The Gobstopper Job: stereotypes, shortcuts and stealing sweets

The Gobstopper Job. Because everyone’s wanted to run a heist at some point in their lives.

We ran the Job in Bristol at the Interesting Games Festival, in Castle Park. We were part of the fringe – a raft of oddball, interesting, quirky games that worked in public and that could be played fairly casually, in a pick-up-and-play sort of way. I didn’t get the chance to play any of the others, due to being on duty in a large plastic police hat with a flashing blue light on top for most of the afternoon, but I saw a balancing game with hand turtles, a slow-motion combat game, a coloured-water-shooting game called Rainbow Rain, and a game about phone hacking that had people in trilbies running around the park looking for mobile phones. It was all rather bonkers and lovely.

Gobstopper Job
A very few of the sweets

Gobstopper worked in part because of the space we played in – bonkers and lovely – and because of the costumes and props Grant put together. I mean, it worked because of a load of other things too, but Grant (who was the main designer on this one) has already written an interesting post about froth and emergent stuff and verbs and what have you, which is well worth reading and has more pictures. This, by contrast, is a bit of an epic.

A note on live gaming vs LARP

Despite appearances to the contrary – such as the name Zombie LARP – I don’t really think of what we do as LARP any more. It differs in some pretty major respects from what I’d instinctively label “traditional” LARP – though I’m certain that’s the wrong label – it’s fundamentally neither plot nor character driven, but situation-led and responsive to player/character action. Its worlds are small, not entirely internally consistent, and exist only for short periods before they are dissolved; their parameters are fluid and they are never fully realised. The lines between player and character are deliberately blurred.

But neither does the term “live game” or “pervasive game” (whatever that is) suit us. Most live games I see on the scene at present offer limited or no opportunity for player characterisation beyond the opportunity to nebulously pretend. They are mostly unaffected by character action beyond the mechanical, and function within the real world – without a need for suspension of disbelief.

Cops and robbers
Cops and robbers

In Gobstopper, as in Zombie, I saw the emergence of distinct play styles. Some people come to Zombie to LARP – they bring characters, immerse themselves in a story of their own making, and use our world and mechanics as a way to play that story out. This is improv theatre with rules. Others come to live game – they are uninterested in roleplay, preferring to focus on the mechanics of the game rather than the story, and lose themselves in the adrenaline of the moment. Some come to do both.

The Gobstopper Job, like Zombie and like a few other games – 2.8 Hours Later and Incitement spring to mind, in different ways – operates on the lines between those two categories. There are no good terms for what we do, and it’s difficult to suggest any that don’t imply that other games that fit the genre boxes more neatly are somehow deficient. Story-driven live game might be one option; emergent short-term LARP might be another. Both are clumsy. At present, we lack the right terms.

Everyone gets a gun on the mantelpiece

Gobstopper was the first time we’ve really run a game outside a dedicated game space. We had a kiosk – the sort you get in parks, that might sell water or ice cream or sweets – at one end, and a base station at the other. Players dressed as robbers (well, with masks and maybe hats) had to get the swag bag into the kiosk, steal as many sweets as possible in 30 seconds, then get them back to base, all the while avoiding the police – four or five people patrolling outside, who could arrest them if they could catch them.

Policemen helping up a player disguised as one of their number
Policemen helping up a player disguised as one of their number and feigning injury

The players for each run had to pick a character class – a single thing they could do – and with it came a prop with which they could do it. Gunmen had a little bang-flag gun that let them incapacitate one policeman. Demo men had a bouncing cherry bomb with the word BOMB on it that could stun people in a radius. Conmen had big ridiculous white masks that disguised them as “Young Gavin”, the rookie, or “Old Bob”, who was only a few days from retirement. And Bag men had the swag bag, and had to get them into the kiosk – essentially a very low-tech hacker analogue.

Chekhov’s gun – the concept that if there is a gun on the mantelpiece in the first act of a play, it must be fired by the end – is a metaphor for foreshadowing, simplicity and dramatic necessity. It’s also a useful way to think about Gobstopper, and why it works. Each player gets a verb – a single mode of interaction with the game and with the world it creates – and extremely limited opportunities to use it. Every player gets to be their own dramatist, timing their climactic moment in their own personal trajectory through the game. This only works if players believe in the power of their gun to affect the world, and understand its importance. Oh, and remember to use it. Occasionally people forgot, in the excitement of the whole thing.

The Familiarisation Effect

A fundamental design problem for us is the necessity of blurring the boundaries for players between the states of being, doing, performing and playing. What we’re after is almost the opposite of Brecht’s Verfremdungseffekt. Rather than using realistic props and costume to draw attention to the unreality of the performance, we’re using unrealistic, very stylised costume and props to create a sort of shorthand of familiarity. We are reliant on caricature and stereotype as tools to build identification between the players and the world we need them to inhabit – not a perfectly realistic, immersive environment in which they can lose themselves, but a working reality with fuzzy boundaries but basic concepts and rules to hold on to.

