A Manifesto for MMORPG Design

      Note: this document is a work in progress.
Note also: this document dates from 2002 and aside from this comment has not been altered since that time. It was an incomplete, academic approach to various perceived problems in the MMO class of games. Modifications and enhancements are long overdue based on my employment since November 2002 as a maintainer of a long-running MMOG.


Massively multiplayer online roleplaying games have achieved a tremendous level of popularity within the broader online community. They are addictive forms of entertainment in no small part because of the companionships and communities that form among the players. However, these player interactions often happen in spite of the mechanics of the game, rather than being fostered by them. Moreover, although the games call themselves "roleplaying games", they often exhibit features and mechanics that hardly represent the state-of-the-art in RPGs. Instead, most hearken back to the earliest forms of roleplaying game, commonplace in the late 1970s and early 80s, where statistics and advancement within a linear hierarchy were more often emphasized than story and characters. Within the modern roleplaying games community, these types of game are now deprecated, even considered slightly pathological, in favor of more free-form or storytelling games.

Perhaps the most common complaint of the online gamer is that the game is "boring", or that there is "too much repetitive work" at "low levels". This is a clear indication that gameplay within a hierarchical system is stifling the entertainment potential. However, many players will also state that they "keep playing to be with their friends", showing that the community ties and friendships are a dominating influence, even in the face of less-than-exciting gameplay. This manifesto is an effort to describe means of organizing and planning an online massively multiplayer game in such a way to emphasize the community-building and player-to-player relationships that are a strength of this genre, while proposing approaches and game mechanics that limit or ideally, eliminate, the boredom all too often seen, particularly among beginning players. It must be emphasized that this type of online game is a form of purchased entertainment, thus the object must be to retain paying customers (players) so that the company producing the game can survive, profit, and continue to deliver entertainment. In order for a game company to retain customers, the players must enjoy the game experience.



AI: "artificial intelligence" Used loosely in this sense, to mean agent-type algorithms that act within the game environment. AI is typically used for infrastructure that players can interact with, such as a shopkeeper NPC, or for adversaries. AI is also used to describe abstract processes within a game that need to evaluate conditions and make decisions without human referee intervention.

alternate unreality: A term for multiple copies of a game world which exist independently from one another. There are usually game-specific terms. For example, the many copies of the Ultima Online game world are called "shards". Each runs on a separate server, with the intent that its players are located physically nearby to reduce network delays affecting game response.

character: The persona or role taken on by a player when he or she enters the game environment.

game: A contest in which there is an objective, shared by the participants, called "winning", which usually ends the contest. It can be argued that most MMOGs are toys (q.v.), not games. However, the term "game" is common parlance and will be used throughout this document.

game company: The real-world company that either designed or publishes and supports an online game. No distinction is made in this document between development and publication, which are acknowledged as different roles usually assumed by two different companies in the real world.

GM: "gamemaster" The referee in a traditional role-playing game. In an online game, this is usually a representative of the game company, with ability to resolve problems or provide support to regular players.

grief player: A player who entertains himself through actions that actually harm the game experience for others.

IRC: "Internet Relay Chat" An online means of communication, in which a large number of people can participate in simultaneous or private text conversation in a "chat room", "booth" or "channel". Methods of expressing action and emotion developed in these environments via text cues and eventually, environment-manipulating commands. This led to the creation of the first online role-playing environments (MUCKs, MUDs, and MUSHes).

level: A generic term for an interval of advancement in a hierarchical system.

levelling: Player slang for focusing one's actions solely to gain levels within the hierarchy as quickly and directly as possible. Also, "powerlevelling".

massively multiplayer: A type of online game in which a very large number of simultaneous participants is possible. This is as opposed to normal multiplayer online games, which typically involve a number of players between two and fifteen, and which may employ a central matching service enabling players to find each other's games.

mechanics: Algorithms that are inherently part of the game system.

metagame: Activities that are directly related to gameplay, but which do not happen within the game environment.

MMOG: Massively multiplayer online game.

MMORPG: Massively multiplayer online roleplaying game.

MUCK/MUD/MUSH: "Multi-User Chat Kingdom / Multi-User Dungeon / Multi-User Shared Hallucinaton" Predecessors of the MMORPG, chiefly text-based descriptive environments which could be variously navigated and/or manipulated by the participants, which very often were predominantly role-players. The environments themselves were often persistent.

newbie: Slang for a player novice to a certain game. Newbie players are usually just starting out in a particular game environment, and their character may be less capable in several ways when compared to the characters of veteran players. Newbies will as a rule also be less knowledgeable about the game environment than veterans, which is a still greater weakness.

NPC: "non-player character" Both a noun and an adjective, this term refers to characters or entities/agents not operated by players. In an online game, this can be AI. NPCs can be infrastructure, like shopkeepers, or adversaries, like AI monsters or other opponents appropriate to the game's genre. NPC adversaries in many games are called "mobs", short for "mobile objects".

persistent universe: A game environment available 24x7, which is theoretically not reset by the administrator.

PK: "player-killer", or "player-killing" Deprecatory term for PvP activity, or players who frequently engage and kill the online characters of others.

player community: The informal organization of all players participating in a game.

player organization: An organization formed by players, that has as its purpose objectives within the game environment itself. Often also called a "guild" or "clan". This is as opposed to a social club made of players of a game, which might have as its purpose interaction outside the game enviroment.

PvP: "player versus player" A style of gameplay in which players directly contest with one another, usually violently.

real world: The non game environment, in which the players and the game company exist. As opposed to the game environment itself, in which the characters exist. Also "RL", or "real life".

