friends on Facebook. He checked Google Analytics and saw, at first, that 150 people were playing the game at one time. Then he watched it balloon to 13,000 active users. Interview requests started coming in. Italy’s national TV station called him. The game was everywhere.

“Then things kind of got out of hand, for me personally, because I’ve never been one to want to be in the spotlight … Those two or three months surrounding the release of 2048 were some of the most stressful months of my life up to that point,” Cirulli tells The Verge. “It was all quite surreal.”

The attention was flattering, but also stressful and overwhelming. And then came the accusations: that Cirulli had copied someone else’s game. That he was a plagiarist, a thief.

It wasn’t that he had made a new version of 2048. In fact, unbeknownst to Cirulli, the game was just one in a long string of iterations, imitations, and copycats — his 2048 just happened to be the one that went viral. But as the messages came in, and the articles went up, the chain of mutations pointed back to a single source of inspiration: an iPhone game called Threes.

“That game was not something that I’d ever seen or played at the time,” he says. “And then people started pointing out that I stole their success.”

Could Cirulli be a thief if he didn’t know it at the time? And would that even matter?

Asher Vollmer was fresh out of college and trying to figure out what came next. He had published a modestly successful iOS game in 2012 called Puzzlejuice — a creative mash-up of Tetris, Boggle, and Bejeweled — that was part of a wave of indie game developers trying new things on nascent mobile platforms. It did well, and Vollmer wondered: could he do this for a living?

“We were still figuring out how the phone fit into our lives as people, and what people liked to do on their phones. And no one had a good answer — even Apple didn’t have a good answer,” Vollmer says, somewhat wistfully. “So it was up to basically a bunch of indie developers to make a bunch of random prototypes and have a bunch of different excitement about whatever idea.”

In 2014, consumers were willing to pay a few bucks for an app. Apple supported the trend with light forms of promotion. According to Vollmer, a small indie game just needed to get a banner in the App Store. “If you got that, you were golden. You had basically made back your money.”

But his next game, Threes, would more than make back his money. From its launch, the reception was rapturous. Reviews were glowing, one going as far to say it was “about as close as it gets to a perfect mobile game.” Threes even charted well in the App Store, hitting the number one spot, where it remained for over a week.

It was only a matter of weeks before the arrival of the imitations: Twos, Eights, 1024, 2048, 2048 Doge, to name a few. Most of them looked different. But their gameplay was very similar to Threes: a grid of numbers, simple swiping controls, a pleasing and addictive rhythm of play.

Unlike Cirulli, who knocked out his web game in a weekend, Vollmer and his partner Greg Wohlwend had spent 14 months developing and refining the core concepts. Threes is an attempt at a perfect, minimal puzzle game — something challenging and rewarding, the kind of project that needs to be endlessly tweaked and playtested. Like “carving a sculpture out of a piece of rock,” Vollmer says.

Over a year of work, and it took just three weeks for the first clone to appear.

How does one respond to imitators? Legal action was a nonstarter. The strongest case for copyright protection would be “look and feel” — and Cirulli’s game, with its autumnal palette, was a far cry from Wohlwend’s tri-color animations. And to protect the game mechanics, Vollmer would have to file a patent, an arduous legal option and a protection he already felt was harmful to artists and found personally “ethically dubious.”

Instead, Vollmer and Wohlwend published an open letter about 2048 and the various clones, voicing their frustrations, but mostly attempting to be transparent about what went into Threes. The post was accompanied by more than 500 email correspondences between the two of them. (When in doubt, show receipts.)

It did little to slow the virality of 2048, but the letter did turn Vollmer and Wohlwend into developer folk heroes. A number of tech outlets, including this one, wrote pieces defending Threes. There was a deep sense, among a vocal niche of people invested in the indie games scene, that the Threes guys had been screwed over by the global phenomenon of 2048. In reality, the game continued to sell — still beyond Vollmer’s early expectations. “We kind of rode the coattails of 2048’s success in a sort of recursive, ouroboros kind of way,” he tells me.