One of the Gobstopper Job teams post-game, with their bag of swag
One of the Gobstopper Job teams post-game, with their bag of swag

So, giving players a tool they can use matters, but giving them a tool they can instantly understand and relate to is even better. Using sweets to tap into the part of an adult that’s still six years old; using BANG-flag guns and BOMB-painted bombs; letting the only people who need to improvise a conversation hide their faces beneath enormous saucer-masks that both make it easier to act and also render it unnecessary – these are all tactics that tap into familiar childhood games, helping to minimse embarrassment and flow-jarring moments. That’s before we get into the psychology of costume itself, and the impact it has on roles – here, we wanted playfulness and fun, a cops’n’robbers feel, whereas when we want to make something more serious we’ll make it look and feel more lifelike and realistic.

Public play as performance

One convenient side-effect of this approach is that it gives the audience a simple way to relate to the game. With crew in giant police hats and players in robber masks, there is an obvious metaphor for viewers to grasp and to buy into. We found both sides got cheers from passers-by, depending on who seemed like the underdog at any given moment. As policemen, we spent a fair amount of time talking to random people who saw us wearing daft hats and wanted to know what on earth was going on. We fielded a couple of people who were oddly thrilled that someone was using the kiosk for something, and directed a few folks to the sign-up desk. Policeman hats and a slow walk seem to create an aura of helpfulness around you – and serve as useful ads for the game, too.

The carnage after players swept through the kiosk
The carnage after players swept through the kiosk

The other big thing they did was help the players feel less weird about getting dressed up. They were absolutely guaranteed that the people running the game would look even dafter than they did. Obvious signalling and the familiarisation effect combined together, along with a healthy dose of physical and mental activity, got players focussed on the game and served to make it easier for them to enter that peculiar place where being, doing, performing and playing all merge together and stories seem to naturally emerge, without anyone deliberately designing them, from the seemingly natural actions of all the participants.

The moment where I knew it had worked was when one run ended and it took five minutes of listening to the players talk through exactly what had happened when and how before I could get a word in edgeways.

Next time

We’re going to be running The Gobstopper Job again, given how well it seems to have worked – and given the fact that we still have about 15kg of sweets in our front room. London somewhere, probably. We’ll be putting out info on the Facebook page as soon as we’ve made up our minds when and where. Do come.

Standing on the Shoulders of Giant Nerds

Notes and slides from my presentation at Develop 2012, on what video games should be stealing/learning from other, geekier sorts of game.

Who am I and what on earth do I do:

I’m Mary, and I’m a massive nerd. Mostly I design live games, mostly involving NERF guns. The biggest game I run is called, imaginatively, Zombie LARP, and it involves live action role playing and zombies.

Zombie was born when Grant, who’s drawn these slides, and Chris, who appears in them pretty frequently, started shooting each other with NERF guns in halls at uni. They invited a couple of us round to their house, and gave us NERF guns, and pretended to be zombies. My first Zombie LARP death happened in their kitchen, as I fumbled to reload a Nitefinder. It was awesome.

The games society at uni kindly let us run it at the 24-hour roleplay. Every year they take over a building at university for 24 hours and play games. Board, card, video, live, war, tabletop – all sorts. That’s where we got our start – playing at 3am with about 15 exhausted geeks hyper on Tesco’s own brand energy drinks. Now, six years later, we run huge games for 150-odd people in abandoned buildings.

But I also play and write about – and sometimes for – video games. And I see a lot of the development processes that we go through with live games and tabletop systems mirrored in what video games are trying to do. So what I want to do today is talk a little about the lessons I’ve learned from wiser geeks than me, playing older, more traditional, and frankly much more nerdy games. Starting with the geekiest of all: Dungeons and Dragons.

What Dungeons and Dragons can teach us about story in games:

The standard game of D&D is a really good model for looking at different sorts of narrative in games. There are four types. Not every game has all of them, but they all have at least one.

First up: the Told story. This is the one the Dungeon Master is actually trying to tell. She might have a whole host of carefully planned and pre-written events that culminate in a fantastic climax. Or, like a friend of mine, she might just have written “BLAM! IS ADVENTURE” on a piece of paper and be totally making it up. But she’s in control of the story, she’s giving out information and framing scenes to create something that adds up. She controls NPCs, she’s the ultimate arbiter of conflicts, she creates the narrative arc and leads the players through it.

Obviously a fair few video games aren’t really interested in this sort of story, or only in very simple versions of it. Things like “Angry Birds invade pig houses” or “Hungry monster wants hard-to-reach candy”. But for others, the story is a big part of the game – the designers want to convey a narrative through the game experience. That’s Told story.

