On Games and the Communities they create

[Junkenstein]

Well this is probably going to get very, very messy. I'll try to keep this coherent but shout if there's anything that needs more detail.

In this post(s?), I am going to try and provide a short version of the nature of Game Theory and Mechanism Design and how these relate to online and offline communities.

What is Game theory? Wikipedia tells us that :

Game theory is a study of strategic decision making. More formally, it is "the study of mathematical models of conflict and cooperation between intelligent rational decision-makers."[1] An alternative term suggested "as a more descriptive name for the discipline" is interactive decision theory.[2] Game theory is mainly used in economics, political science, and psychology, as well as logic and biology.

Let's simplify that a touch. Any situation where your decisions are based upon the decisions of another person can be classed as a game. Anything Single player does not count. We are looking at why you make certain choices based on the actions of others. There is an built in assumption that everyone will act both Rationally and with their own self interest in mind.

Lets define some terms:

Game - For our purposes here a "game" will require
1-More than 1 person
2 - Interactions are possible beween players
3 - There must be some kind of decision
4 - There must be some kind of payoff/reward

Utility - This is what you are playing the game for. In a FPS the utility is to get the most kills. In a race the utility would be to come first. Game theory assumes that all people involved will have the same utility in mind. When players choose to follow a separate utility, griefing or trolling is soon to follow.

Payoff - Highly changable depending on the game. This could be a high score, Gold Medal, social approval. Think of this as what you "win"

Decision - The game must allow for players to make choices. Snakes and Ladders technically isn't a game under this definition. You just roll dice. The same could be said for athletic sports like running, you only have the choice of one strategy, Run Fast. Any other strategy is irrational.

Strategy - Think of every move you can make in a game of chess from start to end. That is one strategy.

Tactics - Think of an individual move in a game of chess. That is one tactic (This is sloppy but will do for the moment).

Perfect Information - This is where you have knowledge of everything that has taken place in the game. There are no hidden variables and win states are clear from both sides. Consider a RTS game with and without the "fog of War". Without the fog of war, you have a perfect information game. With the fog of war, you do not.

Pareto Efficiency/Frontier - I've seen this called both. Think of a fighting game with 3 guys. You have the stong guy, the fast guy and the medium guy. The idea here essentially is "Give something to get something". If Strong guy is stronger and just as fast as Fast guy, it would be irrational to ever pick Fast or medium guy. Many games have balancing problems and it can often be down to this. You'll never pick option X as option Y is always there and is simply better. Think of an FPS where a player is unstoppable with a certain weapon and others are never seen. That's bad Pareto Efficiency.

Solvable/Random/Arbitrary - Games can generally be broken into these 3 catergories. Solvable games are things like Draughts and Tic-tac-toe. There is a clear set path to win. Random games are much more difficult to create effective strategies for. Arbitrary games can only be won by either not playing or acting as randomly as possible (Think rock-paper-scissors)

What is Mechanism Design?

Mechanism design is the set of ideas that the way the game is played can influence those playing it. By choosing (Or omitting) certain game mechanics, game developers directly indirectly influence the end game in various ways. Take the most obvious for a moment, Monetization. I'll get to that. A key point here is that most developers will use various mechanics, little consideration is given to the impact on the rest of the game and player base.

More to come.

[Junkenstein]

So we have established what game theory and mechanism design is. Let's look at some popular games and the mechanics they employ.

Beat-em-up or Fighting games have long been established as competitive tests of skill with clearly defined win states. They tend to be Perfect information and the potential strategies are literally infinite, especially newer fighting games.

Term I should have defined earlier-

Barrier to Entry - This is what must be achieved just to play the game. Money, Time, Location and Skill (Thinking more about ranked matches/ladders) could all be barriers to entry. Some have a very low barrier (Tic-tac toe requires 2 players and a way to draw the grid) and some are very high (AAAA PC title requiring next year's graphics cards etc). The higher you make the barrier, the less people will be able to play. This also has a dual result of building a sense of "eliteness" or "openness" into the respective community. Will expand this.

The barrier to entry to play Beat-em-up game is firstly obtaining the game. For competitive play a second controller or net connection is now usually required. Now why do these games tend to have a bad rep for misogyny/racism etc? Well lets think about the Skills being tested.

Skills - The skill tested depends on the game. Some test reflexes, planning, or logic. Many games test the skill of patience. Monopoly would be a good exactly example of that. Games in the sense I am using tend to be competitive tests of one or more skills.