RPG: "role-playing game" A sophisticated game in which the players enact a story by taking on the roles of participants in the action. Often, this is not unlike communal storytelling or the creation of a dramatic play by the actors themselves. One participant usually serves as a referee, whose task is to enforce whatever the rules of the play environment are. Interaction among the participants is the activity from which entertainment is derived. RPGs have evolved from paper-and pencil simulations or games into live-action and online forms. In the online form, a computer program can potentially serve as the referee, allowing all participants to play.

toy: An environment allowing interaction or manipulation, but in which no objective is defined for the participants. It can be argued that most MMOGs are toys, not games (q.v.).

Key Principles:

  1. Make the Players Do It
  2. Directly incorporate natural metagame activity into the game's basic infrastructure
  3. Role-Playing By Harnessing People’s Lust for Self-Improvement
  4. Avoid hierarchical rankings as a general way of organizing the player community
  5. Ensure players have a functional role no matter their prestige
  6. Construct a dynamic environment, in which player activity makes a visible difference
  7. Establish a persistent game environment - one game for all players, worldwide
  8. Examine the tradeoff between playability and realism
  9. Incorporate player-created Organizations directly into the game’s background and/or storyline activity
  10. Make use of social factors to provide both PvP and PK-free options to players
  11. Incorporate a fame/prestige system - independent of any institutional rank players may obtain

Notes on the principles:

(1) Make the players do it - This is fundamental. Ultima Online has shown that players willingly assume roles that an observer might at first consider infrastructure-related, given game mechanics that allow them to do so. This refers to environmental infrastructure, for example: merchants who move supplies around the game world, builders who construct needed items or structures, even people taking up adversarial roles that create conflict and opposition. Game designers spend massive amounts of effort and thought toward providing and balancing AI and support mechanisms that perform these roles. In an ideal case, this burden can be dispensed with if the players are provided mechanisms within the game that allow them to provide for themselves. The objective is to allow the game environment to be self-sustaining.

(2) A great deal of game-related activity occurs outside the game: examination of character stats on the web, negotiation by email, discussion and propaganda by players about other players or their in-game organizations, arranging tasks to be done by other players. Mechanisms permitting all of these types of activity should be provided within the game environment itself, maximizing player online time. Player online time is generally what the game company charges for, after all.

(3) Role-Playing By Harnessing People's Lust for Self-Improvement (RPBHPLSI) - This is a transitional concept very valuable when considering issues of player motivation. If the players are rewarded for an action in some way, players will perform those actions, often repeatedly. Character advancement is the primary motivation for most players, even in games where character advancement can be shown to be practically meaningless. This realization can be used in game design to manipulate player activity along desired paths. Citation for the RPBHPLSI concept to Greenbean of www.playerofgames.com

(4)(5)(6) Changing the way players are rewarded in-game is a major element of a player-centric design strategy. This is really an interrelated group of concepts that empower the players to participate directly in all aspects of what happens within the game environment. This simultaneously opens up wide gameplay possibilities for the players while potentially reducing game support efforts involved with keeping the game interesting.

(7) This is an ideal, restricted by hardware and network realities. However, the fewer independent "alternate unrealities" of a game world there are, the easier it is to support them, and the easier it is to maintain a coherent player community.

(8) Even the ideal case has to be pulled aside for a sanity check. An extreme and simplistic example: A perfectly modelled real-time simulation of 17th century colonial trade in the Caribbean might sound great. However, no one can or will stay online for the times necessary to sail commonplace distances. Making distances shorter or ships travel faster both violate realism, but solve the immediate playability problem.

(9) This is another facet of Making the Players Do It. If the groups the players create themselves are allowed to align with and come into conflict with the governments and organizations that already exist as part of the environment's design, a significant amount, perhaps the majority, of conflict creation will occur naturally based on human social dynamics. Storyline involvement for the active player organizations is unbelievably powerful as a motivator keeping players interested in the game.

Dynamic Environment

The presence of a dynamic backdrop can make or break any MMOG. One danger of a poorly-established persistent universe is a static background. It is far more interesting to the players if they can see things change over time: from minor, trivial things, like weather, or seeing a building slowly being constructed week after week, to major, fundamental things, like a change to part of the game map or a once-friendly government becoming hostile. Numerous means can be used to bring this sense of dynamism into a game environment. Most basic, however, is the conceptualization of capacity for change during game design. Potential changes in ownership of territory or parts of the environmental infrastructure, phased construction of facilities so they don't just appear in the world instantly, or allowing for the possibility of shifting relations among governments, are all things which should be conceived in the design phase so that no potential change causes a coding crisis for the development team. Ideally, ownership of everything in the game needs to be fluid: control of any resource, location, or object needs to be a changeable attribute. Major portions of the game's background, such as the stances of governments or other organizations toward each other, need to become programmatically modifiable matrices stored in a database.

The game's economy should also be dynamic, with prices rising and falling based upon laws of supply and demand. Additionally, there must be sources and sinks for resources within the environment, with rates adjustable such that the economy can be kept in some semblance of operation even as shifts in player population and activity occur. Controls on the markets must exist so that irrational events violating laws of market behavior cannot occur - such as markets purchasing items when it would not logically be in their interest to do so.

More controversial is the idea that objects within the game environment should be destructible. Infinite supply coupled with limited demand is a very large problem in many game economies. Destructible objects, items, and resources are absolutely vital for maintenance of a value-based economy. Destructibility can include portions of the infrastructure. There is a danger in that powerful grief players may choose to systematically destroy important pieces of infrastructure, yet there is little that will motivate a heroic player more than logging in to find that his favorite settlement or resupply spot has been destroyed. To counteract the destructive players, means to rebuild infrastructure must simply be made available in-game. This can be an excellent catalyst for cooperative play. Moreover, destructible infrastructure is critical for allowing the game environment to grow and change with the needs of the active players, preventing the environment from becoming cluttered as the change occurs.