Secondly, the Experienced story. This is essentially the narrative of the time you spend playing, not the game per se. It includes the infrastructure – the stuff around the edges of the game like phoning for pizza or people being late. Interruptions are part of this, and so are stupid out of character jokes, and those times when the conversation wanders off a long way from where the action is. As Michael Brunton-Spall pointed out on Twitter, it also includes things like critical hits and fumbles, where an unusually lucky or unlucky result on the dice leads to a different story being created. Generally this isn’t a story that ever gets told, as such – it gets experienced, and then your brain chops it up into representative chunks for you to remember.

In video games, as well as taking in interruptions and stuff going on outside the game at the same time, the experienced story includes loading screens, and glitches, and crashes, and dying. My experienced story in Mirror’s Edge, for instance, was about a girl called Faith who kept running off buildings and falling unceremoniously to her death. Assassin’s Creed is a series that copes very well with this stuff, by keeping the player in the story world even during loading and dying and other nominally out-of-game states. Rogue-likes make a virtue of the repetition.

And then, there’s the Interpreted story.

Interpreted can’t exist without Told stories. But the theory goes like this:

  • a story doesn’t exist until it’s experienced. That goes for games, films, and books, too. Otherwise it’s just an object.
  • the story is always filtered through the individual understanding of the player involved. Because they’re human, they’ll have different ideas and associations and resonances, a different understanding of what the various bits of the story might mean.
  • so the interpreted story is different for everyone. It changes in different contexts.

In D&D, that turns up most obviously as players getting the wrong end of the stick about something. Not realising the monster is actually being controlled by the supposedly friendly NPC, or deciding that *this* insignificant detail is actually the thing they should go chasing after for the next three hours. But it also covers players finding new levels of complexity in a GM’s narrative, interpreting coincidences as deeply meaningful – or things that are meant to be meaningful as coincidences.

Authors can’t control audience interpretations of their stories – intent isn’t magic – and there will always be a difference between what a creator wants to say with a story, and what people take from it. Even if only because historical and individual context changes. But it’s true for video games that meaning is collaboratively created, it arises from the joining of a storyteller and a story player, and that sometimes the interpreted story can work against the intention of the creator in unexpected ways – or can reinforce it.

Look at the reams of interpretation that’s been done on Silent Hill, or the readings of Resident Evil’s racial politics. There are elements placed there by the designers that have been drawn out by players to reveal a wider, deeper, perhaps more problematic – perhaps unintended – story. Those readings aren’t invalid just because they might not be intended.

Finally, there’s emergent story. Emergent story is my favourite kind. It’s not quite the same thing as emergent gameplay, which in D&D terms would be doing something like making a Grease wizard who only casts Grease and Fireball spells, and is very good at bringing down monsters with low Dex scores like T-Rexes.

Emergent story is what happens when players get to futz around directly with the mechanics of a game, within a framework. It’s the little, unexpected moments that aren’t directly intended by the creator, but that can be the most memorable bits of the game: roleplaying conflicts among the party, for instance. The decision by one player to take the skill Crafts: tailoring, and then go on a quest to acquire the finest suit in all the land. A group of players deciding to jack in the monster hunting business and go be pirates. Or the moment when the characters react to the death of one of their number.

Video games are not great at this. Live games are brilliant at it, and tabletop is not far behind. At Zombie, this is the sort of stuff we thrive on. We deliberately designed the system to encourage emergent stories – to be flexible enough and simple enough for spontaneous moments of awesome, which is what we originally called it. But the flip side of doing that is that for some players, those moments never happen – and if you fail to balance that right, it can make for boring play.

The games that are doing emergent story well right now are MMOs – Eve online, in a huge way, but also WoW and others. Minecraft & the Sims manages a much more individual emergence – the action of a single player in a procedurally generated world – and perhaps the current pinnacle of emergent story in gaming is Day Z. But this is such a young area in video games. There’s so much more to do.

What Dogs in the Vineyard can teach us about ludonarrative dissonance:

Ludonarrative dissonance is a gloriously useful term that describes conflict between game system & game story. Dogs in the Vineyard is a game where you play, basically, deeply religious teenage Mormon virgin gun-toting priests in the wild West. The whole system is built around the setting, and the types of stories you can tell with it. It provides a way for storytellers to building towns that have heresies within them, and giving your players the tricky job of resolving them. It’s a game about hard moral choices, and not having the right answer, and not being sure, and risking things – gambling your reputation and the happiness, security, and lives of others on being right, when there is no “right” to be.

The system you use for conflict resolution is based around stakes, and bidding. It feels a little like poker. You might start out arguing, verbally, and you roll a pool of dice that runs off your verbal skills, and then you have an argument – you bid dice from your pool against dice from your opponent’s pool, till one of you can’t beat the other. Then your choices are to back down, or to escalate. You could move up to physical violence, or even gunfighting, which gives you a new pool of dice – but your opponent can escalate too, and then you’re risking a lot more damage when everything shakes down.

