It was the best of times . . .

New Scientist vol 162 issue 2181 - 10 April 1999, page 42

We all think we know the difference between a novel, a game and a film. But in cyberspace the boundaries are being destroyed and redrawn, says Claire Neesham

IT'S THE final act and the Prince of Verona takes the spotlight to utter:

"A glooming peace this morning with it brings;

The sun for sorrow will not show his head.

Go hence, to have more talk of these sad things;

Some shall be pardon'd, and some punished;

For never was a story of more woe

Than this of Juliet and her Romeo."

And so the curtain falls on the end of Shakespeare's tragic love story. At least, that's how it has always ended in countless performances over the past 400 years. But not any more. Naoko Tosa and Ryohei Nakatsu, two researchers from the ATR Media Integration and Communications Research Laboratories in Kyoto, Japan, have followed the lovers to Hades where they can begin working their way back into the world of light.

Nakatsu and Tosa's cybersequel is performed with the help of the latest in interactive systems, using gesture, voice and emotion recognition software. It's more like a computer game than a staged performance. Two people "play" the protagonists, each taking control of an avatar which is projected onto a screen where the action takes place. The cybercharacters they meet as they try to return to the living world respond in different ways depending on the performers' words and actions.

Entertainment explosion

Play Cinema, as the two researchers call their project, is part of an entertainment explosion spawned by the increasing power of modern computers. This and other such experimental forms will have to fight for a place in a leisure world overflowing with everything from multi-user domains (MUDs), in which people take part in role play over the Net, to gory games, cerebral adventures stuffed with demanding puzzles and texts designed to exploit the Internet's singular powers.

Already, some of these forms have become incredibly popular, but will any of them ever be celebrated in the same way as a great novel or film? Will Romeo and Juliet in Hades ever be as highly regarded as the original?

One of the curious effects of the new digital forms is that they confound our traditional ideas about entertainment and art. Many of the works emerging from projects such as Play Cinema blur the edges between plays, novels, films and computer games. Of course, some degree of blurring has happened even without the computer. Novelists such as James Joyce and Thomas Pynchon have tried to escape the confines of linear narrative, for example. And readers of fantasy fiction books, such as Star Striker by Steve Jackson and Ian Livingstone, need a pair of dice to know which page to go to next. But the interactivity possible with a computer has confused matters even further. Strangest of all, it is even making the distinction between author and audience disappear.

Not surprisingly, new artistic forms need new words to describe them. At the University of Bergen in Norway, Espen Aarseth is one of a handful of people trying to create a new lexicon. His main concern is the host of online text-based media that have appeared— including MUDs, text-based games and electronic books. All these texts he calls ergodic, borrowed from the Greek for "work" and "path". The term emphasises that the effort the reader must invest in these works is very different from that needed when reading a traditional book. Ergodic texts require the reader to make choices to reach an endpoint.

Take "hypertexts", for example. One of the innovations that gives the Web its unique power is the hyperlink, which allows you to follow a single theme from one website to another by clicking on highlighted words. A number of authors have produced hypertexts that exploit this type of link. One example is Forward Anywhere by artist Judy Malloy and Cathy Marshall of Xerox's Palo Alto Research Center in California.

In 1993, Marshall and Malloy decided to get to know each other by sending e-mails that described snippets of their lives. The result is an entertaining, three-year dialogue that discusses feelings and events from their lives (which they changed slightly to protect friends). These e-mails were then linked to create a hypertext. At the end of each e-mail, the reader is given the choice of going "forward", "anywhere" or "lines". Forward brings up the next e-mail, so the reader can follow the dialogue sequentially. Anywhere takes the reader to another e-mail chosen at random—much like opening pages at random.

Lines is something else again. It displays a search box: type in a word and it brings up lines from their other e-mails that contain that word. The result is to connect the text in a way that no writer of a traditional book could easily duplicate. The choices that the reader must make to navigate through Forward Anywhere is the extra work that Aarseth says is characteristic of ergodic texts.

This type of hypertext highlights one of the big differences between old and new forms of entertainment. You can travel through Forward Anywhere any way you choose without fear of losing the grand plot—because there isn't one. "We didn't plan a narrative," says Marshall. "We just decided a starting point." A text without a strong storyline that you can start and finish anywhere is very different from all but the most experimental novels. Aarseth argues that with this type of work, the notion of narrative makes no sense at all. "It's just a different type of literature," he says.

There is another, more intriguing distinction between ergodic texts and traditional literature. The novelist is virtually all-powerful when it comes to dictating the course of a story. While every reader may see a different meaning in the words, he or she has no ability to change the plot. In his book Cybertext: Perspectives on Ergodic Literature, Aarseth likens someone reading a novel to a spectator at a soccer match: "He may speculate, conjecture, extrapolate, even shout abuse, but he is not a player . . . The reader's pleasure is the pleasure of the voyeur. Safe, but impotent."

By contrast, some ergodic texts allow the choices made by readers to completely change a story. This radically changes the relationship between writer and reader. Suddenly, says Aarseth, the reader becomes "a player, a gambler".