Beat-em-up game would typically test the skills of reflexes and memorisation of set moves to ensure the swiftest victory possible. As these games grow and get more complex, the reflex element becomes less key and memorisation of effective moves and counter moves becomes part of defining the player base by skill set. Think about playing beat-em-up game at home with friends compared to online. You may be GOD at home but you'll probably be beaten like a child in online competitive play. Why?

Well this is where we get back to Game Mechanics.

In the evolution of games, developers typically looked at what aspects of games that came before were fun and then look to recreate that fun in a new setting. This can (and does) result in things that were originally glitches or coding-fuckups being seen as an essential part of the game. Think about rocket jumping in Quake. It makes absolutely no sense to fire a rocket at the ground in front of you to try and jump away. But it was kinda fun. As a result FPS's coded in something akin to this in several iterations since Quake.

Now what happens when the "old guard" of the game and community know about little quirks like that? Well it can go a few different ways.

If the game mechanics have been designed to be welcoming to new players the community will tend to be more supportive of new players. In Team FPS game, some games went down the route of raising the health pool of new players. (Edit - This is exceptionally difficult to do and maintain Pareto efficiency. Also creates griefing issues) Others such as Natural Selection made the co-ordination aspect of the game a key mechanic. NS is essentially a hybrid of FPS and RTS. Both side have a "commander" that can issue weapons to troops and build structures. The understanding is the FPS players follow the commanders leadership.

This made it a very difficult game to troll. If the commander gives you a jetpack and you just fly off the map, you're not going to get another jet-pack. If you got given a shotgun and did a good job you'll probably get a better gun. Communication was essentially a key mechanic of the game.

Now I mentioned Win states earlier, that leads us into scoring.

Stateful/ Stateless - This is about whether the results of the game are recorded and matter. An instance of FPS game used to be stateless in most cases. You played a round or 10, logged off and no real record was kept of your score, name weapon of choice, accuracy % etc, etc, etc. Stateful means your details are recorded and what you have done matters in some way to the overall game world.

So lets look at Modern FPS. In one of the links Cain posted (Link 4?) there was discussion occurring about whether this player was able to achieve these things legitimately. By being able to consider these things, the human brain is very good at being able to place personal failings on others betrayal. "SKKREWUAIMBOTHAXX" just used to be typed into whatever crude IRC kind of chat was built into the game. Now you have microphone headsets so you can shout at the perceived wrong-doer. The barrier to entry for abuse has been drastically lowered. The game has moved to analysis of other players instead of playing the game.

So what happens when you totally remove the ability for players to communicate? Well Nintendo tried this.

Nintendo online games tended to be Stateless competitive tests of skill. Think Mario Kart. Nintendo also recorded win/loss figures. Now, have you ever been playing something and had a perfect score? What's the first impulse when something upsets that? Pull the plug. Keep the 100% victory rate.

Nintendo are pretty much terrified of Child predators. As a result there is a blanket ban on the ability for players to communicate in-game. By removing this ability to create a social connection games tended to never actually finish. Both/all players play until one establishes a clear advantage resulting in everyone else rage-quitting to maintain a "perfect" score.

I mentioned Ranked/ladder games before and that ties in here too. Some games run online dictate that a certain score is required to be able to play in this game. The result here is no new players want to play (As they can't lacking the requisite score) and everyone rage-quits sooner or later to preserve their score otherwise no-one will play with them. Such games have communities that make Mos Eisley Cantinas look like a charming urban nightspot.   

Yet more to come

[Junkenstein]

As you can see, the Nintendo route is probably not the way to go. You should want players to be able to communicate in a positive manner.

Lets look at a Stateful progressive game that allows players to communicate and express themselves - Second Life. In Second Life there is much that you could compare to an actual world community. Various players exercise creative goals for primarily personal reasons, much like Minecraft.

Second life did not anticipate that people freed from standard societal constraints will probably grief like there is no tomorrow. I say Grief as that is how the behaviour tends to be viewed.

This however:
http://www.youtube.com/watch?v=RedLyae4b2s (NSFW, probably)

still makes me laugh like a loon. This is not griefing to me, this is exactly what something like Second Life was made for. Rabid arguments still go on about this, yet very few raise the idea about what utility this guy was going for. I would guess that his reasons for playing were personal amusement and the amusement of others. Contrast that the the interviewee whose utility was increasing their E-fame and wealth. Incidentally Anshe was the first person to make a A million dollars real currency from virtual goods and services. Who is right? Who is playing the game correctly? Both? Neither?


The answer is both.

