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.


I’m Mary, and I’m here to talk to you about toy guns and zombies, with the assistance of some state-of-the-art graphics.

With my partner Grant I run a game called Zombie LARP. LARP stands for live action role play – essentially, games that require the suspension of disbelief in real life, where players dress up or pretend to be the main characters in the game, generally with a conflict resolution mechanic relying on a mix of skill and chance. And often large pretend weapons. It’s sort of a cross between tabletop Dungeons and Dragons and amateur dramatics.

What we do, unsurprisingly, revolves around zombies. We run a game in which a bunch of people pretend to be zombies and attempt to “eat” survivors – or tag, because we can’t actually physically represent eating people. And the survivors get toy guns like these, which shoot foam darts at the zombies.

Each game is different, with different numbers of survivors, often in very different venues. We have to tailor the game to different groups, players, experience levels and so on. The game runs on a rotating basis – people take it in turns to be survivors and zombies – which lets us change things on the fly in between sessions and iterate throug. And we can actually change things during the course of the game itself, if it’s too easy or too hard, by moving zombies around or dropping extra weapons. We call that responsive design.

There are a few constants: we work very hard to make sure the player experience is immersive, exciting and fun; we give the players an objective to complete before they can get out; and across most of our games we have a survival rate of about 5%. I take great pride in that.

What we don’t do is create experiences – the players do that themselves. Our job is to create an environment and a system in which experiences occur. Within the game we work to make a system where exciting, dramatic experiences happen organically – and we do a lot of framing work to help forge a unified narrative out of the players’ individual, fragmented experiences of play. There’s various names for this – some folks think of it as procedural storytelling, or emergent storytelling. We use the concept of creating and operating a story machine – a system within which exciting stories are born through the interaction between the player and the rules, procedures and environment we create. We borrow a lot of UX design to help us do this – looking at flow and navigation, thinking about personas, tackling context, considering emotion. Good LARP creators are pretty familiar with all these things – they just might phrase them differently.

So, LARPs are about turning the real world into a game environment. Some require a lot of pretending about your surroundings, while others aim to be as immersive as possible. For instance, we used to run in a university building – we turned most of the lights off, used overhead projectors to cast weird shadows, and (around the time we started getting serious about the game) dressed up the rooms. Now we run in a huge abandoned shopping mall, which to be honest does a lot of the work for us – and players themselves add a lot to the atmosphere of the game with costumes and makeup.

We keep the core mechanics very simple. Many live games that include combat use things like damage calls, which you have to remember along with your character’s capabilities; or magic spells, that you have to remember and imagine. Some even have completely abstracted turn-based combat systems. Our character options and rules are very limited, work with people’s instincts, and understand that if you’re scared or running around, the last thing you need are lots of rules to remember.

But the main mechanic of our game is combat – you have to fight the zombies in order to survive – so we do abstract that using NERF guns & foam weapons.

These form the basic, point-and-click interface for the game. The starter weapon – the Maverick –  has six shots. A fair few other weapons also have six shots, which is the main reason why one of our zombie behaviour rules is that they swarm in groups of seven – there should always be one more than you can comfortably shoot.

There are a lot of different types of gun, now. When we started there were about four, and now there’s a dizzying array varying from the utilitarian to the downright ridiculous. For instance, there’s the Rapid Fire – not sure why it’s called that, because it’s extremely slow to fire and very unreliable. Its main feature is that after you shoot it, it ejects shells onto the floor. That makes it Badass.

And then there’s guns like the Stampede. Fully automatic massive battery-powered behemoths that give you lots of ammunition but balance it out by making it almost impossible not to just charge screaming into the first combat wasting all your ammo.

For the players, this is the primary way they can affect the game – it lets them control their entire experience. So we emphasise it, talk to players about how it works, demonstrate it – all of which helps to frame the experience in a way that’s going to make sense for them. We treat the guns – which are, in fact, exceptionally silly pieces of kit – with a lot of seriousness. All that context helps players to take them more seriously than they might otherwise. It helps not only to make the game itself more immersive, because people are treating these toys as weapons as soon as they get into the building, but it also cuts the chances of people having NERF wars in the middle of our player rooms.

Every type of gun has its own behaviour, its own usability issues, its own play style and its own impact on the game. Much like a first-person-shooter computer game, we can balance the game using the numbers of zombies and the ammunition available, as well as the capabilities of the weapons. For us, that includes the likelihood of guns jamming, the plausibility of the players losing it and wasting their ammo within seconds of the game beginning, and conversely the propensity for people to hoard their bullets till the last possible moment. You’d be amazed how many people save one bullet, just in case.

But in some ways the game’s framework has less in common with modern FPS run-down-a-corridor-shooting games and more parallels with text adventure games: here is a verb we give you with which to interact with your environment to see what happens. This is all implicit, but the effect of rule systems like ours is to constrain potential actions to a limited set. The main verb we give people is SHOOT.

Other optional verbs that move the experience forward are things like HIT, RUN, HIDE, SCREAM, SWEAR, RUN FASTER, DIE. Even with this limited set of anticipated actions, players sometimes surprise us with things like PRETEND TO BE ZOMBIE.

