Recently I’ve been thinking that a game design is a social contract.  I’m sure this isn’t a new idea, but it’s important to me and I feel like writing about it.

The basic “flow” of game design and play works like this:

• The game designer creates a game.
• The game designer makes that game available.
• The game is transferred to the player(s).
• The player(s) play the game and derive some sort of fulfillment from it.

Some Notes

The “game” includes rules, play materials like a game board, content, software, anything and everything that is provided to let someone play a game.  In this case, “game designer” is really shorthand for an entire game production company.

The game is “made available” by selling it, publishing it on the web, perhaps even providing a client-server arrangement so the player(s) may interact with the rule systems.

When I say the player(s) derive “fulfillment”, that could be “fun” or “enjoyment”…or it could be “interest” or “engagement”.  Like watching a film; take as an example a documentary about war.  It’s easy to imagine such a film that would be most definitely not “fun”, but also a valuable experience.  So there may be games that will provide an experience that is valuable without being “fun”.  (See Greg Costikyan’s analysis of Grey Ranks.)

There might be games that follow models other than this one, and it might be cool to explore them, but I think this model will serve for the purposes of this post.

The Social Contract

So far I’ve just made observations.  What I’ve described is how games work.  But now let’s think about this further.

When a game designer makes a game available, certain expectations come into play.  In essence, the designer is promising that their game will follow the model stated above.  In other words: The game designer promises that playing the game will provide the player(s) with a fulfilling experience. That’s the entire point of selling the game; this concept underlies every game transaction.  The player(s) play the game because they believe it will provide them with a fulfilling experience.  That’s why they follow the rules.

How can the contract break down?

Perhaps the player(s) follow the rules but the game just isn’t fun.  Well, it happens; a failure of skill on the designer’s part, perhaps.  Not every film is wonderful, nor every book, nor every game.

Another failure of the contract is when the designer wants the player(s) to do something within the game, but nothing in the game tells them this action is available.  In a trivial example, if you publish a game with no instructions, you’re breaking the social contract.  Sure, in certain contexts the controls might be “obvious”, but I think it’s unwise to rely so heavily upon cultural context like that.  If someone doesn’t have the requisite knowledge to play the game, they’ll just be frustrated.  It’s in a designer’s best interest to provide instructions.

(An interesting counterpoint to the previous bit is towlr.  Towlr games are intentionally obscure.  However, I don’t believe this is a violation of the social contract, because the towlr webpage says: “Contrary to conventional design wisdom, the towlr games come without instruction and without explanation. Your job is to discover their secret, and score deliciousness.”  In essence, the instructions tell you that if you attempt to solve the riddle posed by each game, you will enjoy the experience.  I think it’s clear that this fits into my social contract model.)

Some games actually attempt to engender behavior in the player(s) that is contrary to the game rules.  Well, I say “some” but I only have one example at hand; Vampires, by Victor Gijsbers.  You can read through the game’s rules, the author’s commentary, Greg Costikyan’s commentary and several posts on the subject (including a couple by myself and a few by the author).  I won’t say any more about the game here, except to say that I don’t think anyone would, or even should play the game “as written”.

This might work as a Flash game or a console game; it requires multiplayer.

The idea is that it’s an ordinary racing game, say like Burnout, Gran Turismo et al..  Or it could be a simpler top-down game like 4×4 Off Road or even RC Pro-Am way back in the NES days.  There is one player-controlled car in a “pack” of racers.

The point of the game is that there are four people playing at once, trying to control the car.  The controller inputs are put together in a voting algorithm that decides what input is actually used at every instant.

This could make for some interesting gameplay.  Racing is the kind of game where, if there’s one path that’s really good and one path that’s less good, a path that’s “between” those two paths will also be “between” them in terms of results.  The function is monotonic, in other words.  Of course that’s not true, but the areas where it isn’t true might be interesting.

I could envision a more skilled player trying subtle maneuvers and being frustrated with the less skilled players…However, at the same time the less skilled players are being shown more skillful ways to play the game!  A team of four really skilled drivers might be in perfect sync…or they might have completely different ideas of what to do!  In the end it might be the teams that work together well that will do the best.  (Perhaps a tournament mode with 16 players driving 4 cars?)

