Attending the PLAYER Festival of live gaming at the London Science Museum this last weekend, I took part in Seth Kriebel’s excellent Unbuilt Room performance, a live text adventure for five players. The game is structured as an 80s-style text adventure (think Zork or Adventure), with the audience taking it in turns to move the protagonist through a maze roughly mapped to the human brain. The game lasts twenty minutes, and after several early fumbles we ran out of time before we solved the last puzzle.
Seth Kriebel’s The Unbuilt Room.
In correspondence after the festival, Seth advised that only 20% of playing audiences ‘completed’ the game. Far from being a frustration or concern, my experience with the Unbuilt Room was that our failure to solve all the puzzles in time was one of the most exciting things about the game. At its simplest level, I am now imagining all kinds of exciting endings to the story, to which the real ending cannot possibly compare. At another level, the Unbuilt Room demonstrates one of the key principles of interactive performance which Boho have discovered/rediscovered through our work: the audience needs to be allowed to fail.
Failure serves a few purposes – firstly, it’s a way of demonstrating that the interactivity is real. As an audience, if I go into an interactive work and manage to somehow get everything right first time, I might be convinced that I (and my fellow audience members) am a genius. But I’m probably going to assume that the game is rigged, and not really as interactive as it claims to be.
Secondly, failure is the best (possibly only) way to learn. Without the opportunity to make (and correct) mistakes, an audience is only rewarded for playing it safe. Trial and error is an intrinsic part of learning – trial without error is just a rather pallid fantasy.
How to build failure in to your games without discouraging your audience is a more delicate exercise, and one with which Boho is continually struggling.
Playable Demo. Image by ‘pling.
One method was in our very first interactive game, Playable Demo (which later featured as a module within A Prisoner’s Dilemma). Playable Demo was an adventure game which an audience member operated using a torchbeam as an onstage mouse cursor. The aim of the game was to help Prisoner 101 escape from his prison cell by collecting and combining objects, and interacting with his guard. We quickly found when we handed the controller over to the audience that the first thing they wanted to do was direct Prisoner 101 to harm himself (no surprises). And, naturally, we wanted that too. So there were numerous easter eggs built into the game where 101 was bludgeoned by the guard for making mistakes. Our fear was that people would simply toy with the character, rather than trying to solve the puzzles we had devised for them.
Our way around that was not to minimise their chance of failure, just to make the pay-off for failure less and less rewarding. It’s as simple as repetition – when Prisoner 101 hit his head on the cell door the first time, it was hilarious. The second time, still very funny. The third and fourth time, producing the same result, less so. The audience quickly learned that they could play the game to lose (the interactivity was ‘real’, so to speak) but that there was more fun to be had in trying to solve it than in trying to break it.
Jack Lloyd, David Finnigan and Cathy Petocz. Image by ‘pling.
One other lesson was to get to grips with the fact that an audience can really, seriously go wrong, and when that happens, you need to let them work it out. In True Logic of the Future, we had a puzzle built around the Logic Piano (a replica of WS Jevons’ 19th century early computer construct). In this sequence, two scenes played out simultaneously – one set in a hospital, and one set at the scene of a crime by the city’s dam. The audience used the logic piano to separate out the two scenes, filtering the messy sequence into its constituent parts. It basically operated with the same mechanic as Mastermind, but with two potential correct combinations – I’m sure we’ll go into more detail with this game in a future post.
Most audiences ran through the sequence between 5 and 8 times before hitting on a correct combination of keys. Some audiences got it within 2 or 3 goes, making the whole thing seem quite easily. During one performance, though, the players ran through the scene 13 times without hitting on the correct combination. As a performer, that’s desperate. You can feel the frustration mounting as the audience are trapped in this same section of the play, and the concern of the players as they fear that they might not be able to solve it. And when a sequence of the show that normally runs for 8 minutes runs on to 16, you begin to freak out that the show is going to run hugely over time, and everyone will be upset.
But when, on their 14th attempt, the audience got the combination correct and solved the puzzle, the exuberant cheer of the audience was worth all the stress and panic. The excitement they had for solving the puzzle was real and tangible, and carried them (and us) through the rest of the show. Afterwards, that was the scene the audience talked about – that was the memorable moment for them. In a very real sense, that was the reason they’d come – to genuinely interact with play with a live performance that really responded to their choices.
How much you let the audience fail, or how much you let their failure influence the show, needs to be dictated by the data you’re looking to get out of the scene. In A Prisoner’s Dilemma, failure in a scene was used to dictate which of the characters in the main storyline had been tortured, so we were able to make the risk of failure quite genuine. Again, in Food for the Great Hungers, failure in scenes correlated to poor industrial relations and multicultural policies in an alternate Australia. By mapping failure to a broader element of the productions, we were able to build a reward of storyline outcomes into the momentary let-down of not succeeding. On the other hand, the purely narrative sequences in True Logic were set in stone, so failure in interactivity was used as a guiding mechanism – by letting people fail quickly, fail often, and rapidly prototype and test new ideas until something sticks – the Angry Birds model (thanks to Deloitte Digital for this analogy). And, if it’s the right story, failure can be total, as with the Unbuilt Room, severing the narrative completely and leaving threads tantalisingly untied – but only if you’re sure this is the feeling you’re aiming to leave people with.
Across all disciplines, failure is not merely an unfortunate outcome of a poor strategy – it’s an intrinsic part of any serious enterprise. People use failure to adjust and recalibrate their strategies for dealing with the world. And in a live gaming / interactive theatre context, failure can be the most fun part of the experience. Don’t underestimate the joy of making a live performer hit their head repeatedly against a wall for your entertainment.
(And sometimes you invent a scene entirely so the audience can fly the performers around stage like an old console game) Flying Dudes. Image by ‘pling.