Exclusive: Will Wright on Emergent Game Design (Part 2)

This is a multi-part post.  Jump to:

Part 1

In a previous post, I discussed a lecture recently given at Microsoft by famed Game Designer, Will Wright. The topic of this lecture was Emergence and Game Design. The first part of this post discussed what emergence is, how it relates to Game Design, and the problems it solves. It also discussed some of the problems it creates, specifically that it’s impossible to predict a priori whether a game will be interesting based on the basic mechanics which form it. It would be useful, nonetheless, to be able to discern particular areas of mechanics which tend to work well together to compose a whole game. That toolkit is the subject of this post.

Will’s toolkit draw many parallels to the concept of game grammar, advocated by such industry tycoons as Raph Koster. There are three independent groups which any given game rule-set will derive from. These sets, called Topologies, Dynamics, and Paradigms roughly correspond to the linguistic concepts of nouns, verbs, and grammar rules. Each of these groups is further broken down into specific techniques. Any game system will draw on a mix of techniques from all three systems, but there is no express order in which the groups must be explored.

Topologies

The first of these, Topologies, is the noun analog. Topologies represent the framework upon which the rules act, and create structure for the game environment. Interestingly, Will considers game communities to be part of topologies. A good example is the advancement progression in most games. Some games have a very linear progression, as you advance through levels and are led from one place to the next (e.g. Gears of War). Others are gated - the possibility space branches outward after each gate, only to collapse to a single node at the next (e.g. Mass Effect).

There are three outlined techniques within Topologies, from most rigid to most flexible: Agents, Networks, and Layers. Agents represent particular objects and beings which perform actions, or have actions performed upon them. In Sim City, individual buildings would be considered agents. Nearly all games make use of agents in some form.

The second, networks, represents the framework that defines interactions between agents. These linkages may be spacial (Buildings can be connected by roads), temporal (an action by one agent causes an event in another), functional (companion cubes can be placed on buttons), or relational (forests and gold mines are resource providers).

The last topology, Layers, is a technique whereby several layers of agent-network graphs can be laid upon each other to create a different facet of the same game. Battle for Middle Earth’s War of the Ring mode is a good example of this, as one game is placed on top of another game, and the outcomes of each affect the other. Different views on information (such as seeing the amount of crime in your city), or statically linked layers of graphs (In Sim City, electrical system, water system and road system) would fall under this as well.

Topologies are the most straightforward of the three concepts, and a similar concept is covered in nearly all books on game design. The next concept, dynamics, brings these simple structures to life. Dynamics will be covered in part 3.