In general, GDC's Narrative Summit sessions focus on the task of improving game storytelling techniques. However, Schell Games CEO and Carnegie Mellon University professor Jesse Schell closed out the GDC Narrative Summit yesterday with a talk called "The Future of Storytelling: How Medium Shapes Story" that took a different tack.
Rather than offer insights on contemporary day-to-day game development, he framed the talk in terms of the future of game narrative, and what it would look like -- and the picture he painted looked very, very different.
Identifying narrative weaknesses
Schell began by posing a question to the audience: "Are we going to have a Shakespeare of games? A game that was told so perfectly, and so well, that 200 years later people will insist we play it exactly as it was?" With that challenge in mind, Schell identified a few of game narrative's endemic weaknesses that stood in between the medium and its Shakespeare. "One of the weaknesses is a problem of verbs. Video game verbs tend to be running, shooting, jumping...watch a movie, and the verbs are different -- talking, asking, pleading...video games are really good at the below-the-neck verbs."
"Another problem," Schell continued, "We suck at tragedy. It's not really a thing for us. If we're doing interactive Romeo and Juliet, what happens? Oh my god, she died. Go back to the beginning. Go us. Not everything has to be a tragedy, but it's unfortunate that this is off-limits to us. We create ridiculous stories, with zombies and monsters and time machines, and we set it up so that there's always going to be a happy ending."
According to Schell, these narrative weaknesses were best summed up by USC Games Institute's Chris Swain: "Film wasn't taken seriously as a medium until it learned to talk. Games are waiting to learn to listen. I suspect he is dead-on right."
Learning to listen
Having identified the fundamental limitation of current game narrative techniques, Schell then ran through a series of technologies that he thought could offer games the chance to "learn to listen," from using player facial-tracking features to read a player's emotional state, voice recognition and natural language parsing tech for conversations (with examples ranging from Hey You, Pikachu!
to iOS's built-in intelligent assistant Siri). "Everyone always thinks [natural language understanding] will never work," Schell said, "There have been people working on [natural language understanding] for 30-40 years, and it's starting to pile up."
Part of Schell's vision of the future of game narrative involved a fundamental paradigm shift in terms of how developers thought of video game characters. "Mario is a cool character, but he frustrates the hell out of me," said Schell. "At the beginning of each game, Mario always says, 'It's-a-me, Mario! Enter your name.' 'Enter your name? Mario, my God, we've played together for 30 years! And you don't remember, do you?' We have persistent databases for each game; why don't we have persistent databases for each character?"
Instead of thinking of characters as in-game avatars for the player, Schell suggested we think of them instead as "Virtual Companions," citing Mass Effect 3
's voice command during an in-game conversation as a simple example. "I say something [to Commander Shepard], and then Shepard says something else, and it's like we're a team. It's like we're buddies, which is a little weird, we've never had it before, but is it really that different from the relationship we've already had?"
Where characters are static, Schell said, virtual companions could evolve: "Normally avatars are made for a certain age, and games are made for a certain age, and when we change, we're going to leave them behind. We're going to want them to change and grow as we change and grow. Which is a huge challenge for us as designers, but if you could do that, damn, that'd be powerful."
Shakespeare of games
"Our Shakespeare won't be a teller of tales, but a crafter of characters," Schell said. "Someone will make characters you want in all your games...as a companion for life...World of Warcraft
is going to be here for 20 years, 30 years. What should we do when characters die? Should we bury them? Or should we pass them on to our descendants? Think of a world where the best way to know our ancestors is through their virtual companions."
Finally, Schell concluded his talk by stressing that this work would be done by game developers themselves. "I know how it sounds weird, but tell me how this will not happen," Schell said. "Who's going to make it? Not Google and them -- it's going to be us. Was Siri first? No. It was Hey You, Pikachu