Certain games allow players to customize aspects of their character or in-game situation.  Consider Cosmic Encounter; each player has a special power representing their particular alien species.  Classically these powers are randomly selected, in the form of cards dealt out to the players at the beginning of each game.  Of course, some play groups might let players choose their alien races, or else negotiate the distribution.

That’s the great thing about powers on cards; you can easily shuffle them up and deal them randomly, or you can let players choose in some other method.  Those two methods would provide for different experiences; therefore, players have an easy way to vary their experiences of they choose.  The really important thing is that you’ve allowed players these two options without any extra work.

What other scenarios might profit from this strategy?  Some games allow customization, so let’s think of doing them randomly.  A randomly chosen character class in a game like Kingdom of Loathing?  How about a randomly generated Master of Orion 2 race?  How about the other side, a randomly generated option that might benefit from direct choice?  This one is actually much harder to think of; I suspect if people play a game where this happens, somebody eventually hits upon the idea of letting people choose; Diplomacy comes to mind.  I’d be happy to hear more examples in the comments!

I recently had a bit of an insight into game design.  Perhaps it’s something that other people know about, but I’ve never heard it articulated before.  In any case, it’s important enough to me that I’ve started thinking of it as “the Chaoseed Principle”.

First, let’s think about tabletop RPGs.  Specifically, let’s think about the Lumpley Principle.  To paraphrase Vincent Baker: “A system of rules is the means by which players agree to imagined events during play”.  This leads to a sort of flowchart of actions.

1. A player proposes that something happens in the game world.
2. Negotiation occurs with the other players, drawing upon the rules.
3. The negotiated action is considered to “have actually happened” in the game world.

There are a few interesting things about this flowchart.  For one thing, when I say “player” I could be referring to the “game master” or other similar position that often is found in game rules.  When I say “the negotiated action”, it may turn out to be something entirely different than the proposed action.  It could be that nothing at all happens.

That’s all well and good, and I could talk about the Lumpley Principle for hours.  However, I want to look at a slightly different facet of this subject.  The Lumpley Principle talks about the negotiation, the second and third steps…

…But what about the first step?  How do players decide what actions to take?  Here’s where we get to my little insight.

One responsibility of the system is to suggest actions to the players.

Here, by “system” I’m referring to the “setting”, “flavor text”, “background story” and even “artwork” involved in an RPG release.  Some people might consider that a little too inclusive.  (Perhaps “product” would be better, but for now I’ll stick with “system”.)

I think it’s appropriate to think of the Lumpley Principle as concerning itself with resolving (or heading off) arguments as to “what actually happens” in the game.  But an RPG has to do more than that, it has to suggest actions to the players.  It has to provide seeds of interesting stories.  Sure, you know how to simulate all sorts of actions—but what actions do you take, and why?

D&D suggests you portray a warrior who decks himself out in powerful magic items and engages in elaborate tactical planning.

Ars Magica suggests you follow the story of a wizard in medieval Europe who spends decades training his magical skills and researching specific spells…not to mention scribing and copying arcane tomes.

Call of Cthulhu suggests you experience the adventures of a Lovecraftian protagonist coming up against supernatural forces that drive him insane.

My Life With Master suggests you portray a monster created by a mad scientist, who rebels against his creator but is so wracked with self-loathing that he is driven to suicide.

These are not stories that players might have considered roleplaying a priori.  A group of roleplayers can sit down at a table and play out any story imaginable without any rules.  But it’s like staring at a blank page; The totality of possibility is a difficult thing to grasp.  It helps to have suggestions from the rules, seeds that will sprout into enjoyable stories.

Now let’s pull our gaze back a bit.  The Chaoseed Principle doesn’t only apply to tabletop roleplaying games.  In fact, it can apply to any game.  Final Fantasy VII, for example, suggests you play through a very specific story.  What I find most intriguing are games that provide interesting “seeds” without constraining the action very much.  Spore, or SimCity, or Dwarf Fortress.  Sure, you could sit down and try to draw an alien monster, but Spore gives you body parts and coloring tools to suggest a host of interesting species.

