There’s an idea that if you put a monkey in front of a typewriter for an infinite amount of time, eventually, its random smashings of the keyboard would produce a text identical to the work of William Shakespeare. Given infinite time, and assuming the typewriter (and the monkey) will not stop working, any and all conceivable combinations of characters will be typed. Now imagine that the monkey is somehow familiar with Shakespeare. Imagine that the monkey is trying to produce Hamlet as best as he can. Imagine that the monkey is a hundred thousand monkeys. Keep in mind, there is only one keyboard. In early February of 2014, an anonymous user of the video game streaming website Twitch.com created Twitch Plays Pokémon, a game that doubles as a social experiment. He set up a stream of an emulation of Pokémon Red Version that anyone can view. In the chatbox beside the video stream, Twitch members can type in a button command like “up,” “down,” “left,” “right,” “b,” “a,” or “start.” After twenty to forty seconds (due to delay that comes with streaming on Twitch), the in-game character will move or act based on that input. If you type “left,” the character will take one step to the left, and so on. The experiment’s purpose, as stated by its creator is to be, “an experiment to test the viability of this format, the way people interact with the input system and the way they interact socially with each other.” More specifically, it poses a question: Can a hundred thousand people beat one game of Pokémon Red?
The objective of Pokémon is to capture and train monsters called “Pokémon” for battle, in order to be the best monster capturer and trainer in the world. The player proves his best-ness by battling other trainers with his monsters until he has defeated the “Elite Four,” a coalition of the strongest trainers in the game. In Twitch Plays Pokémon, one must do that as a collective hivemind of differing thoughts and strategies. We all want to be the very best (like no one ever was), but I think we should do it with a team of my favorite types of monsters, and Joe from down the block wants to do it with his favorites. Miranda in Boise wants to take a shortcut to the next objective, but Sven in Norway thinks the long way would allow for more training. Here lies the underlying challenge of Twitch Plays Pokémon. While more players will add to the collective knowledge of the pack, another hand on the controller will only complicate basic in-game actions. For one person, moving a step forward requires basic navigational skills. Say Red, the in-game avatar of Twitch Plays Pokémon needs to move right one step to align himself with the door of a building. A thousand different people might type “right” in the chat. While the game will not put every command into effect, a thousand inputs where one is needed will certainly make poor Red stroll right past the door. At times, too many cooks can spoil the broth.
Despite the game’s flaws, it is nice to see how it brings some of its players together. While the chat often moves too fast to actually use it to communicate, social sites like Reddit and Tumblr are used to share discussion and strategy. The game also allows different ways to play. One can be an organizer, creating infographics and charts suggesting why one course of action may be best. Many post links to helpful information – drowning out others who post spam. For those who prefer the minimum amount of involvement, there is always the option of just typing button commands and hoping that yours is the one the game choses to put into play. Because of the game’s difficulty, any minor progression feels like a huge achievement. When thousands can contribute even a small amount to that achievement, thousands can bask in the same satisfaction. When a battle is lost (and boy are battles lost), there is a special moment where thousands around the world can cry out in frustration, all feeling the same thing, for the same reason, at the same time.
What I found most amazing, is the community of players’ constant tendency toward creating depth and story out of a game that can be summarized as “a hundred thousand people screw up” For example, the “Helix Fossil,” an in-game item that the hivemind kept selecting by accident, was considered to be a god-like figure for Red. In the players’ collective eyes, Red kept opening his inventory and viewing the Helix Fossil because he was consulting it for guidance. Very often, some form of “ALL GLORY TO THE HELIX” could be seen in the chat. A Flareon, a dog-like Pokémon with pyric abilities, was considered to be a “false prophet,” that went against the will of the Helix and sought to destroy its comrades. The players’ ability to quickly agree on fan-nicknames is impressive. They caught a Ratata, a rat-like Pokémon which eventually learned how to dig. Soon, everyone started calling it “Digrat.” Digrat transformed into a larger version of himself – people started calling him Bigdig. Collectively, it seemed that players of Twitch Plays Pokémon love to give things names, backstories and personalities. They tend to pretend that the “characters” in this game, which in reality can only do what the hivemind chooses, are not only autonomous, but also good or evil. Many negative occurrences, like accidentally deleting a few Pokémon, were blamed on something “evil,” like the “False Prophet” Flareon. Soon after gaining a following, many of TPP’s players created fan sites, fan art, and fan theories. If I had a doctorate in sociology, I might tell you whether or not Twitch Plays Pokémon is evidence of mankind’s natural desire to personify things, make art, tell stories, create a sense of good and evil, and to form scapegoats. However, I do not have a doctorate, so I won’t be doing that.
Even with a connected community of players, it can be hard to efficiently spread information. While it sounds snobbish to say, many players of TPP are ignorant of the complicated inner-workings of Pokémon Red. Despite a widely used webpage where the game’s current status and objectives are posted, there was no convenient way to broadcast exactly what needed to be said, exactly when it was needed. To specify, there was a point where Red’s path was obstructed by a tree which there was no way around. People familiar with any game in the Pokémon series could tell you that in order to progress, one would need a Pokémon that can cut down trees. However, the method of doing so is not the same in each game. In most installments of the series, the player can approach the tree and press the A button. Then, when a menu appears asking if the tree should be cut, the player can press A again to select “Yes.” In Pokémon Red – the game that TPP is emulating – the player needs to press “Start,” select the “Pokémon” submenu, select a Pokémon that knows how to cut trees, and select the “Cut” option. A wave of people in the chat denigrating those who typed “Start” proved the common ignorance of the game’s players. As a result of that ignorance, and the general difficulties of navigating the game’s world and menus, the hivemind spent several hours trying to cut down one tree. With one player in a normal game of Pokémon, such an action would have taken less than ten seconds.