In most MMORPGs, a very extensive game environment is considered an advantage. However, a large game area is also a weakness because it becomes easy for many players to "miss out" on whatever excitement might be available, by simply not being physically present. Communication about events within the game environment is an important aspect of inclusion for players. Every MMORPG needs some form of "news service", tailored to the game's genre. It doesn't matter if a character reads a scroll by the inn door or links to the newsfeed via her HUD. The news service can be used in-game to publicize not only scheduled events, but also to simply report on recently-occurring things that distant players would have had no way to notice. If a character is on the far side of Region A, but reads a news story that indicates a border dispute between Region A and Region B is heating up, the player may even be motivated by curiosity to travel across Region A to see what the fuss is all about (or better yet, get involved somehow). Creation of player motivation and interest is a highly valuable natural side-effect of in-game reporting of distant events. Even if a player is not interested in changing what she is doing with her character at the moment, being able to find out about things going on elsewhere in the game world creates the exciting sense that the world is alive. Destruction of infrastructure and changes in political stance or ownership are major events that should logically make the news. The raw event itself (time, location, and description) can easily be spiced up with a variety of pregenerated text tailored to the environment and the putative "news source", adding local color and preventing the news itself from becoming dull. At worst, one member of the game company's support team might serve as a daily "news editor", taking a list of recent events coughed up by the game database and turning them into short bits of interesting prose for in-game publication.

It is even better if the underlying events act progressively upon the environment's relationship matrix. For example, say a group of bandits (NPCs or perhaps even players) are nominally from Region A. This association with the Region A political unit is an attribute of their characters within the game environment. These bandits are destroying infrastructure within Region B. Not only do the destruction events trigger news reporting upon the destruction itself, but the system attributes destruction of Region B owned infrastructure to Region A characters. The relationship between Region B and Region A automatically begins to deteriorate in the game's sociopolitical matrix. The rate of this shift would depend upon the value and frequency of the destructive incidents. This relationship change also makes the news, perhaps in the form of threats from Region B NPC officials against Region A. The plot, as they say, thickens. Of course, one might expect that Region B has resources which could be used to defend its infrastructure from destruction. That concept will be discussed in detail in the later section on conflict.

From the standpoint of game design, the most active, dynamic part of the environment is the players themselves. This engine of change can be harnessed. Easily. Players by nature want to contribute to what they see around them, to be part of what is going on. Allowing the players to become agents of change in the environment is perhaps the most powerful concept in MMORPG design. This has already been alluded to above, when it was suggested that the "bandits" in the example might be players. The easiest way to get a dynamic environment is to simply make sure the players are able to change and manipulate it. The only secret is to make sure that events caused by the players can be just as significant as events generated automatically, by GMs, or as part of some prepared storyline. In fact, the concept of a managed storyline, given enough active players, can potentially be dispensed with! This idea will no doubt seem radical to game designers who demand complete control over the environments they create, but it will be no shock at all to experienced paper-and-pencil RPG gamemasters, who have for decades recognized the value of the story being a collaboration between the players and the gamemaster.

One key caveat in the creation of a dynamic environment is to avoid the trap of false events: If the news or other description claims an event is happening, it has to be real to the players. Players must see tangible results from their efforts and activities. If the players are told that taking some resource to a particular location aids their government in building something, they must be able to see that something being built, if only slowly. Military campaigns must have result, cannot be pointless, and should not return to the initial state when concluded. If there is a battle going on, players should be able to find the battlefield and see it happening, and risk getting directly involved if they so choose. Structures, settlements, and facilities should be constructed and abandoned over time, leaving ruins or debris in their wake. The ruins or debris themselves should be items of interest that could be explored, salvaged, or significant to future activity in some way. Another key caveat is to avoid linear opportunities for manipulation: At least one presently-running MMORPG has belatedly introduced a mechanism similar to what has been described above, except that only one change to the environment can be participated in by players at any one time. This is ludicrous in a simulated society. Some people will want to work on one project, others will be interested in something totally different. Putting it simply, a player with the proper resources or influence should be able to manipulate the environment any way she wants. Obtaining that level of influence, or those resources, should in fact be a huge part of gameplay. This thinking leads to a new major topic...

Player Advancement and Reward

Player boredom may be considered the significant problem in MMORPGs. This problem is threefold: (1) Players may not interact frequently and form relationships with other players. (2) Players may see nothing in the environment that captures their interest, nothing "new and different". (3) Players may feel required to perform repetitive actions in order to advance their characters. Issue (1) is a broad concern for any MMORPG. This manifesto obliquely addresses a number of ways in which MMORPGs can be improved so as to contribute toward increased player interaction. Issue (2) is primarily addressed above. Issue (3), however, is caused by the all-too-common uncreative inclusion of a hierarchical advancement system which literally restricts player activity, or focuses player action in such a way that the players themselves restrict their own activities. Hopefully, the players are "working" toward goals which allow them to obtain desired in-game capabilities when they are repeating these mindless tasks over and over, but that is an optimistic case. Frequently, the players simply have "advancement fever", meaning that their concept of gameplay is solely to advance their character up the advancement hierarchy (usually called "levels", based on longstanding RPG usage). In fact, this problem can contribute directly to the previously-mentioned lack of interaction, because the "levelling" players are often trying to be so efficient at their chosen task that they can't be bothered to chat with other players. Once these players get bored with the repetitive activity which seems to them the most efficient means of advancement, they will be lost as paying customers.

Game designers are usually aware of these problems, but the chosen solutions are too often flawed. Frequently, game designers will try to adjust the hierarchy to limit advancement boredom - attempts are made to "speed up" advancement by changing the rates at which "advancement points" (usually called "experience points", again based on longstanding RPG usage) are earned, depending on where in the hierarchy a character stands. The usual objective is prevent the hierarchy from frustrating the newbies.