Moving towards more online communites and less away from games, I will use features demonstrated in THIS VERY FORUM to demonstrate why some online communities devolve into shit time and time again.

[Junkenstein]

I'm quite sure that you will all be familiar with other Vbulliten / PHPHB (whatever it is, I forget) kind of forums. Lets think about some of the things found in these places that can also be seen here.

Every post is accompanied by your User Name, Personal Text, Avatar, Signature, Post count and ways to contact you.

Now in this forum, the above tend to change fairly frequently. In general there is no particular feeling towards them either way as far as I know.

Other forums tend to be a lot more invested in their online ID. Take post count as an example. Have you ever seen a thread where it's "This is MY XXX thousand post!!!!?!!1!" Was there anything of value or interest within? Is there any value to posting purely to raise your post-count? Of course not. PD recognises this at least sub-consciously with post counts raising random amounts per post. What is the direct result of that mechanic? Well, have you seen a "This is My XXX post!!1" thread lately?

Consider too, the attitude to Moderation. Many forums thrive on Drama and bans. Here, people tend to Ban themselves. Most fail at that too.

I've rambled a lot here so I'll try and condense this a bit -

Examine the game and the numbers of people playing with an alternate utility to you or the core game. If more people are playing Kill your own team game than kill the other team game, the community is probably going to be terrible.

Recognise that people attach themselves more to Stateful progressive games. If you are dealing with a persistent game world you are likely to encounter less trolls/grief in day-to-day play.

Appreciate that this is a VERY rough overview typed as it came out my head. I've not touched on a ton of stuff such as Real ID (Tying your actual REAL person to a game account) Monetization models and their impacts or a bunch of other stuff. I'll leave this here for the moment as see what questions there are, or indeed if anyone actually gives a shit.

[Q. G. Pennyworth]

I am very interested in the topic, don't think I have anything useful to add.

[LMNO]

The barrier to entry to play Beat-em-up game is firstly obtaining the game. For competitive play a second controller or net connection is now usually required. Now why do these games tend to have a bad rep for misogyny/racism etc? Well lets think about the Skills being tested.

I'm not sure if this was actually answered.

Think about playing beat-em-up game at home with friends compared to online. You may be GOD at home but you'll probably be beaten like a child in online competitive play. Why?

Or this.

There is an built in assumption that everyone will act both Rationally and with their own self interest in mind.

When I first figured this bit out, I immediately understood why a lot of game theory doesn't mean shit in the real world.

[Junkenstein]

I knew I'd miss some shit, I'll try and cover those now.

Beat-em-up's bad reputations of late seem to stem from a combination of an inital low barrier to entry which becomes much, much higher when entering online competitive play. The required skill level and need to know particular game mechanics such as Cancelling at certain points of an animation, particular ranges and speed of attacks etc. Extend this to FPS with equipment load-outs and map memorisations. The level of skill you had was previously acceptable. To play competitively and have fun in the new environment you must be willing to raise your personal skill level. Most will either choose not to do this or will be unable to do this. The result is ragequits, abuse and threats.

Also, the typical demographic FPS/Beat-em-ups are typically pitched to are Teenage-Young males. This is a section of society well known for it's reasoned approach to failure. This is changing somewhat, but "Dead or Alive" style games are still being churned out. The demographics pitched to in this case should be quite obvious.

I hope that's a bit clearer on both aspects.


Most game theory in real life is pretty much useless. That said I do find it to be a useful lens at times. I've found that by identifying when a person is acting irrationally in a game they will usually have self interest in mind, it will just be external to the game in question (Such as letting your spouse win the argument. While you "lose" the game, you "win" by ending the argument. Bad example, I'll expand that if needed). It's helped me understand some decisions people make when I can see no real benefit to them.

[McGrupp]

I think this is a cool topic. The two games that stick out for me are Tribal Wars and EVE online. Particularly I was fascinated by the concepts of 'honor' that cropped up in both of them, more so in EVE. One mindset valued 1 v 1 combat and kill/death ratios above all (and also geared and skilled for 1v1 combat) while another mindset valued winning battles above all (a motto often heard was 'if you find yourself in a fair fight, you planned poorly)

Obviously most players are somewhere in between although I always found it fascinating at the vitriol and hate on the forums concerning two different ways to play a game that expressly has no defined goals.

[Junkenstein]

Honour or Player self regulation is possible in games that allow communication and punishment.