The way that plays out in practice is thematically consistent with the game story – because the game story is running off quite a limited set of themes and ideas. The stories all work off trade-offs, being unsure, raising the stakes because the option of doing nothing isn’t an option at all. There’s no dissonance – the game’s smooth, the experience is unified – because the system, the setting and the story have been developed together, organically. The textures mesh. The experience feels continuous. It’s a deeply fulfilling, immersive, gratifying game to play if you have a half-decent games master, because all the pieces work together.

There are some video games that do this beautifully – Manhunt, for instance. Project Zero. Spec Ops: the Line, interestingly. And then there are games like GTA4, LA Noir, Uncharted – games where the character you control can do things that the cut-scene character just wouldn’t consider. The character that you’re creating and inhabiting as a player is different to the one in the cut-scenes – which breaks immersion, badly, but also breaks the player’s creative, narrative play by disregarding the choices they’ve made. That stops the play experience hanging together, and can make the player feel less invested in the world – after all, if their actions mean nothing once the cinematic starts, then the imaginitive “work” they’re putting into the game isn’t being respected.

What Mage: the Ascension Mind’s Eye Theatre LARP can teach us about ownership:

Mage, as you might imagine, is a game about wizards. Mage the Ascension was about secret wizards in modern-day cities, each with their own understanding of how the world works – their own subjective interpretation of the universal system mechanics. Mage the Apocalypse was a live-action version, where the setting was in a post-apocalyptic world where mages could band together and kill a fairly ludicrous range of monsters of the week – including aliens shaped like flying hats, and vampires and so on.

So. Some people had been playing that game for more than three years by the time I joined – going four times a year or so to events, planning characters, creating costumes, writing backgrounds and events between games, scheming, plotting.

Then, in a forum post between games, the storytellers announced that we’d missed the signs of the impending Cthulhu apocalypse proper, and the unpleasant things crawling out of the sea were actually Dagon-based beasts, and we were all dead.

One very newly recruited storyteller had to deal with what happened next. A group of very invested players who loved that game and felt personally, deeply wronged by what had just happened to their characters got very angry. The problem wasn’t that the game had ended – it was the way it was done. The Etch-a-Sketch end of the world – the perceived unfairness of us having “missed” the signs, and the lack of satisfying resolution for the players and characters. We ended up persuading Jim to retcon the ending and having a massive battle with squid-beasts followed by a barbecue in my back garden, which worked out much happier for all involved – because the players got to have valiant last stands. They got to help, to save some things, to tie up loose endings. They got to make meaningful, characterful choices.

Jim could have told Bioware exactly how the ending of Mass Effect would play with the community of Shepards who experienced it, and exactly how to fix it. Gamers aren’t averse to endings – but if you build a game good enough to have players projecting themselves into your world, and then you destroy it in a narratively unsatisfying way that doesn’t respect their time, their energy and their choices – then the backlash is inevitable. In video games as well as LARP, players feel they’ve contributed to the game world – they feel entitled to a fulfilling story with fulfilling resolution. As creators it’s important to respect that.

It’s easy to forget, behind the buffers of code, PR, marketing, expos, cons and the anonymous internet, that game creators are still sitting down and telling stories round a campfire with their players.

What Fear Itself can teach us about consent in play:

Fear Itself is a tabletop game. It is, when it’s played well, Horrid. There’s a splat book with monsters and other gribbly things for you to include in your games. The book’s called The Book of Unremitting Horror, and it’s got stuff in like a golem made from the remains of people killed in snuff films, and The Motherlode, a horrendous walking vagina thing that births other monsters. And a wolf that comes out of a lake and rapes things. So it’s not a pleasant game. It’s definitely horror. And playing it with a real group is – tricky, because of consent.

In live games, consent can be really quite clear-cut. In LARP, you are your avatar – whatever’s happening to your character, physically, is also happening to you. So certain things are totally off-limits – you’ll essentially agree to fade to black for certain scenes, and agree the details out of character.

In tabletop, it’s trickier. Although your body isn’t the avatar, there’s still a very strong psychological projection of yourself into your character. When they’re directing action or recalling events, People don’t say “my rogue stabbed the dragon”. They use the word “I”, and they confuse other players with characters by using the word “you” interchangeably.

So as a game master, deciding to have a horrendous storyline that puts characters at risk of sexual abuse, for instance, is something you have to talk about beforehand. Gaming groups playing horror games will often have discussions about what they’re comfortable tackling at the table; at any point, someone can say they’re uncomfortable with what’s happening and the game can be steered away or the player can remove themselves without the experience ending. The key point there is “at any time” – you can stop the game, say you’re not comfortable, and work on a compromise that everyone can enjoy.

