Dave Ring is a hard man to find. Given that he launched one of the earliest crowdsourced internet projects—if not the first—one would expect Ring to be something of an online superstar. He was using group power before Huffington had a Post, back when Amazon was just a river.

But when you Google his name—the smell test of millennial relevancy—our Dave is buried beneath Dave Ring the stand-up comedian, Dave Ring the editorial consultant, and Dave Ring the mentally disabled televangelist. Perhaps the web has obscured Dave Ring the internet revolutionary because his project was, from the most straightforward perspective, a bust. In the summer of ’94, Ring led 110 online comrades to beat a video game called FreeCell. They lost. 

If you haven’t played FreeCell, someone at your office has. Microsoft has included the single-player card game in every copy of Windows for the past two decades, making it as ubiquitous as the personal computer. 

FreeCell looks like Klondike, the familiar “red nine on the black 10” solitaire game that grandfathers and bored flight attendants play, but victory in Klondike requires some luck. Played perfectly, FreeCell is all skill, and that’s because every card is visible from the beginning. A talented player is able to assess the playing field and manipulate it for victory.

Ring was introduced to Windows FreeCell around 1993. He was a student at the time, pursuing a Ph.D. in Physics at Texas A&M and working on his dissertation, “Gravitational Effects in S(5) Supergravity GUT and a Neutrino Mass Scenario.” Compared to field theory, FreeCell was like one of those children’s toys where the square peg goes in the square hole. 

Broken PC illo

The game includes 32,000 individually numbered hands, so if you get stuck, you can write down the ID number for your hand and come back later to try it again. There’s incentive to return, too. In the help file, the creator Jim Horne has hid a provocative message: “It is believed (though not proven) that every game is winnable.”

But enjoyable as FreeCell was, a larger puzzle had captivated Ring: the internet. “I was fascinated by how our universe works,” Ring told me in a June 2011 email, “and I was also fascinated by how the new ‘cyberspace’ works.”

In the early 1990s, college campuses were one of the few places in the US that provided reliable internet access. Being online was a different experience then. Most users frequented no-frills message boards called newsgroups, which resided on the umbrella network known as Usenet. Newsgroups were a place where professional students could converge from across the globe, like a virtual coffee shop, each one catering to particular topics like politics or juggling or Tori Amos. 

Ring’s stomping ground was the “sci.math” newsgroup, where he and fellow math enthusiasts regularly challenged one another with puzzles and queries. Posts had subject lines like “Claim: Cantor’s Proof on infinite arithmetic is wrong!” and “Which is greater: luck or skill?”

The idea to beat FreeCell sprang from a message posted by sci.math regular Russell Turpin on July 24, 1994, with the definitive title “FreeCell is NOT always winnable.” In his message, Turpin provided an example of an unwinnable FreeCell deal and asked fellow group members to help prove or disprove his gripe:

Is this (1) a trivial result that almost everyone reached after playing the game a little bit (except for the author of the help file), (2) a wrong result, in which I misunderstand the game or the strategy for playing it, or (3) of sufficient interest that readers here might like a proof that the above hand cannot be won?

Turpin’s proof-of-concept hand was indeed unwinnable, but it also wasn’t one of the 32,000 included in FreeCell—possibly for that reason.

Despite a number of immediate responses, Turpin seemed to grow tired of his own questions. Within a few posts, he got distracted and deviated the topic toward semantics, fixating on that comment in the help file.

It is believed (though not proven) that every game is winnable.

Turpin said the note should read:

The computer program generates 32K deals from a very large space of all possible deals. So far, all the generated games we have examined are winnable. We have not checked them all.

Throughout the thread, Ring popped in, sometimes early as 3 a.m., to push the conversation closer to finding answers to Turpin’s questions, whether Turpin actually cared or not. Ring had made up his mind. They would try to solve FreeCell. All 32,000 hands of it.

Broken Mac illo

The idea had hit Ring after another poster gloated that their friend could beat any hand in FreeCell, like a kid whose dad can totally beat up your dad. Okay, fine, he reasoned, let’s organize a team to prove it. They’d work together like people in an office or in a classroom, but online. They didn’t need to be the gaming equivalent of a Mossad hit squad—just a few people willing to play a few pre-assigned hands of a silly card game.

Since FreeCell creator Horne had individually numbered the hands, each deal was accessible by its ID number. If someone in Des Moines asked the program for Hand No. 600, she’d get the same deal as a player in Edinburgh who requested No. 600. The consistency made global collaboration possible. Here’s how he explained it to his cantankerous cohorts:

This could easily be organized in the following way: Each participant emails the organizer (I would be happy to do it if there is sufficient interest) who emails back with a block of 100 hands. The participant can solve them at his own pace, since the whole project will take a while and there are plenty of blocks. The participant reports back a list of deals he couldn’t do, which are then repackaged and sent out in blocks of 10 to the more enthusiastic participants.

Smartly, Ring petitioned other newsgroups with a similar bent, dubbing the cross-Usenet collaboration The Internet FreeCell Project. Though the Project would last from August 1994 to April 1995, the first and only substantial hitch appeared within the first weeks. Poster Dick Adams unwittingly had the scoop: “The first game which appears not to be winnable is 11982.

Members reported other troublesome hands across the participating newsgroups. Hand No. 285; No. 1,941; No. 21,320. These took time, but eventually they were cracked. Hand 11,982 remained. With 31,999 hands completed, Ring called an all-hands on the final hand. A week later, No. 11,982 still stood unbeaten. It was over. They had lost.

Or not. Ring never regarded winning every hand of FreeCell to be the project’s goal. To him, Turpin’s post, along with Horne’s vague help file, was more a question than a challenge. 

So when that final push on No. 11,982—an effort aided by humans and even a handful of game-solving programs—met with failure, Ring celebrated. Is every hand in FreeCell winnable? No. Thirty-one thousand nine hundred ninety-nine hands are winnable. And one isn’t. He proved that. He had solved one mystery of the universe.

The phrase “I beat the game” connotes a pride in having dominated a game. Ring’s group didn’t do that. Yet Ring’s experience with FreeCell successfully tested a model for harnessing the collective power of a fledgling virtual community. The game wasn’t some docile bit of code, but an inspiration, a launchpad, a muse.

Words by Chris Plante, editor-at-large for Polygon. Illustrations by Brendan Leach.