Many players personify the in-game characters, and as a result, find themselves more upset to lose them. In most Pokémon games, the player is able to go to a “PC,” an in-game computer for depositing, withdrawing, and “releasing” Pokémon. In out-of-game terms, releasing is permanently deleting a Pokémon from the game’s memory. In Pokémon, the option is rarely, if ever used, as a player is unlikely to catch enough Pokémon to fill the PC and need to create space. In Twitch Plays Pokémon, it is used by accident, a lot. Since the game only allows the carrying of six Pokémon at once, a player might need to go to the PC and deposit a Pokémon to exchange it for a new one. Unfortunately, the “Release” button is right under the “Deposit” button. One wrong move, combined with a few bad presses of the A button, and a Pokémon is gone for good. It could be a useless Zubat, or a beloved member of the team. The latter was the case for a Charmeleon and Ratata, affectionately nicknamed “Abby” and “Jay Leno” respectively. On Sunday, February 13, a groundbreaking 12 Pokémon were accidentally released in a 7 hour period. Many players overdramatically referred to the date as “Bloody Sunday.”
Approaching big decisions, players of Twitch Plays Pokémon would get into debates over the next course of action. In Pokémon, one needs a team member that can swim across water in order to get all the way through the game. With a team of five Pokémon and a maximum capacity of six, the question arose: “Should we get Eevee, a fan favorite Pokémon, or keep a space open for Lapras, a Pokémon that can swim?” Some argued that Eevee could be transformed into “Vaporeon,” which can also swim. Others argued that the hivemind had proven time and time again that it lacks the ability to coordinate itself long enough to do anything of that nature. Not to mention, if the hivemind were to fail to evolve Eevee into Vaporeon, the next step would be to deposit something into the PC to make room for Lapras. Even before it had happened, many were wary of accidentally releasing team members. The rival factions found themselves in a tug of war as Red was pulled back and forth on screen. Eventually, the faction that wanted Eevee won. In an attempt to buy a “Water Stone,” the item necessary to transform it into Vaporeon, the hivemind ended up buying a “Fire Stone,” and evolved Eevee into the aforementioned “False Prophet” Flareon. Later, when trying to deposit a Pokémon at the PC, the players accidentally released Abby and Jay Leno. There were many I-told-you-so’s. In addition to disagreements between factions within the game, there are also those between the game’s players, and “trolls” who want to ruin it by appearing en masse to spam inconvenient button presses. In other words, it’s a war between those who want to ruin the game on purpose, and those who do it by accident. Fortunately, there are so many people actually trying to play that the efforts of trolls are negligible.
At a certain point, the hivemind reached a location in the game that proved impossible to get past. In an attempt to make the game viable once again, the creator of the game introduced a new gameplay style. He called it “Democracy Mode,” referring to the original style of gameplay as “Anarchy Mode.” In Democracy, a button is only “pressed” every ten seconds. In the ten second period where Red is not moving, players in the chat can vote for which button they think should be pressed. The hivemind can vote to switch between Democracy and Anarchy at any time. This created another divide in the players, as many found Democracy Mode to be more productive, while others found moving once every ten seconds to be boring. Some believed that Anarchy made the game more interesting, as it made success feel special, rather than guaranteed. Personally, I’ve found that in Democracy mode, there is no risk and constant reward. I wouldn’t want to play a game with no chance of losing. Perhaps the majority of the game’s players agree with me, as Democracy mode seems to only switch on when it’s absolutely necessary. There are still many who find the ability to switch at all unnecessary. To me, the hivemind seems pretty evenly split on that issue.
All calamity aside, Twitch Plays Pokémon, seems to have fulfilled its original purpose. If the goal was to test the “viability of this format,” well, consider it tested. As of my writing this, twelve days have passed since the start of the game. The hivemind has gotten Red to “Cinnabar Island,” which would indicate being most of the way through the game. Even after almost two weeks – almost two years in internet time – you can still find 60,000 people playing at any given moment. Several imitations and spinoffs have been created, like Random Number Generator Plays Pokémon, a game where the button input is actually random, rather than just seemingly. Even if some fine tuning is needed, the clout and longevity of the game prove the format’s viability.
What about the other reason for the experiment? In addition to the format, TPP’s creator wanted to test, “the way [players] interact socially with each other.” From what has already taken place, is Twitch Plays Pokémon a testament to human teamwork and cooperation? Absolutely. Is it a testament to human miscommunication and struggle? Absolutely. It would be naïve to expect perfect collaboration, and pessimistic to expect outright disarray. While not succeeding all the time, many players of Twitch Plays Pokémon attempted to work toward a common goal. Many were trolls, and others just wanted to be part of something. Some just wanted to watch people screw up at a video game. Whatever the individual reasons, over 27 million people have tuned in. A new form of gameplay has been created, and I am excited to see what will happen next.