Crowd Sourcing Play

So Twitch Plays Pokémon  (affectionately known as TPP) has been all the rage lately. For those who don’t know, somebody set up a Twitch stream (Twitch being the game streaming service) with a ROM of Pokémon Red and allowed players to input commands in chat. Soon thousands and thousands of people were participating.

 As expected, having people simultaneously input commands is not a great way to play a game. At first TPP rose in popularity because of how funny the concept is. The character in the game moves completely randomly due to all the inputs (and the 14 second delay between chat and game doesn’t help either). Yet somehow, the players got 3 gym badges in a couple of days . That fascinates me to no extent. If you look at them playing, it looks completely random, but there is enough agency there to actually play through the game. Thousands of players, surely some of them being trolls, are able to to play the game, although slowly. There were of course troubles (releasing the starter Pokémon or other top Pokémon into the wild for example). But all in all, It’s been a success in a way only the internet can make something a success. For example, here are all the religions that have popped up in TPP.

When I first saw this, I told my roommate that I would have done it using a tick system. Every 5 seconds, the chat would accumulate the inputs as votes and do the command with the most votes. Turns out they ended up adding a Democracy/Anarchy slider which allows for the same idea. Players, in addition to typing in commands, can vote for democracy or anarchy. Anarchy mode is how the original system was set up. But if enough people vote for democracy, the tick system is activated. It’s really interesting how the players prefer anarchy. Only when the game is really difficult do they end up going to democracy mode.

I’ve always been a fan of crowdsourcing, even since one of my professors at CMU, Luis von Ahn, talked to us in class about his work with reCAPTCH and The ESP Game (check those things out). Crowdsourcing playing a game is something I never thought of, and it’s fascinating. I wonder what it takes to design a game that’s meant to be played with a crowd, and how other examples of crowdsourcing play do it. Here are my random musings on the subject:

It would have to be more than a simple voting system. There’s a reason the TPP players prefer anarchy. It’s more fun, more interesting. I feel like, with voting, you feel like your contributions to the action in the game aren’t as significant (maybe why kids our age don’t vote in the real elections?). In the TPP model, democracy mode is only enabled when trolls could easily ruin the gameplay.

In the Civilization community, there is an interesting game type that players have made up called Democracy games. A group of players play a single game together from the point of view of a single Civ. But they run the game as a government. There is the finance minister and the military minister. The President gets to actually take the turns. There are election cycles. The save file is passed around from president to president. It’s a great way to add role-playing and social interaction to a game which otherwise doesn’t have it. Perhaps a specifically-for-crowdsourcing game would use a similar system.

Another option might be to have a game where all the players have the same goal, but there are many roles. And people in the crowd would need to be able to perform all the roles to make sure the game goes according to plan. A simple example of what I mean is this scenario: There is button A and button B, and both need to be pressed 100,000 times, but if A is ever pressed 1000 more times B, you lose (or lose hp or points. Now extrapolate that to a better mechanic than button pressing and it might be successful.

All in all, this is a design space that’s pretty open. Hopefully TPP inspires people to try designing more games that can be played by a crowd.

Chess: Great Game or Greatest Game?

Whenever I am asked about my favorite game, I usually mention Sid Meier’s Civilization IV. I usually don’t even consider chess, which I played “professionally” (as in I went to USCF tournaments) in high school. Odds are however, there will be a point in my life where I will stop playing Civilization IV, maybe because a different Civilization game comes out or maybe because I can’t find 10 hours at a time to play a single game. However I will probably always play chess. The fact that I don’t even consider it when thinking about games means it has something transcendent about it. Chess is universal (perhaps behind only soccer in terms of how ingrained it is in global culture). What makes Chess so great? And will there ever be a “new chess,” that is a game that takes the place of chess culturally?

                As Jesse Schell has mentioned a couple of times in my game design class, chess has been perfected over centuries. Rules have been added and changed over and over again until we have the game today. If a good game is the result of iteration, chess has a massive head start. The history of the game adds so much to it culturally. I’ve studied matches from the 1600s that were written down and recorded. You certainly can’t watch replays of Soccer matches from that long ago. Everyone, and I mean everyone, knows chess and knows it is a deep and tactical game.

In general, Chess has been studied a lot. I am very curious to know if there has been a game that has been studied as much as chess. The fact that a game with 5 unique types of pieces is deep enough that there are hundreds of names for different opening variations and end game positions is astounding. The 5 units have been iterated over to perfection, and both sides are nearly balanced (slight advantage to white). There is no randomness, which is crucial, because it means the better player really is the better player.

                What does all this mean in terms of game design? Is it possible to develop a “new chess” that is as well regarded as chess? It’s probably possible, but very difficult. The game would have to have fairly simple rules, but be very deep strategically. There could not be any randomness. It would need the possibility of being able to be studied deeply. Most importantly I think, is the medium of the game. Chess can be played on the computer, but it isn’t (especially in tournaments). There is something about physically moving the pieces and physically taking your opponents pieces that makes the game so much better. There is a reason why all sci-fi movies that have a “standard” game like chess usually have it on a holographic, futuristic board rather than played on a screen. The computer screen aspect may be the reason why e-sports (which are certainly gaining in popularity) are still a niche hobby for “hardcore gamers.”

                Even if a game has all that, it will take a while for it to gain traction (especially in today’s marketplace). Chess didn’t have to compete with thousands of other games. But who knows, maybe in a hundred years, people are playing a game with holographic pieces that hover or something, and no one will know what chess is. Or the much more likely scenario, people are playing holographic chess.