How I made my monster - Part 2

Thus it was all change on my next project, an animated drama called Dilemmas in which the lead character narrated her life and occasionally turned to look out of the fourth wall to ask your advice. You could earn her trust (by showing that you understood her and knew which suggestions would work for her) and her friendship (by showing you were in tune with her). The overall plotline didn’t alter drastically, but the subtle variations were what mattered. Did Cathy get a lost gold pen back by standing up for herself or by lying? She’d walk away from that with very different feelings – about herself and about you – depending on the advice you gave.

But who are “you” in an interactive story like that? I deliberately didn’t want to define it. You’re Cathy’s imaginary friend? Her conscience? It doesn’t matter. You’re you, the viewer, the same you that Michael Caine addresses in Alfie, or John Cusack in High Fidelity. The effect is to draw you right alongside the character, with the added benefit that you don’t have to know any Crimean marching songs. You are most emphatically not a “player-character”.

I applied this model of interactivity in various other projects, including the Sims-type PC game I designed for Microsoft at Elixir Studios, and in a Law and Order Xbox game for the American Film Institute’s digital content lab in 2006:

Characters learn whether to trust the player. This occurs in the context of adventure-type gameplay, either when advising the character what to do or when interrogating a character. Advice that goes against a character’s nature or that turns out to have a bad effect will cost you trust. You might get to a point where Detective Stabler has found some evidence but he won’t show it to you. So if that happened you’d have to find a way to earn back Stabler’s trust.

Around 2009, Jamie Thomson and I started thinking about what we could do with our old gamebooks as apps. Pretty quickly, we realized that apps were a game-changer because we could pack all the nerdy stuff like hit points and inventory under the hood. A few minutes later, we realized that meant that the most interesting things we could do would be all-new gamebooks, where the variables wouldn't be hit points at all, but things like trust and compassion. We could write gamebooks that weren’t games any more but could now legitimately be called interactive literature.

(As an aside: why “interactive literature” not “interactive fiction”? Because I’m talking about books, while fiction includes TV, movies, comics, plays.)

Okay, so all that remained was to get a publisher interested. Read that last sentence as dripping with irony – I had about two years’ worth of publisher meetings with Spark Furnace’s long-suffering but indefatigable agent, Piers Blofeld. We pitched the idea of ghostwriting digital books for brand name authors, of interactive nonfiction, and original interactive novels to be released in serial form. Finally, in July 2011, I met with Michael Bhaskar at Profile Books and unveiled my plan for rewriting classic novels in interactive form:

There are two options for handling the interactivity. The traditional gamebook style is second person: “Tongues of flame dart round the bed: the curtains are on fire. In the midst of blaze and vapour, Mr. Rochester lies stretched motionless, in deep sleep. Do you try to wake him ●, call for help ●, or attempt to extinguish the flames ●?” Alternatively there’s a first person approach, which I favour, in which the narrator asks the reader for advice. This lends itself especially to epistolary novels: “‘Can it be that the count sleeps when others wake, that he may be awake while they sleep? If I could only get into his room! But there is no possible way. The door is always locked. Unless – Mina, dare I attempt to climb the castle wall, as I have seen him do, and enter by his window?’ Yes ● or no ● ?”

The concluding part of the making of the Frankenstein book app is on Wednesday.

When a character asks you for advice

About ten years ago, Leo and I developed a show called Dilemmas for Flextech. It was an interactive cartoon for 9-12 year-olds featuring a teenager called Cathy who would often break the fourth wall by turning to ask the viewer for advice.

Dilemmas arose out of adventure games, but it always bothered me the way those games would focus on things like how to stack up the crates in the right order to reach the rope that you could tie to the hook... You know, puzzle stuff. It treats a story as a problem to be solved. The really interesting choices in a story are the personal ones: white lies, temptations, keeping your promises, letting down a friend, etc.

Not that Dilemmas was about picking the right moral path. That's just another kind of puzzle: "Well done, you score 20 Niceness Points". Instead, in Dilemmas, you had to build a relationship with Cathy. She would almost always take your advice (unless it was really dumb) but the outcome often depended on how well you knew her. Things could go badly wrong if you hadn't judged the course of action that suited her best. She was actually quite an effective liar, for example - so suggesting the fast-talk option often worked. But she didn't always feel good about lying, and that would have an effect too.

Some of the outcomes might appear better or worse, but the important thing was that whatever you suggested for Cathy to do, you'd get a story. There was no fail-and-start-again stuff like in a game. And she remembered the advice you gave her, and whether it got her into trouble, so there was that sense of advising a friend rather than steering a puppet-like character through the hoops of a plot. Over time, you could earn Cathy's trust and then she'd open up to you more.