Punishment - An In-game threat has in game consequences. This is generally more effective in stateful games like EVE as you can ID other players(to an extent) and continuously follow and attack them. This can both raise the level of effectiveness of Trolling/Greifing. It also allows the larger community to intervene should the Punisher be seen to be acting "Dishonourably"

The Monetization model of Eve also impacts on the player base in this regard. You are paying real money for real things and as such things become more real. It might have just been a couple of dollars worth of ship, but they were your dollars. This can result in very basic emotions which some playing these games are poorly equipped to cope with.

I'll pull together another post on Monetization as this really needs to be in the discussion here.

[Junkenstein]

Over the past decade or so there has been a substantial shift in the ways that profit has been produced from games. I'll try and cover some of the bigger ones and the resulting impacts below.

The Traditional model -  By this I mean the standard ways that games have usually been made. A publisher funds a developer and then sells the game. Once the publisher makes their money back, everyone moves into profit and happy times. The pros and cons on the development of the game (and subsequently how well the games mechanics are tested) are fairly well established. Developers get up-front cash and are able to work and eat. The publisher takes most of the risk and potential reward in terms of IP and related copyrights. The potential communities created in this way vary vastly depending on the developer, publisher and genre.

Kickstarter Funded - This is still fairly new. I would anticipate the communities to be very welcoming during development stages for any project because well, they want your cash and to attract new backers. I really need to see more completed games to have a better understanding of what the result will be. I would also expect these games to have a higher degree of attention paid to the mechanics as developers are more concious from the begining of the project that they are developing the game for a community. Opaque strange rules like, Rocket jumps, killing own troops for XP/treasure/just because it's more effective than killing the enemy (I've seen all of these in multiple games) are less likely to occur, or when they do it will be more clearly signposted. What I'm driving at is kickstarter allows developers to establish the barrier to entry for the community well in advance. Everyone is on the same page. This may be overly optimistic, but I've got to have hope in a couple of things. 

Micropurchase and DLC - These two are very related and there's kind of two categories between them, Games that are solely funded in this regard and games that add these element retrospectively after launch.

Microtransactions are very common in online gaming and particularly browser games. These games are focused around the mechanics of getting you to use them as part of your daily routine. This results in the main skill being invested in is time spent playing. Pay to speed things up or get friends involved to speed things up. Or Enjoy the 18 hour wait to build a X. This results in the longest or most popular (or richest) players dominating any competitive aspects. I shouldn't have to tell you what a community that allows you to buy power looks like. Same for ones run around the largest peer group or popularity campaign.

Microtransactions that alter praeto effiency within a game can turn a community to shit in seconds. In FPS game, if you can buy a shotgun, even for 0.01 spacecash, that is even fractionally superior to the standard there will always be a massive divide between those playing for FREE (As you no doubt advertised, eventually, if no one was subscribing. See below) that will invariably descend to slurs and ragequits. This problem will only increase exponentially the further the purchase is from the set balance.

DLC is becoming increasingly common in all games as is DLC released shortly after launch.  Community reactions to these choices again tend to be divisive and negative.


Subscription Based Model - These tend to be stateful games and developers are very concious of trying to get as many people playing for as long as possible. DLC is frequently ornamental or minor designed to provide a new distraction with minimum prateo disruption. This model has been retroactively applied to and been removed from a variety of games. The resulting change to the community is rarely better in either direction. Publishers also love this model as they expect the next WOW but seem to ignore the resulting needed changes in core mechanics.

"Arcade" - Worth noting as these were large and I suspect will come back in a big way over the next few years. These kind of games tended to be much more difficult and exclusive than home versions, designed to extract cash gradually for an amount of play. Old physical arcades sometimes had a good atmosphere with groups of people competing. A win for the Game owner and Arcade owner when there was something new and popular. I'm not sure what this will look like when publishers try and charge X spacecash per play of new game. It's certainly possible with current tech, and I would expect something of a push in this area over the next 5 years.

I guess that's pretty much the core ways

[Junkenstein]

I think this is a cool topic. The two games that stick out for me are Tribal Wars and EVE online. Particularly I was fascinated by the concepts of 'honor' that cropped up in both of them, more so in EVE. One mindset valued 1 v 1 combat and kill/death ratios above all (and also geared and skilled for 1v1 combat) while another mindset valued winning battles above all (a motto often heard was 'if you find yourself in a fair fight, you planned poorly)

Obviously most players are somewhere in between although I always found it fascinating at the vitriol and hate on the forums concerning two different ways to play a game that expressly has no defined goals.

Also, Tribal wars 2 games got MASSIVE. Like 60 on 60. Crazy shit. You always had a couple of retards either side but everyone just ended up treating them like batshit NPC's. Wonderful times.