Another phenomenon commonly seen is an effort on the part of the game company to progressively add levels to the high end of the hierarchy, along with additional capabilities or attributes that players can aspire to obtain for their characters. The developers perceive, correctly, that the players reaching the high end of the hierarchy will have nothing more to achieve, and will thus grow bored, and leave the game. This is the first-order effect. The developers also rightly fear that the player community as a whole will become discontented when the community gradually notices that the most active players, those able to play often enough and long enough at a sitting to quickly reach the upper end of the hierarchy, tend to grow bored and leave the game. This represents the second-order effect. A race condition therefore is created between the player community and the developers: The developers' objective is to ensure that the small, visible fraction of the players at the higher reaches of the hierarchy are kept entertained. The players, however, become intensely motivated to see what the "next big thing" will be, so they strive monomaniacally to reach the high levels as rapidly as possible. This race condition can consume significant development and support resources, as well as result in a player community with weaker internal relationships because of the levelling practices described above.

The existence of hierarchy itself is the problem. Elimination of the hierarchy appears to reduce the apparent value of RPBHPLSI, yet it will be shown later that this is not entirely the case. What elimination of the hierarchy does do is remove the desperate "levelling" practices, and any potential for a race condition between players and developers. In the absence of a hierarchy, something else must be conceived to motivate players to act.

It is advocated that a completely different paradigm for player advancement in MMORPGs be considered. In fact, the term "advancement" itself is improper. What players generally seek is power, an increase in character capability, or new ways to use those capabilities. What can be done is to create a spectrum of capabilities, based on the kinds of activities available to players within the game environment. The various capabilities should represent direct means to influence the environment in minor ways, with the cumulative effect being that players gradually obtain greater influence through their actions and participation in the game.

The concepts of promotion and rank do have a potential role within the game environment, but these are best relegated to internal aspects of in-game organizations of various sorts. As such, they can certainly represent potential player goals, and increased rank should be directly tied to access to and control over the resources of an organization. Promotion and rank should not be an overarcing organizational basis for the game's mechanics, because most games choose to model one or more interacting societies, not simply the internal affairs of a single centrally-controlled organization.

It is particularly vital that players be permitted to set goals for themselves, and that those goals have the potential to be meaningful in-game. Many games advertise free choice of player activity, but the disillusioning truth is that there are profound limitations on what a player is actually permitted to accomplish. Some game designers also confuse "character type" choices with activity choices, which leads to some very misleading marketing using the word "do" instead of "be".

Players should be rewarded with opportunities to influence the environment. Create a literal connection between "Prestige" and ability to affect in-game events.

Consider instead a "Prestige" or "Social Points" system which models reputation within the game society. One or more indices for "Prestige" may be useful to represent social standing in more than one society, if the game world contains several.

Concepts of currency ($) and reputation (Prestige) might be the only significant metrics within the reward system.
Consider direct means in-game to incorporate fame rather than leaving it entirely to the metagame.

Gameplay Topic: Player Motivations

What purposes do players have within the game?
Players need meaningful goals that are relevant in-game

Are players allowed freedom to "change careers"?
Choice of military and civilian roles
military available for the combat-minded players
civilian activity available for those uninterested in fighting
Avoid "quest-type" motivations - especially ones that are the same for all players. Static "quests" that every player experiences in the same way are destructive to the sense of immersion in a living world. Motivation is completely sapped by the realization that whatever epic problem was just "solved" by some player, other players solved it last week, and it will still be around as a problem for someone else to try to solve tomorrow. This situation is the antithesis of a dynamic environment.

Valuable for players to see results in dynamic environment
Consider methods of tying the prestige system into news generation

news events, corporate politics, inter-government politics

the more/significant activity done by a player, the greater the chance of a news event spawning which describes that activity

Prestige does not change available resources, but should change what opportunities are offered (i.e. certain Corps or Govs might not offer tasks to a player unless prestige has reached a certain threshold, or might not permit acceptance of the task if the mission is offered on an open interface. Related thought: military missions may be segregated by actual military rank of potential recipient).

Existence of a realistic economy, with commodity sources and sinks as well as prices (based on real in-game value) which rise and fall according to supply and demand, allows players to act as traders and merchants within the game environment, moving goods around and potentially acting as an entire segment of needed infrastructure. This is more complex, but also more interesting, than having all goods spawn at distribution outlets like shops.

Availability of criminal or adversarial roles provide an outlet for players interested in PvP: both as aggressors and defenders.

Establishing a player Org (guild/squad/company) should be difficult and should require expenditure of considerable in-game resources; it should be a goal in and of itself

Setting up additions to the environment, such as facilities or buildings, should be an option - and may result in a transition to manager-type play if the project is large enough, such as an entire settlement appropriate to the genre.

Depending upon the size or extent of a project, the resources of a single character should simply be insufficient: it should require the efforts of an organization (leads to next major topic)

Player Organizations

A significant part of the player community for many MMORPGs are the player organizations. Various sorts of teams arose as a natural part of the metagame of many online multiplayer games (and were usually called "leagues" or "clans"). Designers realized that allowing the players to carry these affiliations directly into the game environment would lead to greater player satisfaction. In fact, since players were already appending "clan initials" to their names, it was a very logical course of action for game designers. However, many games still do not apply to gameplay the profound potential inherent in these spontaneously generated affiliations.