So, someone I know has created this cool new superhero universe-collaborative writing project-thing called “My Super First Day”.  And I’ve contributed to it.

Whisper

Pretty fun.  This whole writing project thing may even continue.

Equipment.  Why do games have it?  Is it to give the fantasy a bit more realism, or at least plausibility?  Is it because all the other games have it?  The truth is that there are solid design reasons to have equipment in your game.  However, it’s possible some developers don’t even know the good reasons, and that can result in shoddy design.  So, since I’ve been thinking about this recently, I thought I’d go over this basic game design area.  It all has to do with interesting choices.

Let’s take an example; let’s say there’s some item that provides 5 Defense.  Then there’s another that provides 10 Defense.  In some games, these bonuses will simply accumulate; in this case, it’s just a matter of getting every item.  That’s not very interesting, though.  What if these items were, say, helmets?  Then you could only use one of them.  In that case, you’d still want to accumulate every item, or at least the best one you could find at that moment.  There might be some issues with balancing out, say, what you could afford to buy vs. how good it was.  But mostly there’s not much of interest here.  Not only that, once you get the 10 Defense helmet, you forget all about the 5 Defense one.

However, let’s say that there’s a helmet that provides 10 Defense, but also one that provides 5 Defense and 5 Speed.  Now this sounds like more of an interesting choice.  Of course it all depends on the stat system, but here are some of the things we think about: Is Speed more important than Defense?  Is it more important in some situations?  Would it be profitable to wear one helmet in one situation and the other helmet at other times?  What if you can also change your body armor, which has its own varying effects; What strategy will you use in choosing your items there?  Will you specialize in increasing one stat, or will you take a more balanced approach?  With more types of equippable items, there are more profitable combinations.  These all lead to interesting choices.  Not only that, in this scenario the “lower-level” items are not simply superseded by later ones.

Many games allow a character to accumulate items that provide benefits.  It’s definitely worthwhile to think about these systems of benefits, to make sure you have some interesting choices for the players.

After writing my previous article on leveraging user-created content, I’ve been thinking about the subject some more.  I had one more insight I felt I should share.  But first let’s back up a bit…

A lot of times in games you work with sets of data.  Data is just information, 1s and 0s, the contents of variables; it doesn’t have any meaning in itself.  For data to mean something, it has to have context.  You “view” the model in a certain context.

When I talk about “leveraging” content, all I’m really talking about is switching the context.  A classic example (one example where software actually does leverage user-created content) is Streets of SimCity.  This was a racing/combat game published by Electronic Arts; the player went on missions involving driving around a city.  The really interesting thing was that the player could import city maps from SimCity 2000, and these cities would be rendered in full 3D to create custom driving environments for the player.  (SimCopter had a similar feature; the player could fly a helicopter around an imported city.)

The point here is the context switching.  In the context of SimCity 2000, a city is (roughly speaking) an arrangement of buildings and zones that have economic effects on each other.  However, in Streets of SimCity, the city is an arrangement of buildings and zones that define 3D geography.  The underlying data has not changed, it’s just that the games view it in different contexts.

With this in mind, we revisit the question: In what situations would it be appropriate and/or easy to leverage user-created content?  Now the answer is more apparent.  To leverage content in a new situation, you must view it in a different context so it has a different meaning.

I recently picked up Zoo Tycoon: Complete Collection in the bargain bin and I’m having fun with it.  I love games where you get to create and develop things, and Zoo Tycoon does its job well.  (It has isometric 2D instead of 3D, which means less processor load, brighter colors and more visibility!)  However, playing it has got me thinking about “Scenarios”.

When you start up Zoo Tycoon you can play a “Freeform” game, a complete tabula rasa which you can develop in any fashion you see fit.  There is a losing condition–your zoo loses so much money that it goes out of business.  However, there is no true win condition and no restrictions (other than initial conditions like zoo size and starting money).  Instead of playing a Freeform game, you can instead play through “Scenarios”, each of which gives you a very specific set of constraints.  Often the zoo is partially built when you get there, and often your construction options are limited.  There is always a set of objectives that must be completed, within a set time limit, for you to win the scenario.