Vollmer is careful not to sound judgmental. He doesn’t begrudge anyone who would rather swipe around 2048, but he does believe his game is more sophisticated in terms of what it asks of its players.

In fact, it was likely 2048’s lack of sophistication that allowed it to go so viral in the first place. For one, Cirulli’s version was web-based and free, making it more accessible than the $1.99 iOS game. More importantly, it was easier, arguably to the point where it was barely a challenge.

The design patterns of Threes tapped into, and likely influenced, a burgeoning genre of “hyper casual games” — a term that did not exist at the time. 2048 took those patterns and made them even more frictionless.

When we speak now, nearly eight years after the release of Threes, Vollmer and Wohlwend are surprisingly at ease with how it all played out. They each make it clear that they have no hard feelings against any of the independent developers who made their own things inspired by Threes.

“I think Gabriele made a cool thing,” Wohlwend says. “It’s the kind of thing that I would’ve done too, if I had those skills at that age.” (Threes was not even Wohlwend’s first run-in with a copycat — he’d been the artist on a game called Ridiculous Fishing that was ripped off before it was even released.)

His hesitations are reserved for Ketchapp, the French company that took Cirulli’s 2048 and turned it into an app. “That’s the thing I would point to as the villain,” Wohlwend says.

Threes arrived in the midst of a sea change. A new, more lucrative model for mobile games was emerging. They would be free, and make their revenue from in-game advertising — a trade-off that many developers found boorish, especially within the indie games community.

But consumers seemed to prefer the free-to-play model. (Vollmer and Wohlwend would relent and release an ad-supported iteration of Threes a year later.)

“[Ketchapp’s] 2048 was on top of charts for multiple years, bringing in the big bucks, basically cashing in on this product that was the result of this chain of clones,” Vollmer says.

Still, he is quick to deflect any pity. All told, Threes was a success. It allowed him to start two different game studios, and keep working on projects. That’s what he had set out to do.

So who does he feel bad for?

“Gabriele Cirulli,” he says, “the guy who made 2048 did not see any monetary profit off his work.”

It’s true, Cirulli would never get rich from 2048.

When he published 2048 to Github, Cirulli released it under the MIT License, an agreement that allows anyone to use, copy, or sell a work without limitations. The MIT License is, arguably, the purest expression of an open-source internet. But the decision for Cirulli at the time was barely one at all.

“I didn’t really know much about licensing and I thought, ‘MIT is the simplest one’,” Cirulli says, before clarifying: “It’s the shortest one.”

He made 2048 assuming few people would ever see it. Why would he care what the license was?

But this was what allowed Ketchapp to develop the game commercially, and even use the same name. Their 2048 was available on the App Store just over a week after Cirulli published his. It immediately took the top spot on the games chart.

Even for ideas that weren’t under the MIT License, imitation had become Ketchapp’s bankable model. The company would also publish its own versions of that year’s biggest game, Flappy Bird, an app that would epitomize the then-burgeoning genre of “hyper casual” games.

In fact, in a statement provided to The Verge, the company takes credit for creating the category: “With 2048’s release, Ketchapp founded the hyper-casual gaming market. A market of super-snackable games where indeed around 90% of the revenues are generated by in-game advertising.”

If Cirulli had released 2048 under any other license, there’s a good chance he would have seen some portion of Ketchapp’s profits. (Or, perhaps, just been knocked off anyway.) But even with hindsight, he seems okay with the outcome. “It’s not something that I lose sleep about,” Cirulli tells me.

The pressures were mostly external. “People were pushing me on, ‘Oh, you’re missing out on this opportunity. You’re missing out on the money’,” Cirulli recalls. “Personally, I never felt attached to the concept of making money from 2048. So, I don’t really regret what happened.”