The simplest thing that designers must do with player Orgs is to provide in-game support. Interfaces must be made available to the members and officers of a player Org that allow all aspects of the Org to be manipulated from within the game itself. This interface, of course, should be appropriate to the game's genre: Ultima Online (Origin), for example, uses the metaphor of a physical "guildstone" object, which guild leaders interact with to issue commands that set the parameters of their organization and admit or expel members. More recently, Dark Age of Camelot (Mythic) approached the ideal with their guild system, by including commands that allow a guild to customize its internal hierarchy as well as customize member access to special guild functions, such as the ability to admit members, the ability to access guild communications, and so forth. Support for player Orgs can go beyond basic infrastructure through creation of material in the game environment to which only player Orgs have access. This can be as simple as custom (or semi-custom) decals or other ornamentation for Org members, or can extend to more ambitious things like equipment or buildings. Even if there is nothing special in the environment only for Orgs, at the very least an Org should be able to acquire in-game property much as a regular player. The minimal functionality of allowing Orgs to hold "communal" property in some repository for member players is simple, and key to making the Org useful and valuable to its membership. Direct provision for player Org activity within the environment itself can make a huge difference in whether a game's player Orgs are stagnant window-dressing or vibrant catalysts of player enjoyment.

It is worth noting that any support system for player organizations in a game should be constructed defensively. Creation of a player organization is usually seen as a prestigious action in the player community, a way of making a "mark" upon the game world. Unfortunately, this means that if the process for creating a player organization is simple or unrestricted, a vast number of player Orgs will quickly appear with tiny memberships, and with names likely to be more reminiscent of graffiti than appropriate to the game's genre. It is a far better practice to make the act of creating a player Org, not an administrative act of filling out some sort of online form, but instead a process that has a direct tie to gameplay, and which requires an expenditure of in-game resources to complete. Creation of an Org can and should be a major potential goal for interested players, and logically would be an excellent outlet for cooperative play. On the simplest level, establishment of a player Org should involve some sort of transaction with an in-game licensing authority appropriate to the game's genre, such as a government. Org foundation might also require that the founder have some sort of established relationship with multiple other players, guaranteeing that the Org begins with some minimal number of active members. Requiring extensive in-game effort for founding or establishment prevents creation of an abundance of useless Orgs at the whim of players who are likely to prove disinterested in the long term. In keeping which this defensive support philosophy, note also that there must be provisions for eliminating defunct player Orgs. Players have a tendency to want to build something of their own, rather than assume control of something established by others. This tends to result in a large number of defunct player Orgs, with few or no active members, taking up space in a game's database. A robust cleanup process should exist that tracks Org member populations and activity. If an Org falls below a minimum membership threshold, its leader should be notified and the Org should be removed from the game if membership does not increase within a certain timeframe. Similarly, if an Org does not have an active leader, protocols should exist that automatically promote a remaining member into the leadership position, ensuring someone is always at the helm, and empowering that person to take initiative to revitalize the Org. If the entire membership of an Org is inactive, the Org should be treated similarly to an Org with too few members, and the Org should be dissolved after a period of time. Elimination of defunct player Orgs is especially important if Orgs are permitted to control limited resources within a game's environment - unless they are being used actively, those resources must be freed up to prevent stifling more dynamic parts of the environment or community. In addition to a minimal active membership for continued survival, it might be argued that player Orgs should also have a population cap. A cap prevents a single well-developed Org from collecting a majority of a certain type or role of player within the game community, such as all the miners, and ensures possibility of rival Orgs, preventing a Microsoft-like phenomenon. Choosing to place a population cap on player Orgs is more an aspect of environment balance than an Org support issue, and may not be necessary in all game designs. Of course, any rules under which Orgs are established and dissolved should always be made very clear to prospective founders, to reduce customer support problems down the line.


Several major models for player Orgs should be available, with differing kinds of game hooks

A standard protocol must exist for elimination of inactive player Orgs from the game. Failures of an Org to meet contractual obligations due to inactivity of members should be noted by sociopolitical game engine - and fines should be levied against Org accounts.
Should an Org be reduced to poverty and inactivity of all members, the Org should be removed from existence by the game engine.

Prestige should be tracked in relation to several benchmark Orgs, which may be governments

Issues: What if a benchmark Org is supplanted?
What if a player Org achieves sufficient eminence to become a benchmark Org?

Political engine should interact with any player Orgs
Player Orgs must therefore be well-defined database entities that the political engine can link to.
Political engine activity will trigger in-game events or opportunities (news, missions, etc.)

Military Rank should grant ability to send lower-rank members of the same military Org on missions, ability to define reward for said missions (within assigned bounds), access to military intel (within bounds) for choosing/planning said missions.

Need to use ‘placeholders’ of some sort - means to switch among Orgs within the Dynamic sociopolitical structure. NPC Orgs are significant in the initial condition of the sociopolitical system, but it must be possible for the system to operate without any need for specific Orgs holding specific, influential, or preeminent positions.

Cooperative Gameplay

Historically, most multiplayer computer games were based on conflict or contests, such as team games or free-for-all battles. In MMORPGs, wherein societies are modelled, opportunities for friendly and productive interaction can be as common, even more common, than opportunities to fight. While computer games may often be derided as hobbies for an isolated, asocial segment of real-world society, the Internet and MMOGs allow for a converse trend, where interactions in a game can potentially teach useful real-life social skills. Of course, the anonymity of an online character, just like an alias in a chat room, can provide an opportunity for a player to indulge a dark side of his personality through grief play. Ideas for handling the grief play problem within the MMORPG context are described below. It happens that friendly, constructive player interaction is one of the great strengths of MMOGs, and the friendships and sense of community created within the game environment can draw players back to a game repeatedly, even if the gameplay itself is somewhat lacking.

Precipitating cooperation among players for the most part is a problem of defining opportunities within the game environment. Tasks that require multiple participants for completion are an obvious route. The danger is the creation of situations like "We need a fourth for bridge" but where there is no compelling reason to continue working together after the task-of-the-moment is done. Objectives that require not just multiple participants, but a variety of capabilites, are preferred.