Of course, Zoo Tycoon isn’t the first game to implement scenarios; they’re a pretty common tool in strategy games.  SimCity Societies encourages you to attain Achievements, each of which is basically a scenario in itself.  Galactic Civilizations 2 has a number of scenarios that form a history of the war against the Dreadlords…and some that are completely different, just for a change of pace.

So, what is a scenario exactly?  I’ll go ahead and advance a definition:

A scenario is a set of objectives and constraints within a game.  It is possible to play the game in a completely unrestricted way, but while playing the scenario the player attempts to fulfill the objectives while working within the constraints.  There is an initial condition, the situation in which the player starts, and a win condition, which the player is working toward.

The interesting part of this definition is the second sentence, which implies that a scenario places unusual constraints on the player, thereby creating an experience that is somehow outside the norm.  Without this caveat, the definition could apply to anything.  It could apply to Super Mario Bros., for example; the player attempts to defeat Bowser while working within the constraints of the level designs and gravity.  This reductionist viewpoint could be useful, so we’ll keep it in mind.

So, to have a scenario you need gameplay that is restricted in a special case.  A good example of this is the web-based game Billy vs. Snakeman, specifically the Wasteland missions.  During Wasteland missions, ordinary Strength bonuses don’t count; Range, therefore, becomes much more important.  The usual bonuses of allies do not take effect; only special Wasteland allies can be used.  Similarly, Chakra costs are multiplied by 10 (and even, eventually, 100), meaning that only special Wasteland Jutsu can be used.  The player must scramble to find items, allies and bonuses that will allow him or her to solve these special missions.

The Wasteland missions are basically a “side quest”; there is nothing that forces you to complete them.  When you start out, you have no Wasteland gear, so your initial condition is fairly weak.  There is a win condition–one final Wasteland quest that provides a permanent reward.  (There is no “losing condition”, in that the only way you can lose is by giving up and turning your attention to some other part of the game!)  So, in my view, this is a good example of a scenario.

The interesting thing here is that most people might not think of Billy vs. Snakeman as a “strategy game” in the same vein as SimCity Societies or Galactic Civilizations 2.  However, it does allow the player to develop a situation (your character) over time, and it does allow a certain amount of freedom as to which actions to perform.  And if you don’t like your situation, you can start over…by creating a new character, if nothing else!

Another valuable example is Kingdom of Loathing.  Another web-based character development game, KoL focuses on humor, but it does have a lot of content and some elaborate strategy.  To summarize: Once you have completed the main quest in KoL, you may “ascend“.  When you ascend, your character’s status is reduced to starting levels and you do the whole thing over again–with a couple of differences, of course.  You may change your class, which results in a very different experience.  You may also choose a few sets of constraints to make your new playthrough more exciting–in other words, scenarios.  “Casual” allows you to use items from your previous ascension(s); “Hardcore” does not.  “Teetotaler” disallows consumption of alcoholic drinks (a good source of extra turns and stats) in exchange for the promise of a special bonus item upon completion of the run.  “Moon signs” unlock different special areas, each of which provides their own unique items and bonuses to assist you in your new life.  Not only that, players have created their own scenarios to play through to provide even more interesting experiences.

Kingdom of Loathing has enough variety in the player actions that these scenarios can be interesting; the constraints can really create new experiences.  But let’s go further, let’s try to hybridize these two kinds of scenarios.

What if, in Kingdom of Loathing, there was an accessory that provided a stat cap; each stat could not be more than 50 while you were wearing it.  And there was an area that you could only visit if you were wearing that accessory.  And there was a quest that required adventuring in this area to receive a special reward, with monsters that were difficult.  Players would have to find skills or items that provided effects other than stat bonuses to defeat the monsters guarding the treasure.  This would provide an experience outside the norm, and that is the point of scenarios.  This is a tool that could be applied to many games, opening up the range of experiences they can provide.

For years now I’ve been interested in games as tools for creative expression.  I like games where you don’t just develop a skill, you create something as you play.  Once you have created some bit of content, it’s possible to reuse and repurpose that content; to leverage it.  There is one obvious game to talk about, an elephant in the room, but I’d like to start with a simpler example.

