Life in the Virtual World

Players are real people who enter virtual worlds through the instrument of their characters. Chapter 3, “Players,” looked at players and Chapter 4, “World Design,” looked at the world. This chapter looks at where they meet: characters.

Advancement

Players advance in intangible ways, through the knowledge they gain and the friends they make. Characters also can develop intangibly to some degree, in that they exist partially in the minds of the players who encounter them. However, characters are primarily a tangible phenomenon: They are defined by code and data. The virtual world determines what they can and cannot do and how mobiles react to them. When a character “improves,” the improvement is measured in terms of tangible factors rather than intangible ones.

Characters don’t have to advance, of course; for non-game worlds, the aim is often for players to gain or hone skills, therefore the success or otherwise of a character is entirely dependent on what players learn and how they apply this knowledge. Still, though, it’s often the case that players will respond better if there is a reward in the form of some tangible improvement to their character, even if it’s of only metaphorical significance.

For those that do offer advancement, there are basically two approaches: single or multiple. In a single advancement system, one process completely dominates; in a multiple advancement system, characters can improve along several dimensions. At first glance, it looks like no contest: Why would you want single advancement when multiple advancement offers so much choice?

There are two reasons why. Firstly, single advancement is focused and unambiguous. Players know what they have to do, and can therefore set about doing it. In a multiple advancement system, players might have to make choices before being fully informed; they may therefore make mistakes. The second reason is that multiple advancement suggests multiple content, which is more expensive to produce and harder to balance.

As an example, consider a hypothetical virtual world set in Three Musketeers ¹ France. In a multiple advancement system, characters can acquire “military points,” “romance points,” “politics points,” and “fame points.” Thus, they can concentrate on acquiring points of a particular kind to advance their careers.

If more people want to pursue one career than there is content to support, new content must be added for it (and the content that already exists for other careers will be relatively underused). Players who start their characters off collecting military points only to find that the romance approach suits them better will feel that their time has been wasted. A single advancement system, where characters collect “social points” that are an abstraction of the other four, would allow characters to switch focus without suffering, and if content of one type was running low they could overflow into another until the first recovered.

Multiple advancement wins out when advancement involves different mechanisms. If military rank depended on campaign experience and political rank depended on electoral influence, the two could not be reconciled so easily.

Virtual worlds measure advancement in terms of advantageous changes to tangible data. In considering advancement, it therefore makes sense to begin by looking at what can be advanced.

Attributes

Attributes are the fundamental properties that define a character. Because they are tangible, they should properly only rate physical qualities, such as the character’s strength and manual dexterity. However, under the influence of table-top role-playing games they often include non-physical concepts, such as intelligence and wisdom. This just about makes sense to a hard role-player, who can work with “My character is stupid,” but it has its limits: “My character is a genius” would for most players entail pretending to be cleverer than they really are, which is not quite so simple. For most virtual worlds, though, these “mental attributes” are merely convenient labels: Intelligence is used primarily as a fiction to determine how much magic a character can use, and wisdom serves the same function for priestly types.

Attributes also are meant to be relatively fixed: If you have a poor constitution, tough luck, you just have to live with it. Aging effects could perhaps make a difference (older characters might gain wisdom, but lose strength), but on the whole attributes aren’t supposed to change much, thereby preserving basic differences between characters. There has been increasing slippage in this over the years, however, with the old notion that attributes should indicate intrinsic abilities increasingly falling out of favor. Instead, attributes can be improved on by action or training, which makes them more like skills (discussed shortly). There are several reasons for this, having to do with fairness, motivation, and choice.

The result is that after a while character attributes ¹ become fairly homogenized: In the real world, people can be any combination of good-looking/ugly and strong/weak, but in virtual worlds they usually have to start off more one than the other (for balance). They won’t end up being high in both. It may take considerably more effort to raise an originally low attribute to a high value than it does to raise an originally high attribute to it, but that doesn’t present a total barrier. Normally, one stat is raised relentlessly as far as possible first, with the lower ones being neglected until it’s their turn; this leads to a world packed with puny geniuses and muscle-bound simpletons.

The problem with attributes is that they do make sense in real terms, but this isn’t a sense that many players are willing to accept. Some people are simply never, ever going to be as strong or as fast or as tall or as fit or as good-looking as others no matter what they do, but part of the attraction of virtual worlds is that these physical limitations don’t apply. In design terms, fixed or ever-rising attributes are also undesirable: If you decide after some time that you want your character to become a ballet dancer but you built her like an ox so she could become a boxer, well, ballet dancing is out of the question.

For this reason, the trend is to treat attributes as little more than symbols during character creation, so that players have to think about their characters as they individualize them. This gives characters direction, but only in the minds of their creators; in practice, the relative differences between a character’s attributes are not usually all that important in the great scheme of things, especially if values are regularly boosted as a reward for achievement ².

The idea of attributes as important, player-defining properties is probably here to stay. However, whether they remain relatively fixed (as they do for most other objects) or they rise ³ over time (as do skills) remains to be seen.

Levels

In the development of virtual worlds, levels came first (as they also did in table-top gaming). Levels are the classic advancement system—easy to implement and to manage—and are based on the concept of goals and rewards. Characters undertake actions, for which they receive a payment in experience points; because the notion of experience is general enough to apply to pretty well any activity, this approach has wide applicability. When a character’s experience points reach a certain threshold, the player rises in (ability) level. This will typically involve things such as attribute boosts, skill boosts, resistance boosts, greater choices, and higher status (among NPCs and, often, PCs).

Looking at this analytically, boosting is implemented at times dictated by a step function. Why not use a more continuous approach? Instead of levels, raw experience points could be used. By using a finer grain for things such as attributes and NPC attitudes, a character who is one experience point below the threshold for level X wouldn’t have exactly the same stats as a character exactly on the threshold for level X-1. In pencil-and-paper games, where humans do (or used to do) the bookkeeping, this is reasonable—keeping separate track of these things is just too tedious. Computers, though, excel at tedious, so why the need for levels?

The answer is that levels provide goals. They give players something to shoot for, so that their actions have more purpose. They give players a sense that their achievements do have effects. Levels use a step function because players like to feel they have taken a step.

The time taken to reach a new goal ¹ must be appropriate, of course. If characters advance too quickly, levels effectively merge into a stream and offer no more bang than experience points. If levels are too far apart, players become disenchanted that their efforts rarely seem to get them anywhere. Note that although the experience points separating two fixed levels are invariant (at least within character classes), the time taken to acquire the necessary points will depend on two things—what the character does, and luck. Thus, characters do not level ² with monotonous regularity, but with some unpredictability. This leads to what is known as a variable ratio reinforcement schedule, which behavioral psychologists have shown to be the most effective there is for maintaining a high, constant rate of play ³—Las Vegas slot machines are designed this way to keep people pumping them full of money. The morality of using such an addictive system for virtual worlds is discussed in Chapter 8, “Coda: Ethical Considerations.”

Another reason for grouping together several rewards is that this enhances the overall experience; the bulk adds value. Damming a stream and then releasing it has more of an impact than would letting it run freely, whether the stream is of water or of presents.

The relationship between levels and experience points varies, depending on the virtual world. However, it’s almost always exponential rather than linear. MUD1 used a doubling scheme, whereby the points needed to rise from level X to level X +1 totaled the same as the points needed to get from level zero to level X ⁴. Most modern virtual worlds start off approximating a doubling regime, but gradually tend toward a more linear scale. Compare the examples ⁵ from Asheron’s Call shown in Table 5.1.

Table 5.1 Asheron’s Call Level System

A slowing curve enables more levels to be packed into the same range of points, which increases the rate of reinforcement for players of experienced characters. Even so, some virtual worlds allow characters to have so many points that they record scores using a floating-point representation. This kind of level inflation can get out of hand (in worlds without permanent character death, anyway), so there’s often a maximum level beyond which characters cannot proceed. Some virtual worlds allow remorting at this level, which means a character gets to keep its abilities but must start back at level zero as a different (sometimes more powerful) class ⁶. Other approaches are relative levels and ranking systems (“you are 386 th out of 4,034 characters”).

One of the problems with levels (for VWs which make use of them) is that content has to be available for characters of whatever experience. If one character needs a million points to go up a level, they’ll be doing different things to the other character who only needs a thousand ⁷. This means that characters which are of markedly different levels can rarely play together—the lower character will find high-end content too difficult and the higher character will find low-end content insignificant.

Another issue concerns where points come from. Some of the most intense experiences of players result in low overall gains—or even losses—of their characters’ points. If some monster beats you up, but you get healed then go back again and again until you beat the sucker, your character may be worse for it but you, as a player, may be better. Likewise, running around mindlessly collecting easy quest points is going to make your character improve, but it’ll do nothing but bore you as a person. Ideally, points should be awarded for things you find exciting, rather than as some kind of compensation for being fed up.

I’ve been describing levels here as a single-advancement abstraction of other progression measures. It’s possible to formalize this, with level changes being determined using a formula that accounts for a range of multiple-advancement schemes. Thus, a player may suspect from experience that they’re close to going up a level, but don’t know how much they need to improve their swimming skill or their dart-throwing skill to do so. This uncertainty makes a variable ratio reinforcement schedule even more powerful, and the morality of using it consequently even less clear.

Hybrid schemes are also being developed. Neocron ⁸ for example, has typed experience points: When your character is awarded experience points for doing something, the points go to a pool associated with one of five stats. When the pool fills, you go up a level in that stat only, and gain skill points that can be spent on skills related to that stat only. It’s a cross between a level system and a class system.

Designers are thus increasingly becoming aware that the benefits of levels can be used in non-traditional areas. In particular, applying a level system to individual skills can be very fruitful, assuming some provision is made for achiever-type characters to compare themselves against others (and even that may prove unnecessary). Skill systems are present even in many single-advancement, experience-based virtual worlds, and could eventually take over from them as the predominant measure of a character.

However, skill systems are not without their problems.

Skills

In virtual worlds, skills are tangible representations of the abilities that characters have (supposedly) learned.

In real life, a skill is a related series of actions that a person has internalized to the extent that they can apply it “without thinking;” skills are either innate or learned through practice. In virtual worlds, the notion of what a skill is remains the same, but skill acquisition is often only tenuously linked to “practice makes perfect.” In particular, some virtual worlds allow characters to improve skills when they go up a level, irrespective of whether the skills they choose to improve are ones they’ve actually used.

Why do virtual worlds need to make skills tangible? Actually, they don’t: Character skills are very useful as a means for players to individualize their characters, and they provide a good delivery mechanism for achievement, but they aren’t strictly necessary. The most important skills (like it or not) are those of the player, not of the character, and the virtual world can’t hope to rate these ¹.

Skills are interpreted by the virtual world in a number of ways, depending on the nature of the skill, of any tools required, and of the virtual world. The main four are

• Binary skills (feats ²). Either you can swim or you can’t.

• Discrete skills (skill levels). People can swim to an ability described by one of a fixed number (greater than two) of predefined levels.

• Continuous skills. People can swim with widely varying degrees of ability, from not at all to completely competent.