Consider approaches to providing player freedom in career selection

Orgs, Cooperation, and the Prestige System

Perceived power within a game environment is the ability to influence the actions of other players.
Posit a Prestige system wherein players obtain Prestige by performing tasks.
Prestige has by definition as its source, Organizations (player or NPC).
For player-run Organizations, Prestige accumulation per unit time is dependent upon the number of active participants in the Organization.
Prestige accumulates directly upon the officers of an Organization, in proportions those officers can define.
Officers of the Organization can "spend" Prestige to establish tasks that other players can perform.
An Organization which accumulates large amounts of Prestige can thus "sponsor" more tasks (activity) within the game environment.
Environment-specific ways to increase rate or quantity of Prestige accumulation might exist in a given game world.
Possession of certain places or objects might be inherently prestigious, or certain governments might have a Prestige advantage by design.
Potential feedback - the larger an Org, the more tasks it can assign, making it more visible, improving recruitment.

Gameplay Topic: Role-Playing

Consider venues within the play environment appropriate for MUD/MUSH - like interactions among players

Consider roles and methods of implementation for any needed staff activity directors: e.g. NPC Org/Corp leaders

Consider whether it is necessary to expend effort attempting to "enforce flavor" within a role-playing environment.

Conflict and Enforcement

Conflict in a MMORPG is a very sensitive subject within player communities. Described simplistically, one camp seeks to avoid anything that could be called "PKing", while another actively seeks opportunities for "PvP" play. Disputes over what is "legitimate" PvP activity, and what is PKing or worse, "grief play", are commonplace on player community discussion boards. Players from the different camps routinely have opposing views about an identical event. Of course, complicating this picture are those players who form a middle ground between the two extremes. If either extreme is catered to in game design, not only is the other extreme alienated, but the middle ground also remains unsatisfied.

This subject is relevant to game designers for two reasons: (1) For the PvP camp, nothing beats the rush of PvP excitement. If a game does not have it, these players will go somewhere else. (2) For the anti-PK camp, nothing is worse than being victimized by a grief player "abusing" PvP capability. Players that have no interest in PvP, who feel they were attacked unfairly, or players that might enjoy PvP, but who "weren't ready yet", will feel frustrated and also abandon a game. The objective is to satisfy as many potential players as possible.

Regulation of conflict within the game environment thus becomes a primary design issue. It is presently commonplace to use a game's inherent advancement hierarchy as a major component of the regulation system. This document advocates elimination of an advancement hierarchy (see above), so approaches using hierarchy for regulation must be discarded. Instead, it is recommended that processes to regulate conflict be derived from the desired in-game society. Removal of an advancement-based conflict-regulation system has the virtue of eliminating one of the most artificial constructs possible from the game environment.

The concept of regulating PvP conflict based on the game's society can be extended to creation of systems handling many forms of in-game criminal action, of which PKing (equivalent to murder) is merely the most significant. This has a profound benefit beyond potentially providing an "organic" solution to PK activity. Creation of a criminal activity system is permissive and enabling, because it introduces entirely new gameplay possibilities into the environment. It can be shown that a surprisingly large fraction of players in any game environment have a desire to "play the hero". These players wish their characters to be known as defenders of the weak, in whatever sense appropriate to the game's particular genre. Unfortunately, most game environments limit the activity of these "hero" players to fighting AI opponents. In a society-based enforcement system, this role-playing tendency from within the player community can potentially be harnessed to assist in the regulation and enforcement process. Simultaneously, such a system may allow simulation of darker aspects of society, such as theft, piracy, or smuggling, which are also potentially entertaining to some players, and are less destructive to the game environment than PKing.

In any civilized society, means are created to preserve what those in that society refer to as "law and order". The concept of "law and order" is significant when conceiving a game environment, because often the designers will have in mind a specific feel for the environment in this respect, relating also to the desired degree of freedom of action the players will possess. For some games, a "wild west" feel, where anything goes, may be appropriate. In other games, a strictly regulated society with only the most formalized combat, such as tournaments, may be desired. Often, the game designers will desire a variety of subenvironments within the game, ranging from effective anarchy to fully controlled. Broadly conceptualizing these "enforcement levels" is a good first step.

This introduces a concept of "criminality", attribution of a new characteristic to a character that has somehow performed a regulated action.

An additional consideration is the efficiency at which the game system recognizes the performance of a regulated action. Simulating a functioning society in the game world, it is unrealistic to automatically identify 100% of criminals at the moment of the perpetration of a criminal act. Instead, there should be a basic detection probability associated with each act. This detection probability should relate directly to the presence of entities in the game environment capable of detecting the act. A thief stealing an object from a player in a remote location need only be concerned with the chance that the player's character will detect the act. A thief committing the same crime in a populated area where other characters or NPCs are directly visible should be taking a vastly greater risk, especially considering proximity to possible agents of enforcement, such as police-equivalents.

The main element of an organic law enforcement mechanism is to include the players (Make the Players Do It). Once a character has been flagged by an event handler as "criminal" (to whatever degree), there must be a way for other characters to identify the criminal character. Perhaps, if becoming police-equivalent is a gameplay element such as a career choice, not every character will have access to this ability to distinguish. Note that a decision to limit the ability to recognize criminal characters can result in a more permissive environment, because not every player will be a potential agent of the enforcement system. It is recommended that for severe levels of criminality, such as those generated by what the system defines as "grief play", any player should be able to identify the offender and act against him without being in turn penalized by the enforcement system. A grief player thus becomes "outlaw" in the traditional sense of the word, and is no longer under the protection of the game's law enforcement system. In fact, in order to fully motivate participation by players in law enforcement activities, the game mechanics should reward players who act as police-equivalents against criminal characters. Bounty systems, with rewards designed to scale with the accumulated "criminality" of the offender, are a leading possibility for organic law enforcement. Note that not only should bounty-hunter characters be rewarded with in-game currency (a "paid bounty"), but they must also be rewarded with whatever is used for player motivation elsewhere in the game design, so that enforcement activity can compete in desirabilty with other tasks that players may be interested in performing. In MMORPGs common at the time of this writing, that additional reward would typically be "experience points". This document advocates a Prestige reward. See below for another concept that can be integrated with organic enforcement schemes.

