Please note that I don't update this web site anymore. You can visit my new personal blog at

Innovative Game Design - Where I Want To Go

Monday, February 21, 2005

I've always had clear ideas about which direction I'd like to see computer games move in, especially when it comes to storytelling. The Innovative Game Design Symposium I attended last Friday not only reinforced my believes in that regard, but provided me with new points of view on the subject. Not much has changed about my idea of the games I want to make, but I gained a lot of new ideas on how to make them, and how to make them well. In this article I take the lessons I've learned at the symposium and incorporate them into my views on how to innovate game design.

My talk with Celia Pearce, one of the speakers at the symposium, especially opened my eyes to different approaches to building an engaging game environment. I'd like to thank her for taking the time to talk with me and share her views. Much of what you'll read here is directly inspired by what she told me. Which doesn't necessarily mean she agrees with me. :-)

Narrative in games

I like a good story. Since I got involved in computer games, about twenty years ago, I have played many an adventure game. Leisure Suit Larry was the first one I played, Quest for Glory is still one of my favourites. Monkey Island, Day of the Tentacle, Grim Fandango, fond memories. And recently I've been amazed by the thoroughly involving 5 Days A Stranger. But over the years I've come to realise that adventure games do not make good interactive entertainment.

Traditional stories, like the ones we know from books and movies, are linear. They take you from point A to point B to point C. The author comes up with the proper order for the story and you take in the story in that order. This works well for most media, because they are linear in nature themselves. You can't put the paragraphs in a book in random order and start reading. Watching scenes from a play or movie in a way that wasn't intended by the director will probably ruin the experience.

Interactive entertainment works differently, though. The interactive medium doesn't force a particular order upon you. When a computer game ignores this feature of the medium and let's you step through events sequentially, the experience feels artificial, limited and forced. You can't help thinking: "I see no reason why I cannot do this. I should be able to do this."

Designers of computer games that involve narrative - adventure games or otherwise - have tried to come up with a solution to this problem. This has resulted in quite a few games with branching storylines. At a certain point in the story you can choose to either take route A or route B. This approach might give the player an extra choice and slightly increased replayability, but the resulting game is still linear. At any given time you can only play variant A or variant B of the story, so in a single playthrough, you are still limited to a single, linear story.

Adding more branches is not a solution. The path the player decides to take is still a sequence of events in the order the game designer has determined. Replayability is a weak argument in favour of branching stories, because branches can only create relatively small variations on the overall story. The player will probably get more enjoyment out of playing through an entirely new story, even if it is linear.

Game environments

I love becoming emerged in a story world. The concept of non-linear storytelling is hard to grasp for most people, because we are so used to our stories being linear. The first step towards understanding non-linearity in stories is by thinking of story worlds. There are many existing story worlds that millions of people love: the universes of Star Trek and Star Wars, the fantasy worlds of the Discworld and Midkemia, the recognizable worlds of Friends, Sex and the City and soap operas.

If you are engaged in a story world, you look beyond the seperate stories that the world is made up of. Even a bad episode of you favourite TV show can be enjoyable, because at least you get to see familiar characters and experience recognizable situations. Being engaged in a story world somehow gives you the feeling of being part of it.

To make full use of the medium at our disposable, we should design interactive entertainment to be like story worlds instead of like stories. The events that take place in the game world we provide are of much less importance than the overall immersiveness of the environment. We should look for ways we can invoke a certain feel, not unlike the quality without a name that Christopher Alexander describes in his book The Timeless Way Of Building.

One of the things you must understand when you are designing a game environment is that you are not in complete control. If you strive to provide your players with an interactive experience, you should not want to determine the way the player perceives the game world. The player needs to be involved in the game world and whether that happens is determined by the degree in which the player can shape the world to her own likeness, not by the degree in which you have shaped the world for her.

The power of suggestion is an important tool in your design. This means that you should be looking how to leave elements out instead of how to put more features in. An open environment that adhers to only a few rules that are easy for the player to understand allows maximum freedom. This increases the chance that a player can make the world her own.

