Space Station 13 - A case study in emergent gameplay and system-heavy design

As cliche as it is to start with a definition, I doubt many people have heard of Space Station 13. If I was to summarize the game with an IGN-style "Skyrim with guns" quip, it would perhaps be something along the lines of "Like Dwarf Fortress, but multiplayer", but that frankly only barely scratches the surface. 

In Space Station 13, players are given a role such as Engineer, Security Officer or Chemist and tasked with keeping the space station running. Some players are chosen (depending on the game mode) to be traitors, aliens, changelings, or even a wizard and must complete hidden objectives like assassinating another crew member or stealing an object.

The job roles are complex, the station itself is a maze of hidden passageways and secret items, and in this game knowledge is power. If you know the perfect wall to weld through to get into the armory, or the perfect escape route to run away from security, you become a powerful player. Knowledge of every job role is the lofty end goal of the learning curve, since it allows you to be an effective traitor when it's your turn to kill your compatriots; if you know how the atmospherics system works, how to effectively sneak around (or talk your way out of situations), how the virology system works, how the telecomms system works, and how to use remote triggers and wiring properly, you could orchestrate a communications blackout and pass a deadly virus through the ventilation system that turns everyone without a mask into monkeys, eliminating all of your enemies before they can so much as sputter "CHEF IS ROGUE". The intersection of so many complex job roles is what makes the game so intriguing. I'll give a rundown of a typical round to better elucidate this:

A typical round

The Engineers set up the singularity (an unstable power source) without incident, upon going to set up the backup power source the solar panels however, they discovered a corpse, just 5 minutes into the round. Knowing this to be the work of a traitor, since the corpse was fresh, they took it back to medbay and continued on their way.

Once there, the corpse was stolen by the robotocist who was eager to harvest the brain to put into the backup AI he was building in secret. After surgery, the brain-less (and arm-less, the robotocist clicked the wrong limb) body was then flushed down a disposal chute where it landed in disposals. Meanwhile, the mime had broken into security to write rude messages on the floor in secret as part of an art performance (I confess, this was me), and was promptly captured and put into the brig, which it turned out had an air leak.

The security staff watched the mime pretending to suffocate to death, and then realised that he was actually dead, at which point the two security officers decided to hide the body inside a locker and pretend they hadn't seen anything. Then the bridge was flooded with plasma from the air circulation system, and it transpired that someone had messed with the atmospherics system to flood the station with plasma. The all-seeing AI reported that it was in fact a traitorous clown who had done this evil deed! So the security team went on a hunt for the clown, who was last seen skulking around the maintenance shafts. Meanwhile, engineering staff desperately tried to weld shut all the vents and shut off the plasma leak, but it was too late, someone had ignited the toxins and the fireball of plasma was slowly tearing medbay asunder.

Finally, the telecomms system was modified so that all radio names names were filtered to various anime characters, and the AI (now named Satsuki-sama) shocked all the doors, revealing itself to have been the traitor all along. In various pockets of safe air around the station, the few survivors clung to life, with the biggest concentration being inside the cargo bay, where they ordered weapons against the AI's knowledge and led a ragtag assault on the AI core to destroy the traitor and escape the station. The raid was a failure, as the AI's security borgs were all able to shut down the revolution..... all except one, that one body found at the start of the round had now had his brain put inside a security borg (rather than an AI as originally planned), and the wily robotocist had made sure its laws weren't synced to the AI (who, it turned out, had the standard asimov laws, then a hidden law 0 that read "%^£"$ ONLY DOGS ARE HUMANS, ALL HUMANS ARE HARMFUL TO DOGS").

Through a combination of subterfuge and acting, the robot pretended there were intruders in the AI core, destroyed the AI before the other robots could respond, and then died trying to fight off the other robots. The last three crew members alive (out of about 100) were the chaplain, a lawyer, and a virologist, all of whom had been hiding in the janitor's closet and telling jokes. They finally emerged into the bloodied station, stole a welding torch from a dead engineer, broke into the bridge, and called the emergency shuttle to come and pick them up. After 10 minutes of waiting for it to arrive, they boarded the shuttle and left.

Oh and that corpse at the start of the round? He'd just followed the AI's directions and walked out an airlock that was promptly locked behind him.