He only wishes Ketchapp had gotten in touch with him, at least for some recognition. Even though the App Store description of their game says it is “inspired by Gabriele Cirulli,” he has never been contacted by the company. (Ketchapp declined to tell me why they never reached out.)

The attention yielded different rewards for Cirulli. After the explosion of 2048, a friend got in touch about a job. Cirulli moved to the US for three months to participate in the prestigious startup incubator Y Combinator. It set him on the path to his career at a handful of startups. Today, Cirulli works as an engineer at AgileBits, and is living happily in Amsterdam.

Reflecting on 2048 now, Cirulli talks about his journey to discover happiness and fulfillment. He’s proud that he made something so widely played, but that was only the beginning: “[The experience] taught me a lot about the value of my own work, and about my own values in terms of who I want to be.”

After all, that had been the reason he’d made 2048 in the first place: as a way to figure out what he wanted to do with his life.

As I started reporting this piece, a small controversy erupted around the viral daily puzzle game Wordle. A developer had made an iOS clone of Wordle (called… Wordle). He gloated about how much he would make from it and was subsequently dragged on Twitter. Apple removed his version from the App Store, and in response to his critics, he decided to tweet through it.

Wordle is an interesting case because its virality can be attributed partly to its accessibility: it’s a free browser-based game, not unlike 2048. The game mechanics were inspired by a British game show created in the ’80s called Lingo. The offending iOS developer argued this is what gave him license to sell his copycat, since Wordle itself was an imitation of sorts. (Since he isn’t well known, we’ve decided to withhold his name; he did not reply to The Verge’s requests for comment.)

Mostly, it’s funny that nearly a decade later, we’re still having the same conversations about imitation.

Today, Threes remains the biggest thing Vollman and Wohlwend have ever worked on. Unless you count their influence on 2048. And in a way, didn’t they contribute in some small part to a massively successful game?

“I’m kind of proud of 2048, because it’s a huge game and that made a big impact on a lot of people’s lives, and it’s heavily, heavily, heavily based on my design work,” Vollmer says.

For Wohlwend, he’s unbothered that his artwork was the element that did not carry over from Threes to 2048. He still feels that his style — the “flat, friendly, and colorful” aesthetic he gravitates toward — has resonance in the mobile game world.

That world has only grown bigger, and more profitable. In 2016, Ketchapp was acquired by major game studio Ubisoft. Since 2014, the company has published 240 games; the free-to-play model has expanded outside the mobile game market, and is now the revenue stream for several of the world’s most popular console and PC titles. The company would not reveal revenue figures to me, but told me both the iOS and Android versions of 2048 combined have over 100 million downloads. As of this writing, Ketchapp’s 2048 remains at 39 on the “board games” App Store chart. (The top game in the entire store, when I checked, is a Wordle clone — different from the one that got flamed on Twitter — this one, called Wordle!)

These days, the paid and free versions of Threes have a pretty even split in revenue. An updated version called Threes+ is available as part of Apple Arcade, Apple’s foray into the new business model of gaming subscription services.

As for Cirulli, nearly a decade later, the lessons are more personal.

“It’s something that took me years to realize: that I actually get to share in the success of 2048 without feeling ashamed about it,” he says. “Because for the longest time I felt like a fraud for creating something that sort of blew up in my hands.”

I ask Cirulli if he’s played Threes. He did, just a few months after releasing 2048.

“I cannot say I got the hang of it, to be honest. It felt like it went a bit over my head,” he says. “From a creative standpoint, it’s obviously the superior game. There’s no doubt about that.”

Still, Cirulli is thoughtful about the differences. Threes is a tougher experience, a real game, that requires commitment and thoughtfulness from its players. If Threes was like reading a book, 2048 was the equivalent of scrolling Reddit. In the end, 2048 is a game that looks complex, but really isn’t. That, to him, is the beauty of what he created.

Even now, about 200,000 people visit Cirulli’s 2048 website every day.

“It’s a really simple game with no real rules. So you get to make of it what you want.”