Emergent systems

To create a world where the player is free to do what she wants instead of what you as a designer has thought out for her, you need to create an emergent system. I borrow this term from the book Rules of Play by Katie Salen and Eric Zimmerman, but I've seen it used elsewhere, too.

An emergent system is a system with only a small number of simple rules, but the application of those rules allows for an infinite variety of interesting results. An example of such a system is the game of chess. The rules to chess are fairly simple, but the possible strategies that emerge from these rules are virtually endless and understanding the strategic aspect is essential to mastering the game. None of these strategies are written down in the rules, however.

Another example of an emergent system could be a physics-based simulation of the world we live in. You can fill the world with objects that have physical properties, but no pre-programmed behaviour. This allows the player to do with the objects whatever she wants, as long as she stays within the rules of physics. For example, she could find out what's in a box by smashing it with a rock, dropping it from great height or drilling a hole in it to take a peek inside. Of course, she could also just lift the lid. None of these possibilities have to be explicitly designed and programmed. The emergent nature of the system provides all these options without requiring them to be part of the rule set.

Designing an emergent system is not easy to do, because you can't determine where it all leads. By their very nature, emergent systems cannot be fully explored. That's exactly what allows them to provide such an interesting environment. Of course, such a system could be seriously flawed. It's possible the player finds a loophole within the rules that completely ruins the game. That's just something we will have to live with. As far as I know, nobody ever proved that there is no winning strategy to chess, but that doesn't stop millions of players enjoying that game.

Filling the game world

When you have designed a compelling environment there are lots of ways you can have the player engage that environment. I'll mention some possibilities I can think of and I'm convinced there are many more. I also see no reason why you can't combine some of these possibilities, as long as you're careful not to overdo it.

If you still want story in your world, you can disperse story fragments throughout it. This is somewhat like an episode of a series like Star Trek, but the story fragments should be much smaller in scope. You must also let go of the notion of linearity (again). The order in which the player consumes the story fragments is completely undetermined. Don't worry too much about this. Humans have a great capability to fill in the elements of the story that are not explicitly present. They might fill in the story with fragments that differ from the ones you imagined, but there is nothing wrong with that. People will have no trouble making small adjustments in their perceptions of the world if it should become necessary.

A surefire way to give the player the feeling she is involved in a game world, is by having her build part of it. Of course, you'll have to decide to what extent a player can produce her own content. This can range from something as simple as an RPG-like character stats sheet to something as involved as modding the Unreal engine. Somewhere in between are building sims like SimCity and real-time strategy games where you have to build a base of operations. But the possibility span is very large, so how hard can it be to come up with something that hasn't been done yet?

Exploration can be a strong motivation for being part of a game world. There are lots of ways you can have the player explore. The easiest to imagine is probably the exploration of new locations. It's certainly not the only one, though. Looking for story fragments, as described previously, is a form exploration. Interaction with other game characters, whether controlled by the computer or by other players, is also a form of exploration.

Introducing a social aspect to the game world is a good way of providing extra depth. Humans are social creatures. We love to be in the company of like-minded people. Online games offer a perfect opportunity to allow players to interact with each other. You should think beyond just slapping on a chat room and ask yourself how social interaction can strengthen the immersiveness of the environment. If the game world allows exploration, can you provide a meaningful way for players to explore together? If the game world contains story fragments, can you make the players story fragments themselves, allowing them to exchange their knowledge? If one player can build part of the game world, can a group of players build the world in cooperation?

What's left?

What I've described in this article are some aspects of interactive entertainment that I'd like to explore and that I hope others will explore as well. I am by no means trying to define an end point; I don't think there is one. If we ever reach a time where we can create truly engaging game environments, then I'm sure we can find other directions in which we can innovative.

Nor do I want to imply that innovation can only go in the direction I propose. I have my personal likes and dislikes when it comes to games and I draw from them, but you might experience games totally different. Hopefully, you will share your point of view with the rest of us, so that we can learn from you and challenge our own preconceptions. Sure, I like to be right at times, but not nearly as much as I like to be wrong.

Back to blog index


Tell me what you think

Since I'm not updating this site anymore, I disabled comments. You can visit me at my new site: