Will: Imagine two kids playing together. Let's call them Calvin and Dennis. One minute, they're playing with toy soldiers in the dirt, the next, they're building a small dam across a trickle of water. Maybe they go inside and play a board game or watch TV. In each case, they are building a shared world that resides partially in the real world and partially in their imaginations. The play between them is both designing these worlds ("This can be the fort," "Let's make a road now") and interacting with them ("The dam is breaking").

Now, let's add a third player to the group; we'll call her Janus. Except the third member of this group is not a person, but rather a computer of some sort. I'm not concerned here with the form Janus takes. Perhaps she's some sort of VR eyeglasses or maybe a direct spinal tap. I do think she'll be portable, so she can be used anywhere (even outside in the dirt).

Janus collaborates with Calvin and Dennis to design and simulate imaginary worlds. She can also play in them like the other players. She can take on assigned roles, suggest new activities, insert chaos into the play, and over time, will even learn about her friends. She'll keep track of what they like and dislike, what scares them, even what concepts they don't understand yet.

As they play in the dirt, Calvin might ask Janus to make some soldiers. Then, they might see little hi-res soldiers crawling around in the dirt in front of them (via a heads-up image overlaid on the real scene by the VR glasses). Janus might playfully decide to have the soldiers all hide in crevices until the boys can capture them by touch, or maybe she'll play a trick on them by sneaking up behind them with a 50 foot soldier that shows them how big they seem to the little ones.

Whatever Janus turns out to be, I just hope she's not predictable.

5. Now that you've told us where we might be one day, where do you realistically see games in two, five, and ten years down the road? When do you think your vision might become a reality?

Will: In two years, we'll really start to hit diminishing returns from our graphics engines. Additional CPU horsepower will start looking for other areas to improve. The newest areas of advancement will be sound, physics, and behavior (AI).

In five years, behavioral engines will become much more robust and begin to be licensed much in the same way that graphics engines are licensed today. Machine intelligence will still be very much in its infancy however.

In ten years, adaptive systems in all computer applications will become the norm rather than the exception. These systems will observe everything you do and attempt to redesign the games you are playing as you play them. Two players might buy the same game, but after playing it for several months, there will be little resemblance between them, as they evolve to appeal to each individual. In a very real sense, the game players will become the game designers.

My vision in about 30 years away. We're quite close to the interface hardware now, the AI will be the bottleneck.

6. What do you think distinguishes you from other visionaries in the industry? Conversely, what do you share in common with them?

Will: There are two things that have really had the most impact on my work as a designer: reality and creativity. It seems to me like there are so many cool subjects out there in the real world to wrap a game around that we haven't even scratched the surface.

I especially like playing games that change my perception of the real world in which I live. Many people who've played SimCity have remarked on how after playing for a while, they start to see their real city in a whole new way. They start to notice the way things are zoned, the traffic patterns, and the long-term development of their environment. I'm noticing the same effect now when I go home to my family after playing The Sims all day at the office.

Many games are also still rather linear or else the game state space is quite small (all the possible states of the game). This implies a lack of creative freedom for the player. I like focusing on designs that allow hundreds of solutions to the problems faced by the players rather than the more typical two or three. The idea that each game is not only different for each player, but also that it's a reflection of the player's value system appeals strongly to me also.

Next: Will Wright (cont.)