Recently I’ve been playing a web-based game called Billy vs. Snakeman.  It’s a parody of various anime series, but it’s also a fun game in its own right, with some clever features.  (Here’s a link where you can give me referral points by making a character. )  In its most basic elements, BvS deals with developing your character over time.  You increase “your” stats and collect items; these stats and items allow you to pass challenges within the game.  The interesting point here is that you’re not just experience the game, you’re also creating a piece of content–your character.  That character exists within the database whether you’re logged in or not.

BvS has a feature called the Arena where you can “fight” other characters.  In truth, this isn’t really like a player vs. player (PvP) thing; whether you win or lose, you don’t affect the other character at all.  (There are more PvP-oriented aspects of BvS, if you’re into that.)  When you perform the “Fight in the Arena” action, a character is randomly chosen from the database to be your opponent.  That character and your character are compared to see how they perform against a random challenge; ties go to your character.  If you win, you get “Arena Reputation”, a currency that can be spent on certain items (items that can only be purchased with Arena Reputation).

The interesting thing here is that the second character is simply a piece of content that exists in the database.  The other player is not notified and is not affected in any way.  However, that player has spent time building up the character’s stats and items, as well as creating a customized name and possibly an avatar image to represent that character.  So each character is an interesting piece of content, and the characters are leveraged to create an interesting experience for this particular feature in the game.

Now for the more complex example–Spore.  Spore consists of five phases, but in terms of this post they each have the same game flow.  When you play Spore, you are creating something–a cell, a creature, a building, a spaceship.  Usually these bits of content have restrictions on them having to do with gameplay; for example, creatures need legs and feet to move around, so all created creatures have legs and feet (unless the player specifically tried for a pathologically strange one).  With that in mind, and the social and technical design of the “creator” subprograms, most of the content looks appropriate; that is to say, creatures look like creatures that can walk around, buildings look like dwellings where creatures could live and work.  (Whether the content is socially appropriate is another question entirely!)

Once content is created, it usually gets shared to the Spore servers (user settings can change this).  What this means is that your creature gets uploaded to the server, then it can be downloaded into someone else’s game.  Then when they wander their galaxy and explore alien worlds, they might find your creatures living on those worlds.  Similarly, when you wander your galaxy, you find it populated with creatures created by other players the world over.

As we can see by now, Spore was built around the idea of leveraging content created by users.  User-created content is shared to make other users’ games more interesting.  The content is used in a “faux-multiplayer” way.  You meet other users’ creations as if they were other players playing the same game that you’re playing.  They answer the challenges of the game in their own ways, and you get to see the result and compare it to your own strategy.

The faux-multiplayer idea has one big advantage–it’s easy to design.  You can take one user’s data and treat it as if it existed in another user’s game world.  You can have both sets of data following the same game rules.  This is fun because it can inject more interesting variety into the games; the assumption is that the process of play guides the players to create interesting content.  One pitfall is that players might arrive at the same answers to the game’s challenges, resulting in everyone’s data looking the same.  This is an issue worthy of its own post, but let me say that Magic: the Gathering has addressed this problem better than anything else I’ve seen.  Magic is solely a multiplayer game where each player plays with a customized deck of cards; there are well-nigh unlimited combinations of cards that would stand a chance of winning, each with their own strategies.  Therefore creating a deck is itself a piece of creative expression that gets pitted against an opponent.

Now that I’ve gone over the basics, I’d like to speculate about new directions.  The way I see it, content leveraging can be divided into two segments; you encourage users to create interesting content, then you adapt that content in such a way as to improve the experience for someone else.

