How many stomachs does an interface have?

Interaction design for a game is a difficult task, because it requires the designer to create a classification system that has a limited number of classes yet is flexible enough to represent a great number of objects in the game world in an easy-to-access way. Often this is a matter of the distribution of the semantic workload between the graphic interface and the various other input peripherals (such as mouse, keyboard, or gamepad) Since each input device has a unique character, the result of the combinations can be highly different.

Looking at the various input devices that we have at our disposal today while playing games (keyboard, mouse, guitars, steering wheels, Wii controller etc), we see that they can work on very different levels of abstraction. For example a keyboard enables spelling on the phonetic level. On the other hand, the buttons of a mouse or a gamepad are more like ideograms. Combinations of these various input devices will be a basis for various naming and calling conventions, all of them constructing worlds with different qualities, quantities and behavior.

However, often the success of interaction design will depend a lot on how the graphical interface will digest the game world and decompose it into the semantically nutricious pieces that can be than passed over for further treatment to other “stomachs”, such as the menu bars, keys and controllers. In that sense it could be said that interaction design is like establishing a digestive system with many stomachs, like for example that of a cow (hence our initial question: how many stomachs does an interface have).

To digest or not to digest, that is the question!

Here, however, the digestion works on a linguistic and semantic level. Often, the graphic interface will make the first big digestion and break down the virtual world into a group of semantic and narrative cores which will be then ready for processing on the next level, that of the button arrangement on the gamepad for example. The classification that decomposes the game world into units will set the constraints for key assignments or the type of input device that will be sufficient to turn around things (For example interaction design of RTS games for concoles is known as a true challenge). Through this digestion on several levels, the game world turns into a classified system of existents and interactibles that enable the player to take a course of meaningful action in the very context that the same digestion process has constructed. Achieving the appropriate levels of categorical sensibility through this process of decomposition will be mostly depending on the finesse that lies behind the design of this system of stomachs.

Through its system of classification, the interface creates an order: The random mix of objects turns into a world with recognizable patterns.

Let’s just remember some good examples for now (Who likes to think of bad examples anyway?): Diablo had a wonderful simple point-and-click interface which had classified the game world into three categories of existents (of which all could be simply unified under one universal category, “targets”): these were killables, collectibles and destination.  The player who used the mouse had no need to further explain what action she wanted to take. To click on the interface, depending on where the mouse was resting at that moment, either meant “kill this”, “collect this” or “go there”. A further category of actions could arise from an additional decomposition provided from the mouse: left-click and right-click could mean “kill this (with this)” or “kill this (with that)”. Continuos clicking would repeat the action, which meant that the number of actions was also easy to give in. In short: The various “stomachs” had decomposed everything perfectly, therefore the game had a very fluent process of “spelling”. It got particularily well along with the very addictive reward schedule. All this created a great, zero-friction game flow.

Another example is the infamous The Sims. The Sims also had a very digestive interface, but this time one which nested almost all “verbs” into the two interactible classes, “objects” and “characters” (the difference here is one based on humanistic assumptions, technically there is no difference between “character” and “object” in the way The Sims decomposes its world). It constructed a multi-layered deep world with this method. The player wasn’t bothered with much spelling procedures to express the actions she wanted to be done, she just chose them from a palette. The dominant “sentence structure” or grammar that allowed for communication between player and game world was:

Subject  – Object  – Verb

For example:

Character A – Character B – kiss.

Or:

Character B – Fridge – Prepare Dinner

Each sentence that the player constructs was displayed within the frame of the screen via a system of “uniform” icons on top of the screen, so that the player could edit the “paragraph” of events that he had just “written” through the orders she dispatch onto the characters. If she didn’t want a sentence to be carried out, she just deleted it from this “visual list” of future events. This high level of digestive groundwork was reducing the semantic workload of the controller a lot and gave the player great time and freedom for strategic thinking.

Modern gamepads can take over a lot of the semantic workload in todays games though. In fighting games like Tekken a player may “utter” a great number of attacking and defending moves without any procedure hat involves the use of the interface. But as much as it seems like the graphic interface does not do a lot of work in this genre, it is again the way in which the world was decomposed through it, that allows the game design in general to focus on detailed gamepad use during the spelling of actions.