The ultimate goal of an in-game law enforcement system must be to, through systems of reward and punishment, keep player behaviors within acceptable limits, or cause the potential grief players either to be banned or to leave the game of their own volition (usually due to discovery that the grief play they desire is not "fun" within that game's structure). The only caveat is to be certain that sanctioned activities that still carry criminality taint within the game context are not so heavily penalized that the gameplay potential of having criminality within the system is stifled. Of course, extreme examples of grief play, especially social abuse of other players, should result in expulsion from the game environment and banning of the grief player's account. In this regard, a grief player's status as a paying customer is overridden by the weight of the paying customers he may potentially drive from the game.

Catalyzing Player versus Player Conflict

Turning away from issues of enforcement, means of creating potential conflict must be briefly addressed. Usually, the players can be expected to take care of this themselves in several ways. If player organizations are part of the game environment, without a doubt some of these Orgs will quickly come into conflict either due to having opposing goals or because of some clash of egos among players, resulting in organic PvP action. To enrich the game experience, it is best if these conflicts can occur within the context of a dynamic background to which the player Orgs are linked. In a vibrant dynamic environment, Orgs that might not conflict due to personality may find themselves driven to conflict due to causes rooted in the environment itself. This is the ideal state.

Valuable to have potential for realistic piracy
Corollary - nonlethal combat must be possible
Pirates should have option of simply damaging a target rather than being constrained to destroying it outright
Identified pirates should be targeted by either bounty hunters or military missions (should be directly tied to location/severity of piracy incidents or magnitude of reported success by the pirate or losses by legitimate players)
Consider implementation of a criminal/legal mechanism reminiscent of that existing to regulate PK activity in UO.

PvP (sanctioned PK) activity must be regulated.
Best option: clear and focused use of military Orgs in game environment - players interested in PvP will be shunted toward the military both within and outside (via advertising propaganda) the game. Military activity will attempt to coordinate missions for routine PvP action.
Civilian targets will be avoided as a matter of course. War crimes, friendly fire, etc. will be dealt with internally to the military Org structure.
Players who leave the military but attempt to follow a military pattern outside the structure must be shunted back into a (A) military (B) pirate (C) bounty hunter role or essentially eliminated from the game by its mechanics directly.

Lethality and Nonlethality

Per Craig Huber's inspiring essays on next-generation MMORPGs - death in most MMORPGs is really a noninteresting, nonthreatening event. There is a need for this to change, for two related reasons (1) Character Death is a consequence or threat potentially useful for motivating players toward some behaviors and away from others (2) The possibility of Character Death is a source of drama and excitement in any in-game conflict. Huber's recommendation is that a game would be enhanced if designed such that a character can actually be eliminated from the game, instead of being respawned at some safe location or in some manner that requires the player to complete some tasks before the character can be reconstituted. This weak form of death seen universally in MMORPGs is caused by game company fears that eliminating characters will drive away the players. Assuming "death" remains as commonplace as it is in most games, it's quite likely that prevailing attitude on the part of the companies is correct.

What is required is a sort of paradigm shift in combat and character death. While it must be possible to kill characters, it should also be possible for players to mitigate the lethality of a combat. "Winning and losing" should still be quite clear, yet not every combat should result in death, as is now common. The form this mitigation must take within a game is dependent upon genre. Unconsciousness might be an appropriate effect if the game is a medieval first-person-view adventuring game. Disabling a vessel, or vehicle damage, could be appropriate in a space or driving game. In addition, a losing party in a combat ought to have some sort of opportunity to surrender, and that "surrender" function should be lodged in game mechanics. Choosing to fight to the death of one's character becomes a conscious choice, and likely an uncommon one. The option is available for a player to remove her character from the game environment in a blaze of glory - and for the act to be meaningful, in that it would be vastly more significant than being knocked out, disabled, or forced to surrender. Character Death is also an opportunity for a player to construct a completely new persona, if the old identity has become tiresome. Perhaps the character creation mechanism would allow a continuing player's new character to inherit something from the prior character one, using the idiom of a "family relationship" - if the player so desires. That would have the benefit of allowing the replacement character to legitimately retain the same group of friends and contacts the player would naturally keep within the game community.

The concept of surrender, or for rendering a losing opponent temporarily helpless in some fashion, has a number of gameplay benefits. It has the potential to mitigate grief play to some degree, because most grief play goes under other names. If the game has the ability, for example, to allow banditry or piracy without actually killing or destroying an opponent, players who persist in this activity remain villains but are not necessarily grief players any longer. Bandits who slay their opponents, rather than simply defeat them and steal from them, would be treated in vastly harsher fashion by the game's enforcement mechanisms. Bandits who steal "honorably", to misuse a term as it is commonly being misused in game communities, could be subject to less-harsh penalties, and would in fact be a useful source of hazard and excitement within the game environment.


Nonlethal combat or attack methods should be available in some or all forms of PvP combat available within the game - this gives the attacker an ability to avoid killing, just as a defender may have an ability to surrender to avoid dying

Military or police-equivalents might have options to check for and confiscate contraband and/or levy a fine - esp. if smuggling or piracy exists withint he context of the environment

There should be severe repercussions for destroying a surrendered opponent - including court-martial discharge if military, targeting for bounty hunts or military action if civilian.