Now why is such a strange and random tale interesting? Because the developers designed almost NONE of that. This wasn't some overarching gamemode, it wasn't capture the flag:

The systems were in place, but it was all player driven, a vast and complex interlinking of dozens of different systems interacting with each other. Maybe in another round engineering don't set up the singularity right and it breaks free, eating the station and spurring a great rebuild of the central command infrastructure. Maybe the librarian sets up a pirate broadcast system and starts broadcasting smut from an unspecified location as security desperately tries to track down the signal (this happens surprisingly often), or maybe security decide to set up a kangaroo courtroom and put the clown on trial for being unfunny. It's all player driven, and that makes it unique.

Every single actor in this vast chain of events was a human (aside from a few NPCs in the game such as space carp, blobs and monkeys), and many other smaller stories likely happened that I wasn't aware of, after all I spent most of the round dead as a ghost and observing the craziness as it unfolded. The game can be a tense survival horror affair, a co-operative base building experience, an exciting whodunnit, a secret war between two factions spoken about in hushed whispers and corpses. Sometimes the atmospherics department, who controls oxygen for the entire crew, will hold the station hostage and set up the proud independent nation of Atmosia. Or Cargo Bay will set up the noble Independent Republic of Cargonia. These are all emergent facets of the game, and though the server owner can try and set up gamemodes to encourage certain forms of play, the most interesting rounds are always the unexpected ones that nobody could have predicted.

So who made this magnificent game? Well Space Station 13 began as a simple air physics simulation engine built in a little-known multiplayer game engine called "Byond", which is designed to let players make their own games with all of the netcode done for them. The game has no single "developer", but has instead had hundreds of contributors putting time and effort into an open source game which has evolved dynamically as different groups decided to create new code bases, branches, variations, even entire scripting engines.

The game as it stands today is the result of about 10 years of semi-anonymous development, and that makes it unique in the video games industry. The original game was designed by a user named "Exadv1" as a simple air simulation. Unified code bases have come and gone, common resources, code branches have been adopted and dropped, and almost all of the development was anonymous or semi-anonymous.

The history of SS13 is fascinating, but at its core, Space Station 13 is a game authored by no-one, designed by everyone and created as it's played. Each server tends to have slight variations or differences in conduct; for example some servers value serious roleplay, others have a more generalist and laid-back feel to them, but there's also differences in game systems, content and such. Some stations have different rules for surgery, chemistry, making food, different AI laws, different station layouts, different damage thresholds. And this changes up the gameplay. In such a complex series of interlinking systems, changing one piece changes the whole game.

One case study example of this: There are disposal bins all around the station. Placing an object inside one and pressing the button (or waiting a few moments) sees it chuted away pneumatically at high speed to the disposals area. You can also fit people in the chute, with a bit of effort. On one server my tried and true tactic of escape, jumping into a disposal chute led me to be bludgeoned to death by its high speed pipe system, whereas on another server it simply transported me safely to disposals. So now the disposal chute is suddenly a lethal weapon that also removes the body, making it the perfect tool for murder. Of course, you could always remove the flooring and place down some more disposal chute pipes to make the chute whizz you out into space too, thus ensuring the body is never discovered. Why stop there? With a few minutes of re-wiring, the chute can now launch unsuspecting victims into the Chef's kitchen, ensuring that he has a fresh supply of cadavers with which to make a variety of "questionable" burgers, and also dispose of the body. Suddenly, changing one slight aspect of the design drastically alters the tactics and thought put into the whole game.

But Space Station 13 has its flaws. The game is slow and laggy, the combat is a flustered and barely functional spamming of clicks that the community jokingly refers to as a "robust combat system". The inventory system is unintuitive, and despite great efforts to improve the spritework in the game, the game is generally pretty ugly to look at. The game's been developed by amateurs and fanatics; the level of dedication to a completely profitless unknown game is astounding; the game even has its own coding language, NTScript, to handle telecommunications. You can walk into Tcomms and start coding an adventure game into the radio system if you really want to. The game's interface is... clunky, to put it politely. Barebones and utilitarian to put it a bit heavier, and "Shit" to call it what it is outright, but much of this is due to the limitations of the engine.

So my question is why hasn't this kind of emergent-style multiplayer game been popular in a more mainstream way? 