Just as in video games, tabletop role players have two main control modes for character. On one hand there’s the doll mode, where they’re simply directing action, positioning, controlling the avatar. On the other there’s a mode in which people are acting as their characters, emoting, conversing and projecting as them. Inhabiting them. It’s this ability to become a character that makes consent so vital for games – that makes the consideration of difficult material a much deeper one – because a violation or unwanted event happening to an avatar we inhabit is crucially, psychologically different to one happening to a character we watch or read about.

But big, story-led video games at the moment are not so good at acknowledging that they’re not films, and that they have a different psychological link with their players, where consent to experience is just as important as consent to view. Rating systems for viewing aren’t quite enough to let players make informed consent about what they’re doing, and we don’t yet have a good way to warn for or cope with binary stories that force you-as-player to experience certain things vicariously via you-as-character.

This is part of the problem with the way the Tomb Raider reboot has played out so far. Players who do feel they inhabit Lara may not want to consent to experience a sexualised assault; they may feel like they’re being pushed outside of her, like they can’t inhabit her any more. Like that’s no longer a safe power fantasy – no longer fun. That’s not to say that rape can’t happen in games – in fact, tackling it well will be a sign of serious maturity for video games, when it happens. But the issue of player consent and character consent are horrendously intertwined, and doing it well is horrendously hard.

What Zombie LARP can teach us about game experience and memory:

The main thing that makes Zombie LARP different from other LARP games is our focus on immersion and simple, procedural mechanics. But what makes us different from other live events is our story focus.

The win condition in Zombie isn’t surviving. It’s my job, as the closest thing we have to an AI designer, to make sure that the procedural rules we put in place generate a game where about 2% of the player characters manage to survive the game. So most people can’t “win”. But we still want people to feel like they won.

So we very deliberately decided that winning was about getting an awesome story to tell at the end of the game. Being able to go home and tell your mates you’d shot six of them before they finally swarmed you, or that you went down singing the national anthem and saluting, or that you slit your best mate’s throat and threw him to the super-zombies as a distraction.

What we found was that there are two players in every person. There’s the experiencing player and the remembering player. This taps into work done by the psychologist Daniel Kahneman about the way people work – he says we all have an experiencing self, that’s actually experiencing what’s happening, and a remembering self that constructs narratives about it afterwards.

Zombie’s taught us that great games satisfy both. They have to be an engaging, fantastic experience while they’re happening, and the experience has to be reconstructed in memory as an enjoyable, memorable one.

In board games like Risk or tabletop games like D&D, people elide the memory of the dice rolls and the complex maths from their experiences and talk like it all happened without the rules. This is another reason why dissonance between the play and the story matters – if a system gets in the way of the story too much, remembering players will have a much harder time constructing their internal narrative. It’s also got implications for playtesting and feedback – are you getting feedback on both the experience and the memory?

At Zombie, we run the most immersive game we can manage. We aim to scare the pants off people – or to get them stressed, at least, to get them running and excited and completely involved in play. We cut down abstractions. But we also have something called froth. Froth is a LARP term, but it’s applicable across most types of games. It’s what happens after the game, where people get together and talk through what happened. Sometimes it’s about showing people your Pokemon. Sometimes it’s about telling people at length about the politics of your Maelstrom character. And sometimes it’s about how you died.

So we started running a debrief – a frothing session at the end of the game, where we invite people to tell us their stories. This grew from people just talking as they came out of games – desperate to tell us how they died, who they saved, who went down where and how. People start constructing their narratives through conversations – they tell their stories collaboratively, excitedly, afterwards – and we’ve built a process for making that better into the game system.

In my experience, it doesn’t seem to matter whether the experiencing player has fun, so long as they’re not bored or disengaged enough – immersion is crucial – and so long as the remembering one likes the narrative you’ve constructed about it afterwards. So people who are genuinely upset and frightened during a game talk about it afterwards with genuine joy and pleasure. They get their kicks from the memory of the game – it’s an endurance sport.

The experiencing self gets far more pleasure out of busywork and casual gaming than the remembering self does. Pocket Planes and Farmville are brilliant experiences for a lot of people, soothing and calming, quietly rewarding; so’s grinding, if you’re the sort of person who can sink into a reverie and simply enjoy the experience. But those are generally pretty lousy memories – trying to explain to yourself why you just lost the best part of an hour to a game about planes can be really hard to do. By contrast, there are games that don’t feel as enjoyable while they’re happening but that make much better memories. Skyrim, for me, typifies that perfectly – I have more fun talking about the play afterwards than I normally do actually playing. It’s a balance of system and story, mechanic and narrative – and done right, it makes some of the best games in the world.