• Binary-continuous skills. Either you can’t swim, or you can swim with widely varying degrees of ability.

The idea behind a binary skill is simple command guarding. If you have the skill, you get to use the command, and if you don’t, then you don’t. Discrete skills take this a step further, with increasing levels opening up new commands (swimming underwater, swimming in stronger currents).

Continuous skills work slightly different in that everyone has a shot at performing any command. Sometimes, the effects are also continuous (the higher your swimming skill, the faster you can swim), but often it’s used as a form of binary skill based on the difficulty of the task at hand. For example, if swimming across a moat is rated at difficulty 30, then someone with a swimming skill of 30 or more will always succeed; someone with a swimming skill of 25 will fail some of the time, but not as often as would someone with a swimming skill of 10 ³. Binary-continuous skills work the same way, but with the proviso that if your skill rating in the area is zero then you can’t even attempt it (this is in addition to any other applicable preconditions—for example, armless characters don’t get to swim no matter how much they practice).

Skills can be active or passive. Passive skills are switched on the entire time, whereas active ones must be called into play explicitly. For example, if a character has a “deflect arrows” feat then that would be passive—the character automatically attempts to deflect any arrows fired at them. A feat such as “berserk fury,” on the other hand, might be active: Players have to decide when to sacrifice defense for offense, it won’t happen on its own. Most skills are passive, but only come into use as the result of an active decision: Your swimming skill is effectively dormant until you enter a body of water, but while there it’ll be taken into account passively. Because putting active and passive skills in the same list can be confusing for players, there is a growing tendency to refer to active skills as powers rather than skills.

Skill Organizations

The “skill levels” approach is an example of a linear skill progression. To acquire a specialized skill, you have to learn its parent (or prerequisite) skill first. Thus, you can’t learn, say, “diving,” unless you already know “swimming.”

A skill hierarchy takes this idea one step further and allows several specialized skills to be learned following the acquisition of a general skill. For example, having learned a “single-handed weapon” skill, a character may then learn “sword,” “axe,” or “club,” and from there further specialize into particular weapons (longsword, shortsword, rapier, cutlass, and so on).

A skill network is like a skill hierarchy except that it allows for skills to have two or more other skills as preconditions. You can’t learn the skill of “horse archery” unless you already have “horse riding” and “archery” skills; you can’t learn “espionage” unless you have at least three from “stealth,” “disguise,” “bluff,” and “hand-to-hand combat.”

Looking at skills this way, it’s easy to see how attributes came to be regarded as the same thing. “You can’t be a blacksmith unless you’re strong” is just like “You can’t be a pearl diver unless you can swim.” A concept such as “strength” becomes the concept of “being able to use strength;” although this is ultimately dependent on a character’s physical prowess, it nevertheless offers enough leeway to allow for increasing the “attribute” as a device for character improvement.

Doing this has certain implications for skill systems, though. Ideally, all skills should be implemented using the same basic mechanism, so as not to distract the players. If two or more skill systems run in parallel, they must be disambiguated somehow (this is why feats get a separate name). When attributes are handled just like skills, this means that skills and attributes should “look” the same. In other words, if attributes are on an integer scale then they should all be on the same scale and it should be shared by skills.

In practice, this actually works quite well. There are some awkward intrinsic properties of characters that don’t fit naturally into the scheme (the number of genders is fixed, for example) but they can usually be shoehorned in without too much bother. This isn’t to say that there aren’t better ways to do it, of course; however, it does mean there isn’t too much incentive at the moment to do it some other way. As usual, as long as designers understand why things are the way they are, they can make informed judgments as to whether to keep them that way.

It’s worth considering whether skills should be applicable to objects other than player characters. They can quite obviously be extended to NPCs and even some monsters, but what about objects in general? Could not a shield’s ability to deflect arrows be implemented using the same skill that a character might use to deflect them? Or more frequently, shouldn’t a fireball wand simply cast the spell? Skill transference like this is particularly useful for virtual worlds that use magic, where a wand could be given the same spell-casting skill as a player or a player given the same petrifying gaze skills as a basilisk. Could a sword improve, through use, to become better? Could it be trained (that is, enchanted) to learn new skills? If object qualities are implemented using the same system as player character qualities, at least these are options (albeit not necessarily ones every designer will want to pursue!).

Skill Sets

The biggest problem with skills lies in deciding which ones your virtual world should have. How many should there be in total? Is 100 skills too many? Too few? From whose point of view—newbies or oldbies?

We should begin by determining what is meant by a “skill.” Most virtual worlds would admit derivatives of crafting—making or repairing things in the virtual world—as qualifying as skills. Some would include weapon skills, and others would also add spells. The disparity is in part due to the different mechanisms by which acquisition works: If you get crafting skills by collecting “skill points” for going up levels but you get weapons skills by buying “training,” the two are distinct. It would be confusing to call them both “skills,” so commonly some other term is used instead (for example, “weapon specializations”).

Ideally, all forms of skills would use the same mechanism. Why should the system by which you learn to make a barrel be any different from the one by which you learn to swing an axe, to brew a potion, or to let rip a fireball? If they are fundamentally different even at abstract levels, that’s fair enough; I wouldn’t necessarily advocate implementing equipment as skills, for example. However, if they’re the same but are implemented differently, perhaps the reasons for that different implementation should be examined.

Spells, for example, are often linked to a character’s level. When you reach level 43, suddenly you gain the ability to cast sidewinder fireballs in addition to regular ones. The spells you could cast at level 42 you can still cast, but with slightly more efficacy. If this approach is good enough for spells, why not use it for crafts? When you reach level 43, suddenly you gain the ability to make mithril horseshoes in addition to regular ones. Alternatively, if you’re using a “skill points” system for gaining new craft and/or weapon skills, why not apply it to spells?

Using the same skill system for crafting, weapon use, spells, and everything else (acrobatics, horse-riding, speaking Elvish) doesn’t mean they can’t be radically different in effect, of course. However, it also doesn’t mean that the problems of deciding what skills to have go away.

Typically, however many different crafting skills, weapon skills, spells, and general-purpose skills a virtual world offers, a handful will be so essential that everyone has them, most will be secondary skills that people have because they had to choose something and another handful will be utterly useless. In a Wild West world, for example, only people who actively like building unusual characters are going to pass on the horse-riding skill.

If a skill is considered by all players to be essential, all characters should be given it anyway. Characters don’t usually have a “walk” skill, because it’s a skill everyone can be assumed to have. As a rule of thumb, if any one skill is adopted by more than 90% of characters ¹, they all ought to get it. Similarly, if a skill is unused it should also be given to everyone (or derived from attributes)—it costs nothing to do so, and it means the remaining skills should be of roughly equal value to players. Of course, it may be that having done all this you find that a new handful of must-have skills emerges. If this happens, either repeat the exercise or split the skills into disjoint sub-skills.

It’s not too difficult to hardwire a universal skill once it’s been identified (assuming you do so before the virtual world leaves beta test). It’s quite tricky to decide what should be skills in the first place, though. Should “horse training” be a skill? Is it different from “dog training,” or are they both examples of “animal training?” Should there be a skill for tree-climbing, or are so few people going to want to do this that it’ll be a waste of time implementing it? Should there be some weighting scheme so that unpopular skills cost less to learn than popular ones ²? Should some skills be more difficult than others anyway—isn’t learning a foreign language comparatively harder than learning to ride a bicycle?

There are, unfortunately, no clear-cut answers. My advice when deciding what skill set to have is to consider what message it sends to players. I’ve seen designs for virtual worlds that have a weapons skill tree five levels deep, yet which had a single “mining” skill that worked as well for iron as it did for diamonds. The message they were sending was, “This virtual world is all about combat.” Similarly, I’ve seen designs that covered the minutiae of everyday life, with skills covering everything from corn-planting to calligraphy to bee-keeping to savoir-faire ³. The message here was, “Nothing is all that important.”

If you want crafting to be a viable career, the skill tree for crafting should be deep enough for people to take it seriously. If you want people to travel a lot, include several skills that make traveling easier. If you think players will want to play bardic characters, include bardic skills. If you don’t want to attract sneaky-thievy characters, don’t have sneaky-thievy skills. Rate each skill for its attractiveness to various player types; if any types are over- or under-represented, adjust accordingly. If you fear spreadsheet abusers, make key skills mutually exclusive (but be aware that this will create a de facto character class system).

Although the number of skills of a particular kind available to characters is one way to send players a message, it’s not alone. In general, skills should be of roughly the same level of abstraction as one other. If in one area they are represented in greater detail than in others, this also can point to an increased sense of importance. Were the “swordplay” skill to be described in terms of how much it increases the damage done, critical hit percentages, weapon recovery time, parrying capability, and resistance to being disarmed, this would give it more significance than a “lumberjack” skill defined solely in terms of how quickly you can fell a tree. Crediting skills with a false sense of importance this way is a particular example of selective depth, a trap that often ensnares the designers of virtual worlds.

As with objects, skills don’t have to have a tangible purpose. If characters need an “interior decoration” skill to change the wallpaper in their houses, then if they want to change the way their house looks they can either acquire this skill themselves or go to someone who has already acquired it. Either way, the result is a matter of aesthetics: The virtual world doesn’t necessarily ascribe any meaning to how the inside of your house looks. Of course, also as with objects, it doesn’t necessarily not ascribe any meaning to it either. Unless you want to make some kind of “we have some skills just for the fun of it” point, if you can find a way to give tangible value to skills, you should do so. Having your house redecorated to a higher standard could win you the envy of NPCs, say, or reduce the chances of its being infested with termites.

Skills that can be used for more than one thing are generally good to have but difficult to contrive. If you can use an axe in combat, does that qualify you to use one as a lumberjack? Or is that a different kind of axe skill? If they’re different, how many people are going to bother with a lumberjack axe skill? If they’re the same, characters will become lumberjacks so they can increase their axe skills while making money. Multipurpose skills have more justification for existing than do single-purpose skills, but they also can have annoying consequences.

Caps

Levels and (in most implementations) skills operate on a linear scale. A character starts at 0 and improves incrementally. A level 22 necromancer isn’t as powerful as a level 55 necromancer, but would be roughly on a par with a level 22 monk or assassin or whatever and much better than a level 11 necromancer ¹. So what’s the highest value that a level or skill can have?

If you have no limit, characters will improve relentlessly, getting more and more powerful until newbies are little more than insects by comparison. Once-daunting content will appear trivial; players will demand more to satisfy their appetites.

For this reason, many virtual worlds have caps—absolute limits on how high a level or skill can go. For level caps, this stops characters from getting too far ahead of what the virtual world can offer as content; however, it can lead to frustration as there’s nothing to do except wait for other people to catch up in sufficient numbers that new content will be added. For skill caps, players will max out their characters in all the important skills so they still end up ultra-powerful.

The only solution that is truly guaranteed to stop this from happening is permanent death (PD), but this is often taboo—especially in commercial virtual worlds. Some designers have mooted the idea of having PD as the price for cap-busting (if you want to go off the scale you can do, but your character can then die permanently), a solution that has been implemented many times already in non-commercial worlds. The uproar from paying players who want the skills but don’t want their characters to die would likely be considerable, however. Players offered Achilles’ choice between a short life of glory or a long life of obscurity will want a long life of glory.