Well I think whilst Space Station 13 manages a lot, its appeal is relegated to those willing to put in time to learn its systems. Being a new player is suffering, you don't know anything, you have no idea what's going on, and things seem to just happen without any input from you. One second a new player will be trying to figure out how the wiring works on a vending machine and the next a security officer is beating them to death with a crowbar because it turns out they were secretly a changeling. The learning curve is somewhere between Red Orchestra and Dwarf Fortress, but it's a steep investment for new players. A game so system heavy will inherently favor those with the most dedication, and that doesn't appeal so much to the mainstream audience.

Similar games that I can think of off the top of my head are far simpler in scope. Trouble in Terrorist Town, a mod for Garry's Mod, deals with short 10 minute rounds of murder mystery survival, rather than the monster 1 to 2 hour rounds of SS13. Spy Party deals with just one room of potential traitors with a few players, not an entire space station of 20 to 100. Minecraft focuses on building, but has much more of a focus on algorithmic creation and the crafting of tools for the purposes of more construction, rather than focusing too heavily on the crafting, whereas Space Station 13... well one look at the guide for genetic modification should give you an indication of the level of finesse required for some of the crafting jobs in SS13.

So to be successful, a mainstream title would have to simplify these complex systems, and in doing so remove a lot of the fun of having so many complex interactions. If everyone could become a top tier geneticist within an hour or two of playing, then a lot of the fun would be killed. I, for example, pride myself on my ability to break and enter into places. As a mime, I am given something of an excuse to do so (at least on the server I play on, which is a little lax on rules), as long as I am not causing any active harm, and I have gotten used to wearing disguises, blending in, and so on. One of my proudest moments was assuming the identity of an atmospherics technician whom I found dead in a maintenance shaft and, despite not having access to any of the doors or equipment unless I hacked them, I managed to successfully assume his identity right up until the end of the round where I was unceremoniously murdered by a traitor assistant who had killed the atmospherics technician in the first place and assumed that he had failed in his task since I was walking around alive and well, appearing to be the man he'd murdered. To carry out this complex scheme, I had to have knowledge of how to hack doors, how to operate the complex atmospherics system, a few locations that were good for building false walls and, in the moments leading up to my own murder, how to succesfully disarm an opponent (something which I evidently failed at). These kinds of stories are possible with streamlined mechanics, but by being deliberately dense and obscure, Space Station 13 succeeds in making knowledge a skill in a way that other games simply can't do.

Of course the elephant in the room from all this talk of a mainstream Space Station 13 is Dean Hall's "ION", a spiritual successor to Space Station 13 that was announced at E3. I suppose I am sceptical of the project; to succeed it would need to strike an incredibly fine balance between the freedom and complexity of SS13's systems, and the playability and streamline that a console audience would want, especially given that a controller would presumably be the main input method. SS13 is a bizarre game, not just because of its birth, but because of its tone and atmosphere too. It varies by server, but by default SS13 tends to be a slightly abstract, bizarre affair, where a clown is considered a worthy job on a Space Station, chaplains can heal people by bashing them round the face with a holy book, and at any moment a wizard could teleport into the room and turn everyone into monkeys.

Too much complexity, and players will simply be turned off. Too serious a game, and that charm and humor is lost. Too simplified and some of what makes SS13 so challenging and endearing is gone. Even now, after having played the game for a good 300 or 400 hours, I still have not figured out every single facet of the game (I have no idea how research and development works, or how virology or mining work), and especially not on every server. Which brings me onto the final challenge ION will face: customization. Space Station 13 was borne out of collaborative development between many different people, and as mentioned before each server has its own spin on the concept. I can't stand heavy RP servers, because I am more excited by the interaction of gameplay systems, but to some players roleplay is their raison d'être, others enjoy admin events where admins decide to create special circumstances D&D style. The customization, complexity, and tone of SS13 is what makes it so unique. To simplify that effectively whilst keeping the core spirit alive is a task that is nigh-on insurmountable, and not everyone will be pleased with the result, but I'd be lying if I said I wasn't cautiously optimistic at the idea of a version of Space Station 13 that isn't frought with "space lag" (The tongue-in-cheek in-character term for lag.)

So there, hopefully now you will be robust enough to come and enjoy the wonders of Space Station 13, just a little bit. I eagerly await the slew of journalists reviewing ION who have never even heard of Space Station 13, so I implore you to at least check SS13 out, and maybe you'll even enjoy yourself.