Develop title slideStanding on the shoulders of giant nerds:

Video games right now are still in their infancy as a medium. The technology’s changing so fast that it obscures the way that other things are barely changing at all, especially at the top end of the market. Tabletop gaming’s got some very similar problems – where the big titles from decades ago still dominate the market today, while the internet makes it easy for small creators and designers to write stuff and sling it out into the world to sink or swim on its own merits. The big creators are struggling to innovate.

Video games need to get better at stealing things from other genres. It’s easy, when you’re making a game that’s exciting or difficult technically, to be lazy about other stuff – world building, character creation, the smoothness of immersion, the psychological and narrative frameworks of the game. Film is still the default metaphor for big titles – the extreme end of which is Final Fantasy 13, which is essentially a DVD with really complicated menu options, and barely a game at all.

But a lot of the work that video games find so hard has already been done, by LARP companies like Profound Decisions or like the Nordic work on immersive game environments. Procedural generation, emergent play, character-led story worlds – even elements as simple as how to make relationships work in gameplay. Games are starting to draw from all sorts of other disciplines – economics (see Valve), literary theory, anthropology. It would be a shame if they weren’t also drawing on other game forms too.

GameCamp 5: belated thoughts

I meant to write up GameCamp 5 the week after we went, but what with work and writing and venue hunting for Zombie one thing drives out another, as Barliman Butterbur would tell you.

It was different this time. For one thing, last time we felt fairly out of place as live game designers rather than video game designers; it wasn’t unpleasant, but there was a definite sense that we were in the minority. This time round lots of people were talking live experiences, and mixed sessions didn’t carry the same assumptions about the group. That’s a little fascinating – does it reflect a wider uptake in live gaming generally? Certainly seems so to me – there are lots of folks starting to do pervasive games and interesting live experiences, a burgeoning scene that seems to be moving towards LARP from other disciplines and landing somewhere in the middle. Critical vocabulary is missing here; lots of people (like us) are drawing on all sorts of theory and work in other fields and applying it to making games in the real world. That’s exciting.

The other big change was in the way people at GameCamp talked about stories in games. Last time around, Grant and I ran a session about emergent story, discussing the concept of procedurally and structurally generated narratives that emerge through player interaction with the game, but aren’t “told” by the game. This time I think every story-related session I went to invoked the difference between extrinsic and intrinsic stories, usually with an understanding of emergence. Again vocabulary is missing here; I heard the same (or similar) dichotomy expressed as extrinsic/intrinsic, imparted/created, creator/player, and even cutscene/gameplay. Perhaps that’s a reflection of the greater proportion of live gamers there – the split is much clearer when you’re playing with stories in that space, and it becomes impossible to ignore that the story you’re telling is not the same as the one players are experiencing. But I’m hopeful that it reflects a shift in thinking by video game creators too. Not every writer needs to engage with debates like the location of meaning and the nature of narrative – but if none do, the medium will stay shallow. GameCamp made my literary brain happy and my game brain excited.

Oh, and we made some games, too. Hostage was the most fun. I’ll try to write up the rules soon, assuming one thing doesn’t drive out another, again.

LARP design and the problem with geography

We’ve been running Zombie for more than five years now. We don’t really know where it’s going next – our main venue is likely to be occupied every weekend for the rest of the year with Zed Events, which we’ve been helping to organise crew for. That means we’re back hunting for spaces again, which most likely means reinventing the game from the ground up again to work with the new geography.

Level design is very tricky when the physical arena of your game space is laid out in advance for you. Many of the most serious challenges in creating our type of game stem from the constraints of physical space. Navigation, staging, set dressing and crucial game balance issues all arise from location. Different venues take different concentrations of survivors and zombies; they necessitate different objective types; they change the balance between mass play and individual play; they change the ranges and dynamics of combat in many ways. That’s all before you even start on the aesthetics and the safety issues.

Because of all that, each space needs a different approach to the game rules that puts the emphasis in the right place for the venue. (This is also what makes the game difficult to franchise – it doesn’t translate easily across venues without some serious thinking about scenario design.) In reality we’ve built at least three quite different rule sets now, all under the Zombie LARP umbrella, each one tailored to a different sort of space and player base. Now we’re moving again I strongly suspect we’ll end up with a fourth.

A simple point and click interface: zombies at London IA

Demons at ZombieYesterday I gave a short talk at London IA, about one of my side projects: zombies, LARP, morris dancers, demons, creativity, delight, verbs, NERF guns and (ostensibly at least) user experience design. Slides – expertly drawn by @gshowitt – are here, and my notes are below the fold.

Continue reading A simple point and click interface: zombies at London IA

It’s Alive! Fostering emergent stories in Zombie

For those who don’t know, on Saturday I and a team of others ran the seventh Zombie LARP game. We’re hoping the next major event will be a big leap up in size, in ambition and in attendance. But before that happens I want to note down a few of our important principles – and important problems we need to solve.