Without PD, either characters are going to become all-powerful or (achiever) players are going to become frustrated. That really is the bottom line. It’s possible to relieve these symptoms without resorting to PD, though. One way is to have an advancement curve that makes it increasingly difficult to progress to new levels, but not actually impossible. As long as the powers at level N are tangibly better than those at level N-1, players will want to have them. Eventually, though, an effective cap will emerge beyond which few characters will ever rise. Taking the caps off might mean someone can increase their archery skill to 134 instead of being stuck at 100, but they’re never going to reach 1,000. It’s a bit like world records for athletics: They’re always falling, but there has to be a limit. No one is going to run the 100 meters in 7 seconds unless the authorities change the rules currently disallowing the use of motor bikes. Steep advancement curves like this can work, but they really only ever delay the inevitable: Sooner or later, players decide that the treadmill is simply no longer fun.

An approach pioneered by Ultima Online was to make skills atrophy. Characters can learn several skills, but the more they learn then the less attention they give to other skills, which lose their values. This allows for characters to excel in only a few skills at once, while retaining some scope for changing what those skills are. Players can remain powerful, but can use different skills as ways to express their power. It sounds good, but unfortunately players don’t like it: If they learn a skill, they believe that they should always have that skill. They do have a point: It’s not like an English-speaker who learns French will forget the French if they then learn Spanish; indeed, a knowledge of French could help you learn Spanish. It’s true that if you become unpracticed at something, then it’ll take you a short while to get back up to speed, but this will be nowhere near as long as the first time you learned it. Being rusty at something is different than completely forgetting all about it.

The first-century Greek philosopher Plutarch said, “The mind is not a vessel to be filled but a fire to be kindled.” ² Atrophy systems embody the “vessel to be filled” approach: You can only learn so much; if you’re all learned out, you have to unlearn something to make room for new stuff. Although UO players weren’t exactly happy about skill atrophy as a principle, what really annoyed them was the fact that they couldn’t state which skill they wanted to atrophy next. Star Wars Galaxies uses basically the same idea, except that atrophy doesn’t occur automatically: If you want to recover points from one skill to spend on a different one, you have to do it explicitly. Skills you already know cost less to advance in than others, so the temptation is to specialize rather than take the jack-of-all-trades route.

“Vessel to be filled” approaches cap the totality of skills a character can have while leaving the distribution of individual skills relatively free. A more “fire to be kindled” way of doing it is to cap earlier in the process—that is, limit the amount of time that players can (or are willing to) spend in a virtual world (but not so much that they don’t feel it’s worth their investment). Ultima Online also pioneered this method ³, although not until 2000. The idea behind UO’s power hour is that advancement happens faster during a character’s first hour of play each day, but drops off thereafter. Interestingly, the advancement rate to which it falls is the same as it was before power hour was introduced, but players are reluctant to rack up points in it because they’re “paid” less for their efforts than in power hour. Consequently, whereas the average time players spent in UO before power hour was introduced was three to four hours a day, afterward it dropped to around one hour a day. The result was that although characters advanced faster per game hour, they advanced slower per real hour and therefore took much longer in real terms to rise than they had done previously.

This approach worked for UO, but it can only be taken so far: If a character that was played four times as much as another character could only ever be 110% as good as it, for example, that would certainly lead to the loss of many long-term players. It has other disadvantages, too, in particular it discourages players from adventuring in groups and it isn’t easy to make remotely realistic within the virtual world’s fiction.

Skill Improvement

How do characters improve their skills? To preserve the notion of what a skill is, there ought to be some relationship between what you do and how your skills improve; but what?

If skills are rated using a points system (as most are), the question becomes one of deciding when to increment a skill by a point. There are basically two ways to do it:

• Random chance. Each time a skill-enhancement opportunity arises, there is a small chance of causing a skill point increment. This chance may increase over time.

• Cumulative chance. Each time a skill-enhancement opportunity arises, a random number of pebbles is thrown into a bin. When the bin is full, the skill point is incremented.

The advantage of the random approach is that you can gain a second skill point shortly after having gained a first, if your luck is in. The disadvantage is that you may never gain a skill point while your luck is out, and you will complain loudly about this. The cumulative chance approach does have some randomness, so you can’t ever be sure when you’ll gain a skill point (unless you can see the numbers); however, you can be absolutely certain that it won’t be immediately after having gained one.

Where skill nets exist, it may be possible to arrange for some skills to support others. It’s not a precondition of veterinary science that you know human anatomy; however, if you do know human anatomy, then you could expect to pick up veterinary skills a little faster. This can be managed by increasing the random chance of success or by preloading the bin with pebbles so that fewer new ones are required to fill it ¹.

It’s in deciding what qualifies as a “skill enhancement opportunity” that things start to get interesting. There are several ways to do it, none of which are entirely satisfactory.

The traditional method is improve from success. Each task requiring skill use has a difficulty rating; when a character attempts the task, his skill rating is compared against the difficulty rating to see if he succeeds. If he does, it counts as a skill enhancement opportunity; otherwise, well, he should maybe try something less difficult next time.

A variant of this is improve from failure. The theory here is that people learn from their mistakes, not from their successes, therefore characters should get a skill enhancement opportunity for coming close to success but not actually succeeding. If they do succeed, well, they’re happy anyway because they performed the task; this way, they’re rewarded for nearly succeeding too. If they try something so hard that they don’t even come close, well, maybe they should have attempted something easier. This is my own favorite method, by the way, but it’s not for everyone. It can produce inappropriate messages, too: You don’t really want to know that your character’s shield-use skill has gone up if the blow it only just failed to stop was fatal.

Many virtual worlds use improve from leveling. Although some increase in a skill may occur through use, the idea here is that experience points are a sufficiently good reflection of player activity that skill acquisition can be linked directly to character level. When you go up a level, you (may) get skill points to spend on improving yourself (also known by its fiction name, training). This approach has three advantages: Players look forward to going up levels more; it’s much easier to balance for skills that are used with wildly different frequencies (fighters swing swords more often than tailors make robes); characters don’t have to waste their time repeatedly shooting arrows at static targets, trying to learn from being bored. The disadvantage is that the skills that players decide to invest their skill points on may have no bearing on what they did to get the skill points in the first place—that is, it’s unrealistic. Some games try to limit you to only spend points on something you did in the last level (that is, you had to make a piece of clothing at least once to advance tailoring). Players will find and jump through these hoops very quickly.

Star Wars Galaxies uses a very novel approach that impressed designers because of the way it manages to kill several birds with one stone. The idea is that for craft skill points, the measure of success is whether anyone wants to buy your goods or not. Therefore, the more people use the products of your skill, the more skill points you get. This gives crafters a reason to produce the items that PCs want rather than the items NPCs will pay for. It helps stake-holding, it helps community, and it integrates crafters more fully into the economy. Players are as excited over the potential use of their skill as over the actual use of it. Presumably, there could be an improve by failure equivalent, whereby characters gained skill points by making items nobody wanted, but this has none of the advantages of the SWG original. Unfortunately, it does have one very important advantage that the improve by success version doesn’t: It makes sense. In real life, people buy a crafter’s goods because the crafter is good at making them, not the other way round.

There is a final skill advancement mechanism that isn’t used much at the moment but may become more important in the future: improvement from purchase ². If the virtual world’s business model and player base are right for it, characters could gain skills by simply paying for them. This was first done implicitly in the early 1990s by Avalon, which allocated skill improvement points to characters depending on how long their players stayed logged in (at a time when it charged by the hour). Achaea is more upfront about it: You pay real money for credits, which you can then turn into skill improvement points.

When skills can rise inexorably along a linear scale, it’s worth asking what the values mean. Is a character with a strength of 100 five times stronger than someone with a strength of 20? Unlike the case with character levels, it could well be possible to measure this directly—for example, by considering the maximum weight that the virtual world lets a character carry. If something approximating a proportional relationship is established, this raises a number of issues:

• If the principle is established for one skill, should it not apply to all skills?

• Is it realistic to have skill disparities as large as this?

• Are players agonizing at character-creation over whether to give their character a strength of 15, 16, 17, or 18 simply wasting their time thinking about what’s basically just noise?

That mention of player design brings us conveniently to our next topic.

Character Generation

Player characters do not come from nowhere. They have to be created. If players themselves do the creation, they will identify more with their characters right from the beginning. However, if they don’t know what they are doing, then this could easily backfire as they make arbitrary decisions that they shortly come to regret.

Although it’s not part of the virtual world per se, character generation is the first thing players have to do when they start to play a virtual world. It’s therefore very important, as the impression it leaves will shape how players view the virtual world proper when they enter it.

In MUD1, players were asked for their character’s name and gender, and that was all. Nowadays, characters will typically need to have their class, race, skills, and appearance set before they can set foot in the virtual world—a far more complicated exercise. Let’s take a look at what’s involved.

Appearance

The first two questions that need to be answered for any virtual world are still character name and gender, but increasingly character appearance is becoming important.

In graphical virtual worlds, obviously it’s paramount (although even there it’s not mandatory—Lineage succeeds despite having thousands of characters that all look the same). In textual worlds, a character description is usually optional and can be added or changed at any time; a non-default description can often be thought of as a badge of non-newbiehood. Even some graphical virtual worlds allow players to create textual vignettes for their characters, to enhance role-playing.

Almost always, a character’s appearance is intangible: The virtual world itself makes no reference to it, so gameplay effects are non-existent. Although gender and racial ¹ characteristics will affect the image, other attributes won’t: You can be short and skinny yet still have more strength than someone twice your size. Unrealistic though this may be, players like to customize the look of their characters more than they like realism; at this stage, identification with a new character is more critical than its future effect on immersion.

An idea yet to be taken up by any major commercial virtual world ² is that of hyperlinks. It’s a simple matter to associate a hyperlink with any object in a virtual world, characters included, such that activating it opens a browser window to the character’s home page. This gives players even more of an opportunity to flesh out their characters, and can include other useful information, such as objects the character has for sale or is looking to buy. It’s convenient, and it’s extra content that costs the live team nothing to create. However, it’s also 100% immersion-breaking and very open to abuse. Some players will create links to inappropriate web sites, and other players will complain about it; the live team’s workload will increase accordingly. Linking to pages hosted on the developer’s own servers and limiting content to that auto-generated from predefined forms would help enormously, but would lead to increased bandwidth requirements (that is, expense).

Graphical virtual worlds use a paper doll approach to character image design. A basic, randomly featured model of the character is presented, which the player can tweak within certain parameters (the client is going to have to animate this, so changes can’t be too esoteric). Some adjustments concern the underlying wire frame (for example, height), some concern the palette (for example, skin tone), and some concern texture maps (for example, facial features). There are so many variables involved (some of which are interdependent) that most characters will look distinct from one another, except when someone is deliberately trying to create a doppelganger of someone else. Players generally like their characters to be unique ³.