Need for ability for a player to choose to allow her character to end
What about obituaries in the news system? Rewards players who permit a high-prestige character to end

Gameplay Topic: Military vs Civilian Role Differentiation

In many MMORPGs, there is a concept of character "class", or profession. Yet there is usually no real differentiation between military and civilian roles, other than what a given character is capable of. Artificial systems to regulate conflict are usually imposed, to prevent those characters very capable of battle from engaging characters inadequately prepared. Usually the game's advancement system is also employed in this artifical approach, to varying effect.

It is proposed to very simply and overtly include both military and civilian roles in the design of the game, with clear means to migrate legitimately between them. This should ameliorate many in-game social problems resulting from players desiring different levels of conflict, while still permitting considerable player freedom of choice in play style within the context of the game's environment. Intentionally modeling a functional society within the game environment allows for corrective measures without the artificiality.

Simply by declaring some players "civilian", and others "military", two different conduct paradigms are available to regulate activities: Players interested in PvP will gravitate (or be specifically directed) toward military careers or bounty hunting. Civilian characters who engage in PK activity without some in-game authorization (such as sanctioned bounty hunting) will become criminals (preferably over time due to negative Prestige, rather than instant flag-setting), and will be targeted by an in-game mission/job engine for bounty-hunting or military action as appropriate. Military players who disobey orders, exceed their mandate to engage an enemy, or prey on civilian players or NPCs can be dealt with harshly by the military structure they are a part of, or in a fashion similar to the way civilian criminals are dealt with.

For those players choosing military careers or tours-of-duty, gameplay should be clearly different from those players considered civilian. This is an intentional part of modeling military life. Freedom of action may be more limited, though that is more an aspect of the environment rather than necessary on the basis of this role differentiation concept: knights-errant, Space Patrol, or what-have-you, can easily be posited to allow more free approaches to locating dangers and fighting them. Military play should also be focused on defending a civilian populace from enemies or raiders. Bored military players are likely to become a social problem - exactly the case in the real world, and an indication of a good conceptual simulation. Key is arranging activity so that the military players are challenged by fighting each other, not the civilians.

Efforts must be made to limit possibility of civilians being targeted by military action (collateral damage). Communication should be the first, best approach. War zones where military activity is happening should be clearly delineated - civilians must be alerted by the game when and if a conflict area is being entered. Militaries will require clear definition of rules of engagement - and mechanisms to adjudicate any variations. There may be rare occasions where a government makes a decision to allow a military mission targetting civilians. Similarly, there will be situations where civilians come into harm's way through no fault of the military involved.

Enough military activity is required to keep military players busy - repetitive patrols are totally inadequate. Peace between all governments much of the time is not an option. The principle of the Dynamic Environment is useful to keep things moving and changing, keeping the military players interested in what is going on. In addition, another concept can help, if it is appropriate to the game environment's "look and feel": Foreign Service. If a government finds itself no longer engaged in any combat, it may rent its military out to other combatant governments. Military players might choose a FS tour of duty as mercenaries (even selecting which side), be assigned one by their government (in which case the government is probably being rewarded, not the players, by the government receiving the troops) or perhaps the player may choose a cushy (but duller) spate of patrols and escorts back in whatever passes for "home territory". This concept simply allows for continuing military activity to be available even when the dynamic environment causes one or more in-game governments to find themselves at peace.

Including more than one gameplay paradigm: Management-level Play

The "Make the Players Do It" philosophy can be carried much further than simply allowing players to build story and participate in creating infrastructure. Even beyond the level of player-run organizations, players can be used to manage infrastructure. This can drastically reduce what AI-type algorithms are needed within a game.

It is known that a significant population of computer game players enjoy building and management simulations. SimCity gave birth to an entire game genre. The potential exists to use management/simulation game players, employing a separate set of interfaces, to influence and operate the environmental entities in a game world. This would in effect constitute a management "toy" in the same persistent universe as the main online game. Management players could even substitute for AI in a game's sociopolitical system, making decisions about policy and resource allocation for major entities like governments. Players are often good at coming up with novel, interesting approaches. Humans are certainly more creative than even an AI written by the game designers. This would tend to boost the dynamism in a game environment, as well as probably making the governments or other entities seem to the players like they are run by living, breathing people - because they would be.

Management players might be charged on a different pay scale, and should possess a completely different class of game account. It must be realized that in this situation, players are paying up-front for the privilege of helping operate the environment. These players are not GMs, yet their role potentially reduces the administrative demand of running the game. They are a revenue source, not a liability.

Note: it should be totally within the realm of possibility that the dynamism in the game results in the elimination/destruction of a managed entity (a company goes bankrupt, an outpost is destroyed, etc.). This should be made clear to management players up-front, and a new entity offered to a continuing player if the entity that player was managing is eliminated. This is in analogy to the concept of literal character death in a mature MMORPG.

Note also that it will be important to make management players blind to the identities of individual player-entities in the "main" MMORPG to prevent abuses. At best, all interaction with player-entities at the non-management level should be totally abstract. An example of this would be management players having the ability to set up jobs for first-person players, but lacking any control over who in the first-person universe might accept them.

Design Topics: Miscellaneous

Consider methods of adjusting in-game timescales to accept player interest-and-use timescales - keep perspective on player activity patterns that contrast with realism. Student players have more online time than professionals. Is the average player online for two hours, or six at a time? Both modes must be accounted for in terms of achievable objectives as well as access to infrastructure (such as places safe to log out).
Consider whether a role exists for non-player action in the background (passive/maintenance AI)
Consider whether a role exists for non-player action in the foreground (active AI)
Prepare rationalizations and corrections for dealing with disjunctions in the game environment caused by real-world events - such as server outages. -- this may be an extension of the metagame discussion

Gameplay Topic: Physical Travel

Criterion: Reasonable, playable travel times

Email the author of this essay: