Instant AI - Just Add People

Even today, in the age of the Wii, video games still have a predominantly male image. To explain this historic lack of games that appeal to girls, many people point to the extremely high complexity of social or emotional systems compared to the movement of objects and projectiles. Making Space Invaders is a lot easier than making an interactive adaptation of Pride and Prejudice.

This may be true, but this complexity gap need not stop us any longer. Why bother simulating any social or emotional nuances when real live humans are so easy to co-opt into willing participants in a social space facilitated by web technology? Don't waste your time writing advanced AI, just use people.

Say you have some item, like a virtual item in a game. If you are making a social game, you might be tempted to try directly specifying its effects on a player's social network. This would be hard to do effectively. But if instead you structure your players' social interaction and communication to meaningfully involve and be affected by this virtual item, these players will construct a much more nuanced meaning for it themselves. There is no need to abstract it away.

Provide opportunities for the players to create significance themselves out of the elements you provide. Come to think of it, this approach can even apply outside of social games, sort of like how Tale of Tales would like to see more games as instruments.

Now I think that abstracted formal simulations of social dynamics have their place, particularly for educational purposes or to illustrate a certain point. But I wonder if either of these goals might still be compatible with the structured socialization approach, using real players. It's an interesting idea. How far can you go in simplifying these interactions and dynamics before people are unable or unwilling to play in your world?

*image from the excessively cute and catchy Less Than Three*


Unknown said...

You raise an interesting point, and I may be missing your point entirely, but I think there is one significant difficulty with this, inherent in all multiplayer environments:

Cost andwork.

In the real world, an expensive handbag carries high status because somewhere down the line, it cost someone a lot of money. Maybe it was bought as a gift from someone who has money, representing an association with status, or it was purchased directly by the owner, again giving them status.

Without the cost of the bag, or explicit controls on the rarity on it, there is no starting point to develop a narrative significance for it. It is revealed to be a simple Token.

Players might be encouraged to imbue it with some significance, but without constraints specified at some level by a designer, it seems unlikely. Players are Players (and not designers) because Players are often uncreative and a bit lazy. See: Yahtzee's gripe with Scribblenauts and imagination.

axcho said...

Thanks for the in-depth comment! I really appreciate it.

First of all, I agree with what you are saying. When I said that designers don't have to explicitly specify all the effects of an item in a social game, I didn't mean to say that they wouldn't specify any effects. At the minimum, you'd need to constrain cost and rarity, as you suggest. But I also imagined other effects on gameplay. What you wouldn't specify is how these gameplay effects ripple out on the social network.

I did read Yahtzee's blog post on Scribblenauts and I do agree that there is a difference between meaningful creativity and pointless freedom of choice. The point I wanted to get across in this blog post was essentially what Daniel Cook wrote in his recent article Three False Constraints. In particular, I wanted to suggest that social games with real people can easily tap into social and emotional experiences that are almost impossible to recreate in single-player games.