When players create their character’s appearance, it’s the first opportunity they have to engage with their character. It therefore represents a powerful iconic representation of their identity goals. Unfortunately, these goals can change: You might start off as a dark, brooding, menacing figure, but in the course of interacting with the virtual world and its players discover a lighter side to yourself. However, the first impression you’re giving when you meet someone new is “I am a dark, brooding, menacing person.” Changes to apparel can have some impact on this, and there may be hairstylists, beauty therapists, and magic/plastic surgeons able to take it further. Your decision to change has to be conscious, though, which can present a barrier to identity exploration. Short of making characters change physical appearance based on their emotional behavior (which, were it implemented, would lead to a world full of characters looking like their own Dorian Gray portraits ⁴), this is something that graphical virtual worlds simply have to live with.

Appearance is a two-way cipher. It encodes what you want other people to think of you, but it also encodes what you want to think about yourself. When a non-newbie male player creates a female character, that character encapsulates idealized values that the player is aiming both to project and to experience—only very experienced role-players can keep the two separate. The same applies to all other aspects of character appearance—race, hair, eyes, skin tone, height, and so on. These are lexical tokens that combine to create impressionistic symbols meaningful in an emergent, cultural language of identity. Although rooted in the tangible (and, in the case of gender, in reality), it’s the abstractions that players make, and the common vocabulary they adopt for communicating these through appearance, that are important.

This semiotic view of appearance comes with two caveats. Firstly, not all players will invest time creating characters—they’ll take the random image with which they were first presented (unless perhaps it’s outstandingly hideous). The players most likely to do this are

• Newbies who are super-keen to enter the virtual world for the first time.

• Potential newbies checking out the virtual world before committing themselves.

• Players creating secondary characters for some specific purpose.

• Oldbies who are already as one with their online selves.

The other semiotic caveat is that some players use a more concrete real-world symbolism of direct identification. As noted in Chapter 3, Tolkien’s Gandalf character is one favorite role model, but there are plenty more from other works of fiction. Thus, in a Fantasy virtual world you may find a female character with dark, wavy hair, dressed in tight-fitting, purple clothes and calling herself “Deanna ⁵.” However, although this phenomenon is widespread across virtual worlds, each individual world usually has only a small number of people determined to role-play against the world fiction to this degree.

Character Generation Methods

It seems a simple task: Set the initial values for a fixed number of genders, races, classes, attributes, and skills. What is the best way to do it, though, so that it provides enough dilemmas to make players think about what they’re creating, yet can be completed quickly for those who just want to play?

There are many different ways to do it, a surprising number of which can be offered simultaneously. Let’s go through the main ones.

Random

The easiest method to create a character is at random. The server ¹ rolls digital dice to determine all values.

This is fast, and it may produce interesting character ideas that the player had not considered, but some characters will be measurably better than others (depending on how the dice fell). A modified version uses a balancing scheme, whereby each attribute, skill, or whatever is allocated a points rating. Random values can then be adjusted upwards and downwards so that the sum total of points is the same for all new characters but their distribution won’t be. This modified system has the advantage that it can be done on the client, with the server merely sanity-checking the results.

Pick Points

Starting from the modified random position, pick points allow players to tinker with the results. This saves players from re-rolling continuously until they get the result they want. They can reduce some values to gain pick points (equivalent to their rating) and spend them on raising other values. Sometimes, not all the pick points initially available will have been spent by the randomizing process, leaving a pool for players to draw from. This is better, because players don’t feel as if they have to harm their character before they improve it; however, it does mean that a “randomly flesh out” option should be available for players who don’t want to waste time messing around.

The cost of attributes and skills will normally increase (and sometimes decrease) disproportionately beyond a mid-range, to stop players from minimaxing. For example, if a character’s default tracking ability is 40, then it may cost one point to rise to 50, another point to rise to 60, but then two points to rise to 70 and a further three to rise to 80.

One of the ways by which classes and races can be made more distinct is to provide different base (and, sometimes, ceiling) values for skills and attributes. A stereotypical elf, for example, may begin with higher defaults for dexterity and archery than a stereotypical dwarf; the dwarf may have higher strength and axe-wielding abilities to compensate. In these cases, pick point costs may be relative to the defaults rather than to some absolute scale: If the elf’s default archery skill is 60 then it would only cost one point to raise it to 70 instead of the two it would cost the dwarf with default 40.

The total number of pick points available to a character may in some systems be increased by taking a flaw (or quirk). The idea here is that the character has some flaw that the player is willing to accept in return for more pick points. For example, if a player is designing a “desert warrior” character, then it might be worthwhile accepting a flaw of “fear of heights” (on the grounds that the character will never encounter any). Were a giant eagle to pick up the character and drop it at the top of a minaret, however, the flaw would kick in; at the very least, the character could expect a catastrophic loss of attribute points while the situation prevailed.

In addition to flaws, some systems have one-off improvements that can be bought for pick points (for example, “language aptitude”). In this case, the improvements are known as advantages and the flaws disadvantages ¹. Because most advantages fit (or can be fitted) into a skills/feats/attributes scheme, though, they are popular.

Kits

Character kits are predefined pick point spends optimized by the designers for particular playing styles. For many newbies, going through extensive skills lists without having a clue as to what’s useful and what’s not is very daunting. It’s easier to look down a (shorter?) list of character kits and make just one major decision, and then maybe fine-tune the result.

Character kits have a special significance for classless virtual worlds, because they allow players to create a character that is a “sorcerer,” “ranger,” “merchant,” or whatever, but don’t lock them into it. You can start off as a mendicant priest and end up as chief of the assassins’ guild; the character kit you start with points you in a direction of your choosing, but you don’t have to follow it.

A nice feature of character kits is that they can be saved and loaded. Because all the action in character creation happens at the client, this means players can customize their own character kits and give them to their friends, post them on web sites and so on. This makes for a more involved community, although the temptation always exists to create a crock kit and label it “arch-wizard” to trick newbies.

Questionnaires

Questionnaires offer a different approach to character creation. Players answer a series of (usually) multiple-choice questions, from which a piece of software constructs a customized character. Questions may or may not be focused—that is, later questions may or may not depend on your answers to earlier questions. This can be done independently or tied into a kit system; it’s easy to see how a player-written web-based tool could construct characters from forms even if the developers decided against doing it officially.

There are basically two ways that questionnaires can be used: Ask players about the character they’re creating; ask players about themselves. Examples:

• Your character prefers to deal with enemies

  • Face to face.

  • At a distance.

  • Behind their backs.

  • By giving them what they want.

• You view your fellow players as

  • Potential victims.

  • Part of the world.

  • The reason you play.

  • Competition.

The first approach constructs a profile of the character that the player wants to play, then attempts to spend pick points in a way that best reflects this. In the example given: The first answer might suggest a confrontational role such as fighter or lawyer; the second answer might suggest archer or magic-user; the third answer might suggest thief or politician; the fourth answer might suggest crafter or merchant ¹.

The second approach tries to create a character that is right for the player; this may be radically different from what the player would have chosen otherwise. In the (somewhat unsubtle) example given, the answers basically point at player types—killer, explorer, socializer, achiever. Although players can enjoy the surprise results that this technique can deliver (if only to scoff at them) and can have fun wrestling with off-the-wall questions (“If you were a color, what would it be?”), nevertheless it’s a form of psychometric testing; many players find the idea of this uncomfortable ².

Tabula Rasa

This method creates characters by constructing a backstory for them. Starting as a callow youth with no skills, no quirks, and no unusual attributes (that is, an empty slate, or tabula rasa ¹), the player chooses a profession from a list. This adds some skills and perhaps changes a few attributes. The player is then presented with a second list that omits many of the original choices but offers new ones instead (depending on what the character decided). The process iterates, with new opportunities and specializations presenting themselves. When it stops, not only can a character be created, but a short history, too.

For example:

“Zayre spent much of her formative years helping her carpenter father. The call to adventure was too strong, however, and so she joined a merchant train as a traveling cooper. Soon, she turned her talents to fletching, and became adept with the bow. Her reputation grew as she defended the caravan from bandits and raiders, until eventually she left to join the army as an archery instructor. Finding the discipline too stifling, she volunteered to become a scout, patrolling the barbarian wilderness with only her horse, her bow, and her thoughts for company.”

I first encountered this idea in the single-player role-playing game Darklands ², where it was very effective. It doesn’t seem to have made much of an inroad into virtual worlds, though, perhaps because it implies that characters age. This is a shame, as it remains my personal favorite.

Essay

High-powered role-playing virtual worlds turn character creation into an art. Players of games such as EverQuest and Anarchy Online who regard themselves as “role-players” would do well to consider whether they could have written an essay on their character and its background before ever playing, to be submitted for approval to the world’s staff members. That’s what’s required of some MUSHers.

Role-playing virtual worlds work consensually. People who do not buy into the fiction or who are disruptive of it have no place. This makes all players simultaneously guardians and perpetuators of the fiction. To gain admittance, newbies must therefore demonstrate that they both honor the fiction and have an aptitude for role-playing. An essay forces a player to create a fully rounded character, in the process satisfying both these goals. As mentioned in Chapter 1, “Introduction to Virtual Worlds,” some virtual worlds will conduct interviews, too.

This kind of freeform character creation only works for small-scale, intimate virtual worlds with very dedicated players. Given that most players regard essay writing as a form of torture that should be outlawed by international convention, this is perhaps not surprising. Nevertheless, in its place it can bring character creation to the highest planes. It’s not by chance that some of the best role-playing MUSHes have waiting lists ¹.

Physical Differences

Elves are taller than dwarfs, but not at tall as humans. Dwarfs are stronger than elves, but elves are more dexterous. Elves and dwarfs can see in the dark, but humans can’t. There are physical differences between character races, which are reflected in their default attribute settings.

Okay, so what about character gender?

Ah. In virtual worlds, the differences are usually purely cosmetic. Male and female characters look different, but apart from that gender X is every bit as good as gender Y and vice versa. There are three main reasons for this:

• If characters of one gender were more able than characters of another gender, it would greatly annoy real-life representatives of that gender because it would seem to be saying things about them.

• On the whole, ability differences between genders really are negligible. In day-to-day situations (as opposed to extreme ones), tasks are usually gender-independent.

• Virtual world designers are a liberal lot who don’t want real-life prejudices either seeping into their virtual world or being reinforced by it.

It may surprise many students of gender studies to learn that the third of these is perhaps the strongest. It certainly applied to MUD1 ¹, which made no tangible distinction between character genders except in three instances that I’ll discuss shortly. Such gender-blindness is all very noble, of course, and it promotes attitudes that can only be good if carried out to the real world. Unfortunately, there are some real-life physical differences between men and women that really can’t be dismissed in virtual worlds.

In real life, genetic females can have babies, but genetic males can’t. Few real-life women can grow a beard, most real-life men can. Any virtual world that suggested that either of these statements were untrue would have a hard time maintaining immersion.

What about differences in height? That’s a little fuzzier. The average adult male is taller than the average adult female, but some women are taller than some men. I’m taller than my wife, but her brother is shorter than his wife (who is also taller than me). Nevertheless, I’m noticeably taller than most of the women I know. In a virtual world, would I have to be the same height so as not to disadvantage female characters?