Zombies, get verbs too, though much more limited set: SHAMBLE, HUNT, ATTACK, EAT and a few in response to player actions like FALL BACK, COLLAPSE, DIE and then later on RESURRECT.

For a while, we experimented with a magic system, but it didn’t work very well – mostly because it required people to remember things in a systematic way, when they were under quite a lot of physiological, zombie-related stress. But we did manage to bring in a power that sort of works: in some games, religious characters called Believers can scream freestyle religion in order to stun zombies. During one game, a player decided he wanted to play an atheist Believer, and managed to stun a crowd of zombies by telling them loudly and repeatedly that they were scientific impossibilities.

This is the sort of thing that happens a lot. Players want to use the verbs we give them in unusual ways. They want to try them on everything within the simulation. You see this tendency for users to muck around just about everywhere, but most clearly in video games, because they have the freedom to respond. Text adventures tend to have stock phrase responses for when you couldn’t do something, and as we’ve moved through point-and-click adventures those have gotten more sophisticated. Action adventure games have slightly better ways to deal with players testing the boundaries of the sim, though they’re not always perfect – you could drown Lara Croft; Uncharted 3 has safe areas where commands that would normally make Nathan Drake punch people instead make him wave or shake their hands; but you still can’t set fire to children in Skyrim.

But in a real-life, responsive environment you can’t provide stock responses to boundary testing. You have to make sure there’s a safe way to test boundaries without actually risking hurting people, so there has to be clear communication about where the hard, out-of-game boundaries are. And within the game, players, given a SHOOT option, will attempt to SHOOT everything. Including, and especially, things you don’t want them to shoot, like information-givers and quest-givers and each other.

For a while, we dealt with that by using unkillable god-like quest-givers, but the players didn’t enjoy that so much. People really, really want their boundary-testing to be rewarded by something delightful. So we started introducing things like hideous carnival-style puzzle games in which you could shoot things, but there would be obvious and unpleasant consequences. We started a reward system that built achievements on some of the most startling, brilliant things our players had done – so we have the “Over a man’s head mind you” award for gymnastics under stress, or the Cactus “Bastard” McPhillips award for astonishing bastardry, which gets awarded for things like sacrificing your best friend so you can run away. But only if it doesn’t contravene rule zero. I’ll come back to rule zero in a moment.

There’s delight in the unexpected interaction. In one event, there was a substitution puzzle in which players had to solve a cipher in order to determine which of several symbols they had to write on the walls. We had symbols for angel, demon, heaven, hell, kill, reanimate, and various other things – including symbols for our big unkillable god-like characters Emmerson and Kramer. After the game, players asked what would have happened if they’d written the symbols for “kill Kramer” in the ritual circle. That would have been a fantastic moment to go off-script, do something spontaneous and delightful.

So since then we try to look out for unexpected ways the players might use the tools we give them, and reward them by making it work. In the last game, in order to get out of the complex, the players had to find a demon and lead him through the zombie-infested mall to a magic circle. They could control him if they had a particular occult book. So they got the book, and rather than take him straight to the circle they moved him around the mall getting him to stand in doorways killing zombies for them.

Then there are always people who will push the boundaries of permitted interaction in other ways – changing the objectives, or the parameters of play, or the verbs involved. Part of playful exploration for many people is the creation of arbitrary goals and limitations, in an attempt to see what happens. You see this sort of creative reimagining all over the place, but again perhaps most clearly in games. Speed runs of Super Mario, or the Chronicles of Nondric, where a gamer played Oblivion as a commoner and tried to avoid doing any quests at all. Minecraft is built on this urge to create something amazing, unique and fascinating within a limited system. Zombie too has people who try to get different experiences out of the game by limiting or altering the verbs they use to interact with the environment. My personal favourite example of this: a group of four people turned up to the last game kitted out as morris dancers – complete with jingling, zombie-attracting bells on their legs – and went into the sim only with melee weapons, no guns at all.

They broke the game.

But that’s fine, because next time we’ll balance it more effectively. We’re looking at responding by taking away most of the melee weapons so people who do want to limit themselves in that way get to be a bit more special, and find the game a bit more challenging. And they’re talking about coming as some sort of Highland marching band. The wonderful thing is that everyone has fun.

That’s where we come back to Rule Zero. The most important rule of the game is very simple: don’t be a dick. By framing the whole event in those very simple but quite far-reaching terms we create an environment where everyone gives each other the benefit of the doubt. Rather than making rules that force players into certain types of unfun experiences when they’re applied rigidly, it lets us say: be flexible. If someone tries something unexpected, go with it. React in a way that makes sense in the situation you’re in. If it breaks the rules of the simulation, if it uses actions or verbs you’re not familiar with or you haven’t built in, that doesn’t matter, just so long as it’s fun. And if it’s fun enough, we’ll try and build it in for the next event.

Fundamentally: shooting zombies is fun, and pretending you’re a badass while doing it is even more fun. Everything else is about reinforcing that core mechanic, and trying to make it the most fun possible.

Published by

Mary Hamilton

I'm a journalist-type tech-ish geek person, working in that interesting ambiguous place where reporting the news meets all sorts of peripheral skills. In my spare time I herd zombies, design games and write stuff.

What do you think?