As I said, Dilemmas was targeted at 9-12 year old girls - not, in 2001, considered a very big potential games market. And it suited a style of play where a bunch of viewers would sit watching on the sofa, calling out suggestions or letting Cathy get increasingly impatient till she did something off her own bat. Back then, not a chance. Flextech asked us to work on a steampunk thing called Dr Mysterio instead. But I always liked the idea of interacting as confidant to the lead character, and in the end I got to use it in a very different kind of story: Frankenstein.

The images above are not from Dilemmas, however, but from the game I designed at Elixir Studios, Dreams. That too was about making relationships with the characters. Our design motto was: "This is my town. These are my friends. Here are our stories." And the scene here (a snippet from a longer story arc involving a love triangle that you could nudge, chivvy along and generally meddle in) is not a faked-up sequence; it was a genuine, working part of the game. More about Dreams next time.

Stepping into the story

With my book app of Frankenstein out next week, I’ve had several reviewers asking about the first-person interactivity. Unlike a traditional choose-your-own gamebook, where “YOU are the hero”, in Frankenstein the main character narrates the story and turns to you for advice. You’re his Tyler Durden – or maybe I mean his “Jack”.

This form of interactivity is new to gamebooks, as far as I know, but certainly didn’t leap into being just for Frankenstein. It’s an idea prototyped in various interactive drama projects that I’ve developed over the last ten years. I’ll talk some more about some of those projects next time, but first, by way of introduction, here’s a piece I wrote in Game Architecture & Design (1999) that started me thinking along these lines:

Most of the time when you’re playing a game, what you’re doing is just that – play. You’re exploring, running, jumping, or thumb-twitching furiously to unleash a barrage of bullets. Which can be lots of fun, but a game that only had that going for it wouldn’t be fun for long. Gameplay enriches the experience. It’s the difference between idly kicking a football about in the back yard, or setting up a pitch with goals and having a match.

But gameplay is only one way of interacting. Consider a sandbox-style driving game like GTA or The Getaway. If it consisted of only city streets for you to drive around in, that kind of game would soon pall. The first thing you want to do is be able to interact with your environment. Bump another vehicle off the road, smash the doors off a police car, run somebody down. It’s violent fun – but it’s still fun. Kids play the same way, building bricks and knocking them down. So you need to have enough opportunities to interact that the setting becomes more than a backdrop; it becomes an environment.

A game like that will be better still if you can, for instance, knock over a fire hydrant and leave it spouting water. Then that pool of water makes the road slippery so that a passing car skids onto the sidewalk. A pedestrian has to dive out of the way. He hauls the driver out of the car and punches him. Now, what if the driver identifies you, the player, as the source of the trouble? He points, the pedestrian looks your way, and then both characters start yelling at you. That way, the interaction has created a story with tension, the promise of action, and emotions running high.

As an aside, this cascade-of-events kind of story emergence may seem an ambitious goal, but it was part of the Dreams game I worked on at Elixir Studios. More on that later.

When you are playing a game like Black & White or The Sims, most of what you are doing hardly involves gameplay, by which I mean strategic choices. You might just be routinely uprooting trees or steering characters out of the door in time to catch their bus. You don't mind (you probably don't even notice the lack of gameplay in those activities) because you are having fun. But if these products weren't interactive, you would certainly notice that. Then you might start wondering if your time wouldn't be better spent reading a book, watching TV, or throwing a frisbee around in the park.

Consider the ways that a player could potentially interact with a game:

• Affecting the game world, either by changing the difficulty settings or (in simulations and god games, for example) as part of the game itself
• Directly controlling the actions of a character or group of characters
• Influencing a character's actions at one remove (for example, by giving the character advice, clues or items, like Zeus aiding a favoured hero)
• Influencing a character at two removes (for example, by leading him somewhere or pointing something out for him to look at, like a Muse providing inspiration)
• Deciding who to follow, rather than what happens – an invisible observer flitting between narrative strands
• Selecting what is interesting to you personally and making the game give more time to those elements - as any child will, when told a story at bedtime

How many of those kinds of interactivity do you commonly see implemented? The first, certainly, but only insofar as I can make the game easier or harder, or change the speed. Why shouldn't I also be able to say to a computer opponent, as I can to a human player, "Hey, let's build up a big army before we start fighting," or, "Don't attack me because I just want to have fun building a city"?

The second kind of interactivity in that list – direct control of the characters – accounts for pretty much everything else. But interactivity doesn't have to be about what happens. It is often more interesting to think about why something happens, or how it caused other things to happen.