Well, perhaps not, because in a virtual world height can easily be regarded as cosmetic. Humans won’t bang their heads in dwarf houses, and dwarfs can reach things from the top shelf of a bookcase. Graphical worlds might have a more practical problem in that super-tall races simply won’t fit in small places, but if they do it’s a problem of their own making ². Nevertheless, in general female characters can be made to look smaller than male ones while having collision-detection boxes the same size.

However, what about strength? Men have significantly greater upper body strength than women. I may be shorter than my sister-in-law, but I can easily beat her at arm-wrestling. Should male characters have a higher base strength than female characters? If so, should female characters be compensated with a higher dexterity and/or stamina? This is precisely what MUD1 did: Initial attributes for characters were set using a random-number generator, then female characters had 10 subtracted from their strength and 5 added to each of their dexterity and stamina. It was possible to get strong female characters, but strong males were more likely.

In reality, though, the differences between male and female dexterity and stamina are marginal compared to the male advantage in strength (dexterity may be more related to size than gender). I was balancing genders in MUD1, but not in a realistic fashion. How far should virtual worlds reflect reality as it is, as opposed to reality as people wish it were?

There were two other ways that MUD1 made a distinction between male and female characters. One of these was to have a matched pair of puzzles that you had to be a certain gender to complete, namely kissing a sleeping beauty (for males) and kissing a frog prince (for females). I was never happy with these, though, because they were unfair on gay players: Both the sleeping beauty and the frog prince were heterosexual, and didn’t take kindly to advances from members of their own sex ³. I did not migrate either of these puzzles to MUD2.

The other tangible way that MUD1 used gender was that it had magic to flip it: Male characters could become female, and female characters could become male. This was non-consensual—other characters could cast the necessary spell on you whether or not you wanted them to. It wasn’t something that people did very often: It had a high chance of failure, the penalty for which was the caster’s falling asleep; besides, most people didn’t actually care if it happened to them ⁴. Having their character’s gender changed commonly angered only two groups of players:

• Newbies and recently ex-newbies, who hadn’t yet grasped the fact that virtual worlds can have a different culture to the real one.

• Players deliberately playing characters of the opposite gender who viewed it as an interruption of their experiment.

It’s possible to have tangible but neutral differences between genders that (in theory at least) hurt neither. For example, plate armor breastplates could reasonably be restricted so only characters of the gender they were made for could use them. This might lead to economic imbalances, though: If supply and demand differed by much, one gender could end up paying more for their breastplates than another. Should a designer strive to eliminate such unfairness, or accept it as an inevitable consequence of increasing the virtual world’s depth?

To conclude: There are significant physical differences between men and women in the real world. For the short-term periods that characters exist in virtual worlds, these can be embraced fully, embraced cosmetically, or conveniently ignored. Unless they’re making a deliberate point about gender, most virtual worlds will stop short of adding tangible gender differences between characters. There are too many real-world consequences otherwise.

Yet character gender isn’t really all that different to character race. Why is it fine to suggest that one race might be stronger or smarter or more pain-resistant than another, but it isn’t fine to say the same thing about gender? These may be made-up “races,” but it’s easy to see how such casual references to “racial differences” can perpetuate false notions of racial differences in the real world. Chapter 8 considers this issue in greater detail.

Long-Term Characters

Should player characters age? There are some characters in text MUDs that have been played for so long that, were they alive in the real world, they’d be entitled to vote.

For virtual worlds with moderately low persistence, it’s perhaps not an issue. If the world periodically resets, it never ages; it’s not inconsistent to suppose that the characters entering such a timeless world shouldn’t change either.

For greatly persistent worlds, it’s not this easy. EverQuest’s night/day cycle is 90 real minutes in length. That means 3 years of real time is nearly 50 years of virtual time. As the virtual world never formally resets, shouldn’t 3-year-old characters be in late middle age by now? Everyone must sup regularly from a fountain of youth.

Aging, like gender differences, is one of those inconvenient facts of real life that designers often omit from virtual worlds. They don’t even address it cosmetically: Beautiful characters are forever beautiful.

The reason for this, of course, is that aging implies decay and, ultimately, death. The decay part might be acceptable to players if it’s balanced by something else—for example, dexterity might drop but intelligence might rise. What they are unlikely to accept without a great deal of persuasion is that they themselves are responsible for supplying the main gain that comes with aging: experience. Their character will be better when it is older because of what it experienced when it was younger—that is, how they played it.

Character death is not something that sits easy with players, even when it’s “natural” through aging. Would people buy a music CD if it was guaranteed to stop playing after five years? Why would a player want to start in a virtual world knowing that their character would be taken from them when at the height of its powers? What about the players of elf-types, whose characters are practically immortal? Would female characters get an extra two or three years, as they do in the real world? It’s almost inevitable that in such a climate, some fiction would have to be created to assuage the concerns of long-term players. Characters don’t age because, er, there’s something in the air.

Okay, so they don’t age. Time still passes, though. Characters get married all the time in virtual worlds—it’s a common ceremony (and one I’ve performed many times myself). Do they get to have virtual children?

Virtual children pose so many problems for virtual world designers that most will avoid the topic at all costs. Unfortunately for me, as author of this book I can’t do that. Here, therefore, are the main issues:

• Children have to be conceived. We all know how this happens in the real world, but….

• Women bear children and men don’t. Is this unfair on male characters, unfair on female characters, or unfair on both?

• It takes nine months from conception for a human baby to be born. Even at accelerated time, this is plenty long enough for the player to have second thoughts. Are virtual abortions allowed?

• After a few months, pregnant women look different to non-pregnant women. Do their character models change?

• The final few weeks of pregnancy can be quite disabling. Should pregnant characters still be able to hack dragons with the best of them, or would suggesting this be irresponsible?

• Labor and the birth: These are going to be faked up to happen when the player is logged off, right?

• Naturally all babies are born in perfect condition without any complications. So, who looks after them, and how?

• Are children shown in the virtual world? If so, they grow, they change shape—all that needs to be modeled.

• How are children’s looks and attributes computed? Are “designer babies” possible? What about something basic, such as choosing a child’s sex?

• Parents don’t age, but their children do?

• Can you sell your children into slavery? Every real-world parent has thought about it….

• Have you any idea how expensive it is to bring up kids?!

• Are children educated automatically? Do parents have any say in the matter? What if they disagree?

• Can children die before they reach adulthood? Can they be killed?

• Who gets custody of the children when the parents split?

• Can parents abandon their children? In worlds with permanent death, what happens to orphans?

• Who gets to play a child once it reaches adulthood?

• What tangible reasons are there for having children anyway?

There are serious moral, practical, and implementational issues concerning virtual children. They can all be addressed, but never to the satisfaction of everyone. In most virtual worlds, the cost of implementing virtual children so outweighs the benefits that it makes sense to put the idea in the “forget about it” pile. The main exceptions are

• Virtual worlds with a strong, realistic theme (particularly ones in an historical context). Children may be necessary for gameplay reasons, and the prevailing morals of the modeled reality can be used to answer some of the more awkward questions.

• Virtual worlds actually about children—for example, educational ones.

• Virtual worlds with permanent death, where children are used in an inheritance fiction to replace dead characters ¹.

And that, thank goodness, is quite enough of that.

The Virtual Body

Players enter virtual worlds through the medium of their characters. Although in general the player determines the appearance of a character, characters are not merely ghostly projections of the spirit: They are tangible entities within the virtual world, as real (to it) as any other object. As representations of sentient beings, they have attributes and skills; however, they have other functionality merely by virtue of their being objects. From this point of view, the virtual body is a machine that the character lives in. What should this machine do? And what happens when it stops doing it?

Maintenance

You wake up, you visit the bathroom, you wash, you brush your teeth, you have breakfast.

You wake up because you went to sleep. You have to go to sleep because human (and most other animal) brains won’t work if you don’t.

You visit the bathroom because your body is powered by chemical reactions that generate waste products.

You wash and brush your teeth for cultural reasons. You don’t have to, although the chances are it’ll prolong your life considerably if you do. People used to die if they got an abscess on a tooth.

You have breakfast because you need fuel. You noticed you were breathing too?

This is an example of a maintenance schedule. It varies from person to person. Some people might include a bit of exercise, others might count their morning prayers as “maintenance for the soul.” The point is, it’s a routine. Except in special circumstances (for example, illness or vacation), it’s something you do daily; it’s not for fun, nor for gain, but to avoid the consequences of not doing it.

Should virtual bodies have maintenance routines like real ones? Characters have no need to sleep or eat or anything else unless designers say so. Why lock them into the same tyranny of routine that players must observe?

The only strong argument in favor is that of realism (that is, it helps immersion). Unfortunately, in this particular case it’s very powerful. Players don’t actually want to undertake a routine for their character, but they do need to know that their character has a routine, and that it’s being followed. It’s a general principle of virtual world design that if every player has to do exactly the same boring thing, none of them should have to do it. Thus, the solution suggested in Chapter 4, “World Design,” of abstracting the maintenance routine is a good one; furthermore, the fact that characters are charged for it means that players will notice, thereby reassuring them that these strongly ingrained but wholly irrelevant activities are taking place ¹.

There is a difference between abstracting and removing, though. A character who puts on make-up might want to wash it off right there and then; in this case, it’s not an act of general maintenance, but a normal action to achieve a specific purpose. Similarly, just because characters by default empty their bladders when their player is logged off, that doesn’t mean they shouldn’t be able to do it at other times: Urinating on a small campfire is as easy a way of putting it out as any other (at least for men) ². If characters want to eat, drink, and sleep explicitly rather than behind the scenes, they shouldn’t necessarily be prohibited from doing so.

To summarize, then: Routine maintenance should always be abstracted, but the actions that it abstracts should also be available independently (cultural circumstances permitting).

Survival

Don’t go to sleep on the ground if it’s covered in snow; if you do, you’ll get hypothermia and die. The same thing will happen if you swim in icy water. In a desert, if you don’t freeze to death at night you may bake to death during the day.

That’s in the real world. In virtual worlds, you will either suffer some minor, temporary skill or attribute degradation or you won’t notice any ill effects at all. Characters are generally immune to the environment.

So? Why be unpleasant to players just for the sake of it?

Why indeed….

Well, it’s because there are subtle side effects to this kind of automatic survival. For example, people live in houses primarily for shelter. If they don’t have anything from which to shelter, they don’t have to live in houses. Sure, they might want them as warehouses, but that’s hardly the stuff of which communities are made.

There are other survival reasons that cause people to want to live close to one another. If characters are going to live anyway, though, these don’t apply. Easy access to food and water is only an issue if players have maintenance schedules that need them; healthcare is only important if people can fall ill; mutual defense is unnecessary if attackers can’t hurt you.

In other words, some of the more primitive and tedious aspects of the real world that players don’t want to experience act, unfortunately, to set up some of the more advanced and enjoyable aspects that they do want to experience. This conflict between what players want and what they are prepared to put up with to get it is a constant that runs throughout virtual world design ¹.