Take the case of Uncle Buddy's Phantom Funhouse by John McDaid, a software consultant and postgraduate student at New York University. Readers of Funhouse become the literary executor of a vanished science fiction writer, Arthur "Buddy" Newkirk. They receive the contents of what was on Uncle Buddy's hard disc at the time he disappeared. These include such items as a digital notebook, music lyrics, a dictionary, a deck of tarot cards and a proof copy of a short story Uncle Buddy was editing.

The readers work their way through the objects rather like detectives, trying to develop hypotheses and connections that may explain why Buddy disappeared. "There is no one `right' path through the fiction," says McDaid, "nor is there any one right answer to the question of what happened to Buddy."

This form of adventure is somewhere between a game and a story. While the writer creates a world, the readers are free to explore that world any way they want. Two readers may read about totally different objects and events. "The writer sets the rules, but the explorer may experience something that is completely new," says Aarseth, "something the designer hadn't thought of."

If cybertext is converging with game playing, are computer games being given stronger narratives? Dominic Cahalin, a games designer for Sony Entertainment Europe, is doubtful. "I don't know anyone who bought a game for the story," he says. The narrative of a game, he argues, is seen as something onto which the creators can hang the game-play mechanics. The most attractive elements are competition (whether it's against the computer or a friend), the feeling of accomplishment when a task is completed, and the power element—the ability to control the system and the destinies of the "characters".

Robyn Miller, one of the creators of the incredibly popular computer game Myst, seems to agree with Cahalin. Myst is a graphical game set on an island in which the player is left totally alone to wander around, discover clues about the island and solve puzzles. Last year, Miller decided to stop making such interactive games. "Interactive is an incredible medium, but I don't necessarily believe it's a storytelling medium. It focuses on environment," he said at the time. "People are not what it portrays best, and character is what drives story."

But this may be an overly bleak view. According to Janet Murray, director of the Program in Advanced Interactive Narrative Technology at the Massachusetts Institute of Technology, the stories underlying games are improving. One of her favourites is The Last Express, a graphical adventure game that takes place on the final outing of the Orient Express before the First World War.

As the train travels inexorably towards Constantinople, you—the hero—are forced to uncover the strands of an elaborate, melodramatic plot. Whatever you do, certain events happen to a preset timetable—just as they would in a conventional story—which introduces a real sense of tension and release. "It's more like a movie," says Murray. "You can talk to the other characters, take things from them, give them things, hide from them and overhear them. It's a very rich world."

And it's not just storylines that have improved. Though Myst's storyline and characterisation may be questioned, it set new standards for graphics and music. Ever since its debut in 1993, says Murray, people have realised that "the medium can be discussed in the same sort of aesthetic terms you would use for films".

Graphical computer games, ergodic texts and Play Cinema are just a few of the myriad forms that the computer now makes possible. "The computer generates an endless number of different genres," says Aarseth (see "Once more with feeling"). What we're seeing now is the wild thrashings of an infant industry. In her book Hamlet on the Holodeck, Murray likens the situation today to the early, sometimes wildly experimental days of the film industry at the turn of the century. Yet, "In the first three decades of the twentieth century, filmmakers collectively invented the medium by inventing all the major elements of filmic storytelling, including the close-up, the chase scene, and the standard feature length."

Digital entertainment needs to go through a similar process. "Now, we're at the stage of pounding on clay," Murray says. "We're still exploring the medium." The people who are really helping to create new media are those who are trying to give people what they want by harnessing the unique power of the computer, rather than merely copying traditional forms.

And what is this mysterious digital power? Aarseth and Murray agree on the chief aspect, though they call it different things. "Simulation" is Aarseth's term. "Books are great for telling stories," he says, "but computers can create a dynamic model in which you can experience things in different ways." Murray calls it "immersion"—the sense of being in a complete world and being able to move around. No longer are people left to read a representation of a world, they can actually enter it.

Is anything likely to grow out of this experimental period that will ever become art? Aarseth reckons so, and that there will be many different genres to choose from. Murray agrees: "It's reasonable to have that expectation." After all, new media have a habit of spawing great art. Could the pioneers of film in the last century, for example, ever have predicted a film like The Godfather?

Once more with feeling

EVEN poetry may benefit from the computer's interactive powers. If your verse is lacking feeling, Naoko Tosa and Ryohei Nakatsu of ATR in Kyoto, Japan, may have just what you need. Their experimental Interactive Poem System helps people to write poetry in the mood of their choice.

At the centre of the system is a software agent called Muse, which appears on the computer screen as an animated face that expresses emotions. Muse utters a short poetic phrase to a human, who may respond with one of the phrases given to them on screen or by creating their own poetic phrase. These words are identified by a speech recognition system while a neural network, trained with the utterances of many speakers, identifies the emotion in the speaker's voice. It can spot eight emotions: joy, happiness, anger, fear, teasing, disgust, disappointment and surprise.

Once Muse has analysed the human's words and the emotion in their voice, it adds an appropriate emotional phrase and changes its expression. Human and Muse then take turns, a line each, until they have created a new poem.

Claire Neesham
Claire Neesham is a writer based in London

<<< back