"If you are a process-intensive designer [...] then the characters in your universe can have free will within the confines of your laws of physics. To accomplish this, however, you must abandon the self-indulgence of direct control and instead rely on indirect control. That is, instead of specifying the data of the plotline, you must specify the processes of the dramatic conflict. Instead of defining who does what to whom, you must define how people can do various things to each other."

Chris Crawford, writing in Interactive Entertainment Design, April 1995

Games guru Chris Crawford talks about process (the laws of the game world) as opposed to data (the details of what happens in the game world). To illustrate this, think back to some of the early discussions of interactivity using broadband TV. Systems were proposed in which the viewer could decide what the characters did. Essentially, they were talking about that unlamented genre the interactive movie.

Would that work? Imagine an interactive version of the television series ER. You get to choose. Does George Clooney obey the hospital rules, does he give the experimental drugs to the sick kid, or does he just go outside and play basketball? Is that a rewarding way to experience interactive drama?

No. It assumes that the point of all drama is simply to find out what happens next. If it's good drama there ought to be a whole lot more to it than that. In the hypothetical example, is the point of the player's choice to guess what George Clooney’s character would really do? Then the interactivity reduces merely to an empathy test. Is it to make him do things he wouldn't normally? Then the interactivity is just a nerdy way to test the storyline to destruction. In neither case is characterization explored in any detail, and the only theme we are apt to uncover in this story is that it's kind of weird when people behave inconsistently.

Now consider a different way we could have played our interactive ER. This time we don't get to choose what the characters do or say. Now all we can do is choose which character to follow around the hospital. Dr Carter quarrels with Dr Pratt over a diagnosis - both storm off. We follow Carter, who gives his version of events to a third character. Later we switch back to Pratt, who by now has spoken to various other characters and is pursuing another agenda. It is only later, if ever, that we may stumble upon the resolution or aftermath of the quarrel that we witnessed at the start.

This example in itself isn’t a sufficient condition for interesting interactive drama. But it illustrates an important point. All drama, whether interactive or not, derives momentum from the viewer’s relationship with the characters. A rule often quoted in writing classes is that a character has to change and develop – hence the ubiquity in screenwriting of the redemptive story arc. In fact, characters don’t necessarily have to change to be interesting, but our understanding of the character has to change. We need to see things that will build on our first impressions and, ideally, that will surprise us and force us to reassess those impressions. That’s because the viewer is part of the system – and if that’s true of non-interactive media, you can bet it’s even more so in the case of games.

The aim of all storytelling is to make the audience care. Interactivity is a new(ish) and powerful technique in the writer’s repertoire for achieving that goal. The ER example is only arms-length involvement, in that you are still interacting as an external observer. It’s better than pressing a button every time you see a white rabbit, or brushing virtual dust off text so you can read the story – but it’s only just better. However, it is also possible to offer interactivity that draws the reader or viewer into the story, by developing a direct relationship with the protagonist. That’s what I used in Frankenstein and, ten years earlier, in a teen drama for Flextech called Dilemmas, which I’ll talk about next time.

Interesting choices

This is from a project Leo and I worked on in about 2001 called Dilemmas. It was an interactive animated TV show about a teenager called Cathy who would often break the fourth wall by turning to ask the viewer for advice.

Dilemmas arose out of adventure games, but it always annoyed me the way those would focus on things like whether you could stack up the crates in the right order to reach the rope that you could tie to the hook... and so on. Those are the least interesting elements of any story! The interesting choices are the personal ones: white lies, temptations, keeping your promises, etc.

Not that Dilemmas was about picking the right moral path. That's just another kind of puzzle set by the designer: "You score 5 Niceness Points". Bah. Rather, you had to build a relationship with Cathy. She would almost always take your advice (unless it was really dumb) but the outcome often depended on judging the course of action that suited her character best. She was actually quite an effective liar, for example - though she didn't always feel good about it, and that would have an effect too.

Some of the outcomes might appear better or worse, but whatever you suggested for Cathy to do, you'd get a story. There was no fail-and-start-again stuff. And she remembered the advice you gave her, and whether it got her into trouble, so there was that sense of advising a friend rather than steering a puppet-like character around.

Trouble is, Dilemmas was targeted at 9-12 year old girls - not, in 2001, considered a very big potential games market. And it suited a style of play where a bunch of viewers would sit watching on the sofa, calling out suggestions or letting Cathy get increasingly impatient till she did something off her own bat.

Back then, not a chance. But in the post-Wii era, who knows?