Sensing the Virtual World

Traditionally, human beings have five senses: sight, hearing, taste, smell, and touch. These are the means by which individuals can detect the world about them. Virtual worlds that have human or humanoid characters will usually want to account for these senses, with perhaps some additional ones for special cases (bat-like sonar, insect-like infravision, and so on).

Most virtual worlds, whether graphical or textual, emphasize vision. When a character enters a location, the player experiences what the character can see. They may also experience what their character can hear. Textual worlds also will make occasional reference to smells and tactile sensations (for example, heat from a fire), but taste only really comes into it when something is eaten or drunk.

In graphical worlds, the players’ senses double up as the characters’: Players actually see what their characters see ¹ and actually hear what their characters hear (plus the annoying soundtrack). In textual worlds, players are told what their characters see, hear, smell, and feel (through the skin—not emotionally). The scene is constructed in the imagination of the players, which paradoxically can lead to a more vivid experience. Although most players of textual worlds will of course use their own sight to read what is sent to them, this does not have to be the case: There are numerous examples of blind players using speech synthesizers or non-blind friends to read the text aloud; graphical worlds are not conducive to this.

Although textual worlds can utilize more senses and deliver a more compelling experience, graphical worlds have three main advantages over them:

• The graphical experience is more immediate: You don’t have to interpret what you see, you just see it.

• Graphical images can display more information of a certain kind than can textual images. Scanning 50 faces for a friend is easier (for most people) than scanning a list of names ².

• Graphical worlds use multiple channels. You can hear while you see. Textual worlds are essentially linear: You’re told what you see before or after you’re told what you hear.

For newbies in particular, text is unappealing. The fact that you can take photographs of smells in textual worlds is less relevant than the fact that you can see where you’re going in graphical ones.

Sometimes, senses are impaired or ineffective. In real life, for example, you can’t see in total darkness. What happens in virtual worlds?

Well, there are a number of solutions. Here are the most common:

• Suggest the impairment, but don’t enforce it. A graphical world may darken the screen, but not actually black it out.

• Provide a trivially easy remedy. Give everyone a portable light source.

• Make it temporary. If you enter a dark room from a light one, your eyes need to adjust to the dark before you can see; when you return, your eyes need to adjust to the light first. The process only takes a few seconds, but the wait can be scary. Popular in textual worlds.

• Play it straight. Actually black out the screen. Don’t send update packets, either, so players can’t hack the client to make it think it’s full daylight. Only relent if they’re under attack.

Impairment is one thing, but what happens when one or more of your character’s senses isn’t working at all? What if, as a result of accident or magic, your character loses the ability to see or hear?

In graphical worlds, the answer is simply “It Does Not Happen.” It would be possible to continue playing without sound, but without vision there simply isn’t enough other sensory information available to compensate.

In textual worlds, it’s a different story. You can try things and get feedback. If you walk into a wall, you can be told there’s a wall in the way (rather than a tree or a door or a dragon). Things that make noises or have smells or give off heat can still be described. You can pick up things and know what they are by their shape. Losing a sense could even be essential for solving a puzzle—for example, if you’re blind, then the snake-hair ugly woman can’t turn you to stone.

The way to do this is to use dynamic descriptions. Most textural virtual worlds, it should be noted, use static descriptions and therefore also have problems implementing lost senses (generally they take out too much information—blind characters can’t hear either). Here’s an example to illustrate what I mean.

A description of an object might be “Ticking in a corner is an antique grandfather clock.” If a deaf character (or one wearing ear defenders) came into the room, they wouldn’t know it was ticking. Should they see the description of the clock or not? Well, strictly speaking, not. If it were a static description, though, either the whole description stays or the whole description goes, neither of which is satisfactory. Dynamic descriptions individualize the text depending on the character’s sensory abilities. Thus, a deaf character would see “Standing in a corner is an antique grandfather clock;” a blind character (or one wearing a blindfold) would hear “Ticking nearby is a grandfather clock.” If the designer doesn’t consider ticking to be enough to identify an object, it might just be “Something ticks nearby.”

Dynamic descriptions allow for changes due to a character’s skills or knowledge, too. In the preceding example, the use of the adjective “antique” may depend on whether a character is an historian or not, for example. If a character is running, they may notice less than if they’re walking; if they’re concentrating or searching, they may notice more. A virtual world can incorporate this information into a dynamic description appropriate to each character.

Although senses can work below their normal levels, they also can be implemented to work beyond them. Characters can be given the ability to see like they were looking through a telescope, or to hear heartbeats at a hundred paces, or to taste the recipe that was used to create a dish. Superhuman senses in a world that contains superhumans do make, er, sense. Although there is no problem with this for active senses (ones you have to decide to invoke—for example, touching something to discover how it feels), passive senses can be more problematic. If you can hear heartbeats at a hundred paces and there are fifty people within a hundred paces, that’s a lot of heartbeats. For this reason, even passive senses will usually be considered active beyond a certain threshold. Switch on your bionic ear when you want to hear how many guards there are at the other side of the door, then switch it off afterward.

Body Composition

When a character picks up something, strangely this does not mean that the object teleports into their backpack. To make sense, they need to use their hand. Similarly, when they walk, their legs and arms move, when they look to one side their head moves, and when they’re fleeing in mindless panic every part of their body moves. The issues involved here extend beyond mere animation: If characters can move their hands, that implies they have hands to move.

It doesn’t take long following this line of argument to conclude that bodies must be regarded as assemblies, as discussed in Chapter 4. Following on from our look at senses, though, it’s natural to ask the question: What happens if a body part works above or below normal levels?

As with senses, the default is that every body component works to its full potential. Because of the way that they interact, it’s more difficult to make individual body parts superhuman than it is for senses; there may be comic potential in giving a character a hugely muscled ring finger, but it wouldn’t be a lot of use unless the rest of the hand and arm was muscle-bound too. For this reason, most of the consequences of having improved body parts in virtual worlds are abstracted as character attributes instead.

When body parts don’t work for whatever reason, it’s a different matter. Tedious though it may be from a programming perspective, it is nevertheless possible to put a guard on all commands to restrict their individual use if a body part (or assembly of parts) that is out of action ¹. For example, if your hands are tied behind your back and you want to pick something up, you can’t. You might be able to “get key with teeth,” but then you wouldn’t be able to use your mouth assembly until you’d let go of it. Note that some body parts are connected to senses. Most importantly, if a character’s eyes are shielded or missing, then they can’t see anything.

It’s tempting to use this idea to build an abstract anatomical model of a human body so that the effects of various events can be worked out. “Ah, when the nose and mouth are both blocked that means no air can reach the lungs, therefore no blood can be oxygenated, therefore no muscles can work, therefore death results.” In practice, though, this can be handled explicitly without the need for deduction. A spear through the chest means your character is dead—that’s really all you need to know.

Body parts can cease to function for three main reasons:

• Restraints. You can’t run because you’re wearing diving boots.

• Damage. You can’t run because a tiger ate your leg.

• Flaws. You can’t run because you have a club foot.

Damage can arise in some combat systems that allow for “critical hits” or “targeted damage.” Although ostensibly permanent, there will almost invariably be some means by which a lost limb can be recovered.

Flaws are the quirks that some virtual worlds allow players to assign to their characters in return for more pick points. Again, although these are supposed to be inherent aspects of a character’s identity, they can frequently be cured. The fact that quirks can (literally) “add character” is only of long-term interest to hard role-players; soft role-players may regret an early decision to take a flaw, and in a virtual world without permanent death they will complain rather than start anew.

Note that a character lacking one or more organs, limbs, or senses can be considered disabled ². There are ethical issues arising from this, discussed in Chapter 8.

Groups

In virtual worlds, being the social beasts they are, it is imperative that players be able to form groups. The more populous the world, the more important it is that there is hard-coded support for this activity.

Chapter 3 introduced the idea of groups as the foundations of community. The point was made that people should be able to join multiple groups, so they can participate in multiple communities that are not hermetically sealed from one another; this gives players a greater sense of belonging to a virtual world (translation: They’ll find it harder to leave if they can’t take all their friends with them). When players can join only one group, that group will develop to maturity quicker but can leave as a unit; also, individuals who leave their group will often find hooking up with another group difficult, which could mean a crisis for them.

Single or multiple group membership is just one of the dimensions that can be used to describe groups. Other important ones are

• Formal or informal?

• Temporary or permanent?

• Flat or hierarchical?

• Hardwired or softwired?

Let’s take a closer look at these distinctions.

Formal or Informal?

In small-population virtual worlds, everyone knows everyone else and there is no need for formal grouping commands. If you want to create a group for whatever purpose, you chat to prospective members and there you have it, a group.

This is an informal group. The main problem is finding someone willing to do the organizing, rather than the virtual world’s lack of support for it. Informal groups arise in all virtual worlds. In small-scale ones, they’re all you need; when more players are involved, some notion of a group as a tangible entity usually surfaces.

If members of a group are to communicate on a special group channel, or are to accept mutual concessions (such as seeing one another’s basic stats or pooling experience points gained), or if the group as an entity is to possess anything (money, property), the virtual world needs to be able to reference that group. There must therefore be a formal process by which the group is created, which gives us the concept of formal groups. It is quite common for formal groups to be created from informal groups.

Note that it’s conceivable in either case for a group to have NPCs as members, although for informal groups the AI necessary to support it would be considerable.

Temporary or Permanent?

Sometimes, people want to group together for only a short period—a single session. A group of characters are hanging out at a bar, someone comes in and announces that there are wolves attacking the village, so everyone bands together to beat off the wolves. This would be a temporary group.

Groups that persist (in particular, that would continue to exist when all their members were logged off) are classified as being permanent. Even if the group has a definite aim and will cease to exist after a certain point (for example, a consortium to sell virtual pumpkins on Halloween), it counts as being permanent if it would survive as an entity were all its members to log off.

Temporary groups tend to be (but aren’t always) informal; permanent groups tend to be (but aren’t always) formal. For formal groups, there may be differences in facilities depending on whether they are temporary or permanent: It doesn’t really make sense for a temporary group to own property, for example. Although it’s possible to implement a temporary group as a permanent one, there are two principal reasons why a designer might decide not to do this:

• People are often more willing to sign up to a group on the understanding that it’s a one-off commitment than they would be if it were permanent. They might want to join a lynch mob, but not the police.

• Players may be limited to one permanent group (to help it develop quicker) but allowed to join any number of temporary groups (to assist the development of a wider sense of community).

It is important that temporary groups be quick to form and quick to disband. Players regard them as a superficial convenience, therefore anything that makes them inconvenient (even having a hard-to-remember command name) will reduce usage.

Flat or Hierarchical?

When all members of a group have the same powers as one another, that group has a fiat organization; if some members have more powers than others, it is hierarchical.