What on earth is Zombie?

First, though, an explanation. Zombie is a live-action simulation game where people take it in turns to try to survive in an industrial complex overrun by the living dead. We run several scenarios over the course of the game, with a different group of people “surviving” in each one. When players aren’t trying to get out alive, they’re pretending to be zombies so that someone else can have a turn.

The game is a sort-of bastard child of traditional live-action roleplay (LARP) systems, fast-paced video games like Left for Dead, and the kinds of cowboys and Indians/summer water pistol games you played when you were a kid. The combat resolution system is based on Nerf guns (players shoot zombies) and a low-contact mechanic (zombies touch players on upper arms to represent biting, mauling etc.)

If you’ve read this and you still have no idea what we do, please leave a comment to tell me. I’m trying to improve my ability to explain the game to people who have never played a LARP or a video game before, so the experience would be useful.

How are we telling stories in Zombie?

Video games almost always have plot. Sometimes that plot is stretched over 50 or more insanely complex hours; sometimes it’s over in minutes so you can get on with killing things. Sometimes the storytelling is so deeply entrenched in the game that it’s inseperable from it; sometimes it’s abstracted from it so that the gameplay and the overarcing story are essentially separate entities. And sometimes the plot is about football.

Almost all LARPs are plot oriented. Some big games have top-down storytelling systems where world-changing events are affected by the big players in the game, while others have grass-roots player-oriented plot systems that allow even the most minor player characters to affect the universe.

In Zombie, plot takes a back seat to gameplay. Players might have twenty minutes at most to survive, and most of them won’t. That time seems a lot longer than it really is thanks to the game pacing and the adrenalin (much like the experience of riding a rollercoaster) but long-term character development is not an option, and neither is sticking around to watch the game world evolve. Zombie does have a wider plot system and the players can and do affect what happens, but when you’re running screaming down a corridor pursued by the undead trying to eat you, it’s impossible to take that in.

As refs and storytellers, we do several things to try and work with the game elements to make the game story rewarding. Most of these were worked out through trial and error and getting it badly wrong before we worked out how to get it right.

  1. Broad brushstrokes.We talk in bold black-and-white hyperbole. Every run is all-or-nothing, do-or-die. Players are given missions that affect the fate of the wider game world, so their actions carry weight and the game retains a sense of urgency.
  2. Metaplot and wider world. Zombie has an overarching plot framework that makes it possible to slot game events into place. There are several organisations in the game’s world – a shady scientific corporation, an armed resistance unit – and the real-time games take place within a framework created by the actions of those organisations.
  3. Sandboxing. Runs in Zombie are set up to be sandboxes where the players can take many different routes to the goal. We have set pieces for players to encounter – a room full of injured survivors, or a super-powerful zombie intent on taking them down – but those are never static events that play out in a pre-defined way. They are elements of the game world that add authenticity to the run without scripting players’ actions or requiring them to act in accordance with anything.
  4. Emergent stories.This is a common concept in video game design but in my experience is used much less outside specialist gaming environments. It refers to narratives that are uncovered or revealed during gameplay, and which require input from players to understand and piece together. For Zombie, I commonly use the term to describe stories about moments in the game that are unpredictable and unpredicted, that form unique and structured narratives, and that are the result of player interaction with their environment.

    And this is the important one. We try and make sure that after the chaos of the run, players have their own, personal stories to tell. We give them space beforehand to construct back story for themselves – encouraging team action – and we give them briefing time and attention afterwards to help them construct individual and group narratives about what happened. We try to give them tools and communities in which to tell those stories, we respond to them and retell them and incorporate them into the structure of the game.

Some stories filter out and fall. Others become local legends – the tale of the player who leapt six feet over a group of zombies only to later be mauled to death in a dead end, or the player who hid from the zombies successfully for twenty minutes before his mobile phone went off, alerting them to his presence (he died shouting “Now is not a good time!”). Last night one player managed to obliterate about 40 zombies with a heroic show of power – that story too will be permanently recorded in the mythology and mythos of the game. We give people awards for creating brilliant stories – often those awards are nothing but a shout out, a retelling of their story and a biscuit or a sticker, but they carry value and people strive to obtain them.

What’s so good about emergent stories?

Zombie is an activity that, at heart, is very difficult to share. It’s designed and conceived as a completely immersive experience while you’re playing, making it very hard to film video or take pictures. Backchannel chat, feedback and social sharing in real time are impossible. Very few images or films survive from our early events (though a couple of Youtube videos do get a steady stream of views and bring in occasional new interest three years later).

But even in the first game, our players found a way to share their experiences. They told stories to each other and to their friends, passing on their favourite experiences orally. Almost everything we’ve done with our storytelling framework since then has focussed on creating the brilliant moments that make those stories, and encouraging people to tell them.

In planning meetings we make lists of “moments of awesome” that will be memorable if they work right, things that will stick in the mind. We put single zombies in weird situations just in case a player stumbles across them. We make tableaux, design interesting characters for players to meet and memorable situations for them to meander into.

We try not to dictate the stories. More often than not they happen organically. We can’t make the player team split up and get lost; we can’t force someone to go to incredible lengths to avoid in-character death; we can’t ever guarantee that what we do will be the focus of player attention. More often than not our efforts simply go to create a better atmosphere for these experiences to occur. We make it easier, but it’s the players who make it work.

And we can’t dictate how the players ought to tell stories. We try to give them as many routes as possible online, both by creating our own community area and by using Facebook (and Twitter to a lesser extent) to curate and collect and encourage. Stories like this are ephemeral, and while we want people to tell them and we want a long-lasting record, we know we can’t rely on ever having one.

Many non-gaming events rely on video and images for a record. Increasingly, conventions and similar (relatively passive) events are relying on backchannel chat and the wider analysis of that conversation to provide useful data and a lasting record of what occurred. For us, the record lies in memory and in oral channels that are hard to replicate online – because of the immersive nature of the experience along with various technical issues, it’s impossible to get an idea of “what it’s like to be at Zombie” from any one medium. But when our players tell their emergent stories, that has immense value for us. It’s the best marketing possible because it comes with a direct endorsement and genuine enthusiasm. It’s an elusive currency but it’s vital to our survival and it’s been integral to our growth.

There are four main areas of uncertainty for me that arise from our approach, with questions that I don’t yet know how to answer. They are:

  1. How do we continue to foster personal, individual experiences and therefore stories while scaling our game upwards? If there are 180 players instead of 60, how does that affect our model?
  2. How can we encourage people to create and share content online that resonates with their emergent stories without sacrificing our immersive in-game experience? We already have teams going in with cameramen to film them, but the footage is necessarily low-quality and shaky and never reflects the full experience. How can we depict the game in ways that encourage emotional response and act as anchors for emergent stories in the same way that text can?
  3. How can this model apply to other events? How does it fit with (un)conferences and industry events? Networking events? Rallies? Fetes and carnivals? Riots and demonstrations? Is this another way of looking at and describing oral history? Or does this work to foster, encourage, document and curate emergent stories have journalistic potential?

If you have any suggestions for answers, or any more questions, please share them in the comments.

Emergent thoughts on emergent stories

Alexander de grote, in de Hermitage tijdens Museumnacht in Amsterdam After an interesting conversation with @harryharrold and @MrRickWaghorn yesterday, I’ve been mulling a few thoughts on emergent stories and how the social side of the web could make it possible to curate and (to some extent) formalise them.

As @harryharrold pointed out, one of the big problems LARP events haven’t solved yet is how to deal with emergent stories after the fact. Unless you’re physically present at an event it’s next-to-impossible to get a coherent narrative of what actually happened, particularly if it’s a big fest event.

In my experience narratives split into two general categories after LARP events – the big picture and the little details. For some players/characters and events, the big picture is what’s important; they have enough of an impact on the overarching plot  plot and the overarching plot is reachable enough for them to be satisfied with a global grand plot update.

But for most events and most players, what’s important and relevant is the nitty-gritty of their immediate social circle. Who said and did what to whom; the “little” stories of betrayal and intrigue and death and love. And that’s incredibly hard to track.

A big part of what we do at Zombie is making sure that individual players get their stories straight before they forget anything. We have “debrief” sessions so that both the refs and the players can get a handle on the emergent stories that have happened on each run, and we do our best to encourage storytelling online after the game. But that only works because every run is self-contained; with a larger plot edifice and continuing characters such a simple system is simply impossible.

But LARP isn’t the only area to have this problem. Festivals, conferences and conventions do too; in fact any large event where multiple trajectories are possible and no individual experience is large enough to express the story of what happened.

Here’s an idea I was contemplating pitching for Greenbelt this year or next. Hand out 10 relatively smart phones to 10 people, picked as a good spread of the demographics at the event. Kit them out with Twitter feeds if they don’t already have them, photoblogging, moblogging and microblogging kit, basic video, audio and still image hardware, and make it as easy as possible for them to upload anything and everything, wherever they are. Track everything they do, geotagged and timestamped, and from that collate and curate a livecast of the event.

I’d love to see if it’s possible to combine @harryharrold’s ideas on curating LARP stories and map them onto other events too. Emergent stories are popping up all over the place and the social web is making it possible to collaborate and build interactive, explorable maps of events that previously have had linear narratives. The stable social experience is exploding and we all want to choose our own adventure. Perhaps we can apply this to more than just LARP.