At this point I’d like to talk about a couple of web-based games I’ve created.  First is Phantasma; in this game, players portray wizards inhabiting a magical castle.  The emphasis is on developing your stats through “research”, learning spells and creating enchanted items.  Next is Chaostorm, a more abstract sort of game focused on creating items with procedurally-generated “recipes”.  Players can ultimately create “Scopes”, which assist them in finding items they need, and “Battle Items”, which boost their stats for PvP-ish contests.  Of course, because I’ve created both these games, I have access to all the content for both of them.  In Phantasma is a location entitled the “Kipatsu Shop”, known for selling items from “other worlds”.  In this case, the shop sells items from Chaostorm!  Randomly selected Chaostorm items are used as templates to create Phantasma items with appropriate power levels and prices.  Chaostorm Scopes are sold in Phantasma as “Elemental Scrutinizers” that assist a wizard’s research into the magical discipline of Elementalism, and Chaostorm Battle Items are sold as “Elemental Projectors” that increase a wizard’s spellcasting ability in that same realm.  The name and description of the item are imported directly from Chaostorm (with “Elemental Scrutinizer/Projector” prepended to the item’s name).  In this way, the content from Chaostorm is used to create interesting new items for Phantasma, in a carefully controlled process.

Chaostorm is a game designed to encourage users to create interesting content.  However, even less “experimental” games can yield intriguing bits of content.  As we saw with BvS, a player character itself is the sum of the player’s choices, their answers to the game’s challenges.  A long-time player of BvS has created an elaborately customized piece of data that represents a personality within that world.  This would hold for all sorts of games classified as “RPGs”, whether multiplayer or not.  Other games yield different types of content; SimCity is an obvious example.  (Will Wright came up with the idea for SimCity while designing maps for the background of a helicopter game; he liked designing cities so much he made a game out of it.)  With SimCity the player is tasked with creating a city.  There are a slew of city-building games that imitate this design; Caesar, Cleopatra, Stronghold, et al..  However, strategy games such as StarCraft have city-building elements as well, even if they’d be more likely to call it base-building.  There are any number of space-trading games that feature customizing one’s starships.  Oblivion allows one to purchase dwellings and fill them with furniture, although there isn’t much in-game encouragement to customize your home exactly how you wish.  As I see it, there are two ingredients for interesting content creation; there must be restrictions to guide your users into creating content that makes sense, and there should be enough possibilities that not all content is identical.

The second part of the process is adapting the content for new purposes.  I believe this is the part where there are still great possibilities for advancement.  Once you have a city, for example, you can have the character walk through it–but that’s easy.  What if there was a game where the player bought a city in a bottle?  SimCity Societies allows the player to create cities that have different “stats”, such as Spirituality or Industry.  Perhaps the city in a bottle is an item–the character could wear it around their neck in the “necklace slot”.  And cities with high Spirituality would provide bonuses to MP or Magic stats, while cities with high Industry could increase Strength.  Or, you could have an item that represented another character–made into a voodoo doll, or maybe an item that provides a link to their strength.  And that would provide some customized bonuses depending on the other character’s stats.  The point is that there are many different types of content that games require, and many of them can be “filled” with data provided by other games.

(EDIT: I wrote a little more on this subject for Part 2 of this article.)

Previously I talked about using images in AS3. Now I’ve figured out a further enhancement to the procedure I described.

In that previous article I talked about adding an image to the library and instantiating it as a BitmapData object.  Then I talked about creating a Bitmap object, but there’s a problem; in the AS3 reference, on the Bitmap page, there’s this nugget of information:

Note: The Bitmap class is not a subclass of the InteractiveObject class, so it cannot dispatch mouse events. However, you can use the addEventListener() method of the display object container that contains the Bitmap object.

This makes it impossible to turn a Bitmap into, say, a button.  However, there is hope!

Instead of making a Bitmap object, make a Sprite object.  Then fill the Sprite’s graphics member with the bitmap.  Specifically, use the beginBitmapFill(), drawRect() and endFill functions.  Here’s some basic code to get you started:

var t_IconClass:Class = getDefinitionByName("picture0.png") as Class;
var t_BitmapData:BitmapData = new t_IconClass(0, 0);
var t_Sprite:Sprite = new Sprite();
t_Sprite.x = 64;  // Position it wherever you want
t_Sprite.y = 128;
t_Sprite.graphics.beginBitmapFill(t_BitmapData);
t_Sprite.graphics.drawRect(0, 0, t_BitmapData.width, t_BitmapData.height);
t_Sprite.graphics.endFill();
t_Sprite.addEventListener(MouseEvent.MOUSE_DOWN, SpriteClickFunc);
addChild(t_Sprite);