Most formal groups are at least nominally hierarchical, as they usually have a designated leader who (minimally) can expel individual members but can’t be expelled by them. In practice, though, there is a big gulf between this shallow tree structure and a full-blooded hierarchy with multiple levels, promotion, and demotion between levels, and increased powers the higher up the hierarchy you go. This degree of structure is almost always only present in formal groups, and aspects of it may be supported by the underlying code. For example, if the group has money and can own things, there may be provision for “treasury powers,” allowing a designated individual to determine who is allowed to spend the money and within what limits (that is, give a budget).

There have been many experiments in virtual worlds with customizable hierarchies, with textual worlds (as usual) at the forefront. They have the benefit, however, of relatively small and innovation-friendly populations compared with those of graphical worlds, so even poorly thought-out ideas can work well enough in practice.

For the big, graphical worlds, it’s more hit and miss. Meridian 59 had a system whereby ranks could be created within a guild, allowing extra hierarchical levels that enabled groups to restructure themselves as they grew. This was a step in the right direction, although the powers associated with the new ranks weren’t easy to fine-tune.

The bravest early attempt at building a hierarchy was made by Asheron’s Call with its allegiance system. In this, a vassal character swears allegiance to a patron character. The patron may already have other vassals (up to a dozen), and may have their own patron (hence the hierarchy). The primary tangible benefits of the relationship are that a proportion of the experience points earned by all a patron’s vassals is also passed to the patron. There also is a ranking system, whereby titles are conferred on characters the more vassals they have beneath them ¹; there are tangible benefits to a high rank, in that some magical artifacts won’t work below a certain threshold.

Asheron’s Call only allowed characters one patron (that is, they could only be in one hierarchy). It did permit them to join temporary formal groups, although in practice these were usually composed of members of the same near branch of a hierarchy. Although it’s easy to criticize AC for limiting characters to a single social group in this way, actually that wasn’t the reason its allegiance system fell short of fulfilling early hopes. The problem was, although there were tangible reasons for a character to have vassals, there were only intangible reasons for a character to be a vassal. Although intangibles can hold together vast clans in (and sometimes across) virtual worlds, AC had an unbalanced system. By trickling up experience points, it asserted that allegiance was a part of gameplay; by trickling nothing back down, it denied it.

Most people need no incentive to be the boss of other people, but they do need incentive to be bossed by them. Critical to this is the belief that they themselves may be able to rise in a hierarchy, where “rise” means “getting nearer to the top” rather than “getting further from the bottom.” AC got it the wrong way round. It was close, but no cigar.

Hardwired or Softwired?

A hardwired group is one that is programmed directly into the virtual world. A s oftwired group is one that is created by the players. All hardwired groups are permanent, and therefore they can be thought of as a subset of permanent groups. Similarly, although it’s possible to conceive of hardwired informal groups (built around predefined, labeled communication channels in the style of CB radio), in most virtual worlds that have them all hardwired groups also are formal; they can therefore be thought of as a subset of formal groups, too.

It used to be that all formal groups were hardwired into virtual worlds. Nowadays, players can set up groups themselves, of course, and virtual worlds are much the better for it. That doesn’t mean there’s no room for hardwired groups, though. What softwired groups gain in flexibility, they lose in specialization. It would, for example, be unwise to allow softwired groups to impose taxes on the inhabitants of a town; it would perhaps be acceptable, however, to allow the town council to do so. The town council is an example of a hardwired group.

Hardwired groups exist mainly for functional purposes; although social networks can benefit from having features that are coded in (for example, you can’t join the local bandits unless your horse-riding skills are sufficiently high ¹), the more practical groups tend to dominate. They’re particularly useful for newbies, because they guarantee there’s an immediate place for them in the virtual society.

Note that there is a difference between hardwired features for groups and hardwired groups. All formal groups have hardwired features (it’s what makes them formal), but these are configurable. Even if new groups can choose arbitrary values for arbitrary properties to use as entrance criteria, that would still make them softwired as the players can do something about it. Hardwired groups are fundamentally inviolate: They’re as much a part of the virtual world as the landscape and the mobiles.

One of the main advantages of hardwired groups is that it’s much easier to arrange for NPCs to join them. Few players will care to stand around bored for hours on guard duty, but (presumably) someone has to do it; NPCs fit the bill perfectly. In this respect, NPC members of a group can be thought of as resources that the player characters can control—an option not really open to other kinds of groups.

Typical hardwired hierarchies (yes, they do tend to be hierarchies) are

• Ruling/legislative. The people that make the laws. Monarchs, councilors, presidents, and so on.

• Military. The defense (or offense) forces. There may be different branches for land, sea, air, space, or whatever. Captains, sergeants, wing commanders, and so on.

• Religious. Those who concern themselves with the spiritual well-being (as they see it) of the virtual world. There may be competing religions, with different temples offering different services and worshiping different gods. Priests, mullahs, bishops, and so on.

• Judicial. My gimp takes A’s gold and B’s silver and passes it to my smith, who makes it into an electrum brooch. The judiciary sorts it out ². Judges, magistrates, lawyers, and so on.

• Guild. Craft- or service-oriented groups of people pooling resources, determining prices, offering employment. Whether these are hardwired or softwired depends largely on how the economy is run ³. Guild masters/mistresses, branch secretaries, shop stewards, and so on.

• Educational. People who teach and train. Some of these may be part of other hierarchies. Professors, drill instructors, gurus, and so on.

• Law enforcement. Those who prevent crime and apprehend criminals. This hierarchy could be part of the military in some cases. Strictly speaking, law enforcement only has to enforce the laws made by the legislator; some hierarchies may have their own law enforcers with a very specific remit (for example, the religious police). Inspectors, inquisitors, speed cops, and so on.

All formal groups have powers. The extent of these powers is mainly over the members of the group, but in the case of hardwired groups this can extend significantly to nonmembers.

Furthermore, although players can take their characters out of softwired groups and set up with someone else, this is not an option with hardwired groups.

Although some of these powers can be limited by the code, they remain highly susceptible to abuse. Judges, for example, are always working with intangibles; otherwise, the virtual world could sort out the problems itself and judges wouldn’t be needed. Low turn-out elections can be hijacked by special interest groups (for example, griefer clans), guards can be made to abandon their posts, training can be withheld or offered only to friends; the list goes on.

Because there are significant issues concerning the control of hardwired groups, community management staff should be consulted about their design. Players who have an important position in such a group are often assumed by other players to be acting with the live team’s blessing, whether or not this is indeed the case. Even if the players who hold high office are vetted for suitability first, there could still be problems: Just how long can you make a character’s jail sentence be before it amounts to permanent death? With power comes responsibility; sadly, those on the receiving end of power rarely have the same notion of “responsibility” that those wielding it do.

Common Configurations

Although there are many different ways that softwired groups can be set up, some tend to be more useful than others. Designers may therefore want to have easy-to-use templates for these in place, so that players don’t have to worry about setting flags that don’t interest them. As long as the basic models are all configurable, this is fine.

Common configurations are:

• Fellowships. Temporary groups of characters that get together for a single session then go their separate ways.

• Parties. Permanent or (more usually) casual adventuring bands. People who play together regularly.

• Clans. Large, disparate groups of characters. Although mainly a social network, individual members may come together occasionally for activities that require numbers. They’re generally too unwieldy to organize for extended periods.

• Guilds. These are like clans, but with a supporting game fiction. Some virtual worlds pursue this vision more than others: Dark Age of Camelot, for example, allows players to design their own guild crests, with the result that guilds feel more integrated into the virtual world than they otherwise might.

• Alliances. Groups of guilds or clans acting as if they were one. A single, huge alliance that completely and aggressively dominates its virtual world is known as a überguild.

Many mid-sized groups are run by politician-type players, who chivvy people into doing things and generally keep the group together. Such groups can rapidly fall apart if their leader quits or moves on. Some designers regard this as a good thing, because it allows for players to join other groups with a better chance of survival (an evolutionary argument); other designers don’t like it because any failure of community could cause players to leave (the “other groups” that players join could be in another virtual world). The latter designers will institute automated rules of succession so that groups always have leaders, and votes of confidence so that inert leaders can be replaced; the former designers won’t.

Combat

Combat is an important feature of most game-oriented virtual worlds. It embodies conflict, drama, risk, bravery, friendship, honor, glory—it has everything! Indeed, in some virtual worlds it is everything; there’s very little to do apart from move from one battle to the next. As explained in Chapter 3, by no means all virtual worlds are like this, of course: Merely having combat ¹ doesn’t inevitably mean that the virtual world will end up being “all about” it. Nevertheless, many virtual worlds are indeed “all about” combat; this causes an ideological conflict with worlds that are “all about” something else (for example, role-playing), and annoys the balanced worlds who are lumped in with them even though they may be anti-conflict ².

The subject of combat in virtual worlds brings out some of the most passionate arguments in the whole field. Let’s take a look at it in detail.

How Combat Works

Taking a wide perspective, combat involves many activities: Magic-users cast spells, archers shoot arrows, clerics patch up the injured. Important and battle-winning that these may be, they are nevertheless mere adjuncts to the central conflict: Two individuals standing toe to toe, slugging it out with close-up weapons. The reason for this is simple: Hand-to-hand combat is more thrilling than ranged combat. The monsters are in your face. Yes, the wizard cranking up the spell that will nuke the bad guys is excited, but it’s an excitement borne by what’s going on at the front line: If the tanks aren’t excited by the fight, the wizard won’t be either.

For this reason ¹, most virtual worlds that have combat put hand-to-hand at the forefront. This defines the basic mechanism and the measures for success or failure. Ranged combat conforms to the same system, as does magic (whether the offensive or defensive kind). In discussing combat, therefore, it makes sense to focus on the hand-to-hand variety.

Here’s how MUD1 did hand-to-hand combat:

• One character attacks another player in the same location.

• An automatic fight sequence begins, in which characters take turns hitting each other.

• The chance of hitting is based on the character’s dexterity attribute.

• The damage done is based on the character’s strength attribute.

• Damage is deducted from the character’s stamina ² property.

• When one character’s stamina falls below 1, that character dies and the fight ends.

This is still pretty much the paradigm. Damage can be increased by weapons and reduced by armor and shields; spells can buff attributes, repair injuries, and do the opposite (to opponents); some combatants may have multiple attacks (dragons can breathe fire while clawing at you); skills may allow for additional special actions. In all these, the underlying mechanism remains basically the same.

Why did MUD1 do it this way? It looks suspiciously like the way that Dungeons & Dragons combat works—so similar that many virtual worlds openly refer to characters’ health as “hit points,” in the D&D tradition.

MUD1 did it this way because it delivered the gameplay. Roy wanted combat to proceed as exchanges of blows—I swing, you swing, I swing, you swing—because that was the most appropriate level of abstraction. He wanted it automated because otherwise it would have been tedious for players continually to type “hit them ³.” When one character swings at another character, there has to be some chance of hitting: This was encapsulated by the dexterity attribute. As for what to do when a hit got through, well, obviously there had to be some damage.

Roy and I discussed how to handle this. Although I knew the D&D combat system well, I also knew other ways to do this kind of thing (I didn’t only play D&D, after all). Come to that, I’d already designed board games and a computer game ⁴ that used different mechanisms. Overall, though, the classic approach won on the basis of simplicity. When you’re in a fight, you want an instant readout on how well you’re doing. Anything too complicated wouldn’t give players enough time to absorb it; anything much simpler and combat wouldn’t have been any fun.

We decided, therefore, to give characters a number of “lives.” Whenever a character was hit successfully, one life would be removed. To make this less predictable than it sounds, we scaled up everything by a factor of ten and added some randomness. Thus, although your character could usually survive ten blows, sometimes they’d hold out for longer and sometimes they’d capitulate sooner. We did some tuning to make sure that battles neither ground on too slowly nor ended too quickly to be fun, and that was that. Everything else—spells, weapons, artifacts—followed.

As a game mechanic, this worked well. As a representation of reality, it was somewhat off the mark. Real-life combat with hand weapons is not a gradual wearing down, a cut here, a nick there, until eventually someone dies of attrition. Instead, most blows are blocked, dodged, or parried, and when one does finally land it either has little effect or it effectively incapacitates the recipient. In real-world medieval warfare, once someone suffered a telling wound they could take no further part in the battle and were fairly certain to die even if no one finished them off (medical science couldn’t deal with infection). Virtual worlds set in different periods or having some kind of magic available in them don’t have to stick with these facts, of course, but they still need a fiction to rationalize it.

So what other combat systems are available that retain the gameplay but make more rational sense?

Enhanced Combat Systems

Actually, the MUD1-style approach does make some sense if, instead of stamina/health/ hit points, armor/shield is used. You bash at people until their armor comes to pieces, then the next blow hits their body and they’ve lost. Where characters’ injuries would heal with rest after a fight, “wounded” armor is repaired.

Of course, to be “realistic” some hits should possibly get through despite the armor, but even so they might not do fatal damage, yet fatigue can set in and…blah blah blah.

Combat systems can rapidly get very complex. However, the same approaches crop up time and time again, not only in virtual worlds but also in single-player and tabletop games. Here’s a run-down:

• Reaction and recovery times. Traditional combat is round-based, in that characters trade blows either in turn or simultaneously. This is unrealistic, because some weapons take longer to recover from than others. In the time it takes you to raise your two-handed sword, I might have been able to run you through twice with my rapier. Defense also can have reaction and recovery times: Your heavy shield may take more damage, but if I switch the point of my attack you might not be able to bring it into play before I strike.

• One-off attacks. Instead of having fights proceed automatically after the first blow has been struck, each one must be performed manually. If I throw my axe at you, I can hardly do so round after round anyway—it’s a one-off attack. Special boosted attacks (such as stunning punches or called shots) are usually treated as one-offs.

• Stances. These are a crude mechanism for trading offense against defense. If you want to hold out until your allies come, you increase your defense but pay for it with a reduced offense. If you want a fight to be over quickly and can take excess punishment, you raise your offense ¹ but your defense suffers as a result.

• Hit location. The damage you do depends on what you hit. A blow to the leg is not as bad as one to the head. This brings in the possibility of having hit points for individual body parts instead of one value for the whole character. Where a blow lands depends on the weapon, the characters’ skills, the relative sizes of the characters, the relative sizes of the body parts, and (the strategy bit) where the player aimed.

• Damage types. The theory here is that different weapons do different kinds of damage, to which different armors (or, combined with a hit location system, different parts of the body) are differently susceptible. Heads can take less crushing damage (for example, from a mace) than they can slashing damage (for example, from a longsword). If you’re armed with a mace, you should therefore target the head rather than, say, the midriff. Characters may have special quirks that make particular blow type/location combinations unusually effective (for example, a “glass jaw”).

• Wound levels. This approach gets away from the hit points idea entirely. Wounds are instead rated by category—for example light, medium, heavy, disabling, destructive. They compound, so several light wounds make a medium wound, and so on. Heavy wounds may lead to some loss of functionality, and disabling wounds will lead to an entire loss of it. Destructive wounds disable permanently. Thus at light level your hand may be cut, at medium it may be gashed, at heavy it may be fractured (you couldn’t write with it, but you can still hold a sword), at disabling it may be smashed (you can’t do anything with it), and at destructive level it might be severed. Different weapons have different wound profiles for different body parts. A mace may have a greater chance of disabling a head than does a sword, but a sword has a (small) chance to decapitate.

• Critical hits. Critical hits incorporate hit location and wound levels into a hit points system. Combat proceeds in the traditional “death of a thousand cuts’ style,” but if a character hits well enough (maybe one time in 20, depending on the location), they do special damage (usually equivalent to a disabling or destructive wound). Opponents who weren’t even breaking into a sweat can suddenly be knocked unconscious by a well-aimed blow. Thus, players who are losing a fight might stay rather than flee, in the hope that they can pull out a critical hit and win.

• Afflictions. These are temporary side effects from blows/ spells that impact on a character’s ability to conduct combat. If combatants accumulate enough of these, it makes taking them down a lot easier. A few knocks on the head won’t necessarily do much damage in themselves, but if they make you dizzy then that could reduce your ability to parry something more serious.

• Combat scripts. This approach owes much to pencil-and-paper role-playing games such as En Garde! ². Combat commands are entered into a queue, and executed in order simultaneously for each combatant. Combinations of commands can be prewritten and hot-keyed, so that players can have time to think rather than relying on first-person shooter twitch skills. If you have a script effective against a particular kind of opponent, you can call it up instead of your default one. Of course, if you don’t but they have one effective against you, it could be problematical. This “program your battle” approach was adopted by Star Wars Galaxies, so it’s likely to become more popular as clones of this game appear ³.

All these ideas can be (and often are) incorporated into gloriously complex wholes. The level of detail can become so great that the rest of the virtual world pales by comparison. Few virtual worlds with an immensely complicated combat system avoid being completely dominated by it—Achaea is the best-known exception.

Although these combat systems are intended for melee, they can be used for other forms of combat, too. In particular, if characters control or are crewmembers of fighting vessels (spacecraft, pirate ships, mechs, and so on) then similar mechanisms can apply. “Number One: Take out their hyperdrive.”

As I mentioned earlier, whatever combat system is adopted for hand-to-hand combat underpins all other forms of combat. If you’re using a hit points system for sword fighting, then arrows will also cause characters hit point loss; if you’re using a wound level system, then damage-causing combat spells will have to use it too. These secondary combat activities come with their own issues, of course (arguments over the range and accuracy of crossbows versus longbows being particularly prone to starting flame wars). Weapon statistics aside, though, combat systems face in general a common set of difficulties of which designers should be aware.

Problems with Combat

There are many social problems with combat, which I shall come to shortly. First, though, I’ll briefly address some purely practical concerns.

• Equalizer weapons. Anyone who can flip off a safety catch and pull a trigger can kill someone else. Highly trained individuals have a better chance of success, but even complete novices—children playing with daddy’s pistol—are in with a good chance. Equalizer weapons brook no defense except a physical barrier (a good shot is no better at avoiding bullets than is a poor shot) and the damage they do tends heavily towards the disabling and destructive levels. Unless a convincing fiction for why this isn’t so can be produced ¹, equalizer weapons take a lot of the fun out of (combat-oriented) virtual worlds.

• Multiple weapon types. You might find that swords work best with a hit points system, but muskets work best using a wounds-based approach. Which do you use? What do you do when someone using a sword attacks someone who is using a musket?

• Over-robust defense. How long does it take for a newbie to kill an old-timer who isn’t putting up a defense? Typically, it will take much longer than it ought to. The unarmed veteran stands yawning while the newbie repeatedly swings their sword; when they hit, the damage done is merely a fraction of that necessary for victory. It would be like chopping down a tree if it weren’t for the fact that you frequently miss the target completely.

• Concurrent combat. How many creatures can you fight at once? Do you get the same number of attacks against each one, or do you have to share them out. Do you defend equally against all of them, or against some more than others? How do you stop people from ganging up in large numbers and beating the stuffing out of their foes?

• Targeting. Do I aim at an individual or at a point in space? As pointed out in Chapter 2, “How to Make Virtual Worlds,” there are implementational considerations here. There’s also the issue of consistency: If I aim an arrow at a point rather than at a character, then the same solution should be used for hand-to-hand combat. What about semi-autonomous weapons, though—for example heat-seeking missiles? Are both relative and absolute targeting systems needed? Who is hit if someone fires a bullet into a crowd?

• Area effects. If I throw a grenade or cast a fireball spell, the result affects everyone in an area. The whole point of such an attack is to avoid the targeting issue. What happens, then, if there is someone in the area who is on your side? Can you be blown up by your own bomb? This is mainly a problem for graphical worlds, because textual ones don’t commit to exactly where anyone is standing.

• Twitch. First-person shooter games rely on the skill of the players. There are two components to this: experience and twitch. Experience comes from having done something often enough that you know what to expect. Twitch comes from having fast reactions, good manual dexterity, and reasonable vision. Both experience and twitch transfer to virtual worlds, but neither ought strictly to be there: The knowledge and speed of a character shouldn’t be that of the player. Twitch is the more annoying ², especially when coupled with client hacks or go-faster macros that can make someone almost unbeatable. Where should you put the threshold, though? How many seconds is it reasonable to make people wait before you process their next combat command? What effect will lag have on this?

• Combat skills. My character can do things I can’t. What are these things, and how do they affect combat? If there are too many or too few, it could send players the wrong message. What can I do while I’m waiting for the next combat round? Will it occur instead of or in addition to my attempt to hit my enemy?

• Disengaging. When fights are handled automatically, under what conditions do they end? Can I simply run away? Do you get to hack at me while I do? Can I surrender or be captured? If so, how could I escape and why wouldn’t you just kill me anyway? What happens if I’m stunned, but not actually dead? Does the fight end there or does it wait until you strike the killing blow?

There are, of course, many other tedious-but-important decisions that designers have to make about combat. If the planned system is deep and complicated, these can take up disproportionate amounts of time to resolve—not because of complex interactions between components of the system, but because designers can become so involved that they try to treat what are essentially subjective choices objectively.

The main thing is to be able to step back, look at a combat system, and ask three things:

• Will it work?

• Will it keep players immersed?

• Will it be fun?

If the answer to all these questions is “yes,” go with it.

There is a fourth question we could ask, which on the face of it looks like it follows from the preceding three. It has deeper implications of its own, however.

Will players accept it?

To find out why this is important, we need to consider two notions: opposition and consequences.

Opposition

Conflict concerns opposition. Combat is the most obvious manifestation of this, but it occurs throughout virtual worlds. Two people wanting to get through the same turnstile at once are in conflict; two people aiming for the same job are in conflict; two merchants pitching to the same potential customers are in conflict. Combat may be one way to resolve this conflict, but others tend to work better.

In virtual worlds, conflict is typically categorized in terms of opposition: Who is in conflict with whom? There are three main possibilities:

• Player versus Environment (PvE). Players are opposed by the environment—that is, the virtual world. In a combat situation, this means player characters (PCs) fight monsters.

• Player versus Player (PvP). Players are opposed by other players. In a combat situation, this means PCs can fight each other.

• Group versus Group (GvG). Players are members of groups that are in conflict with other groups. In a combat situation, this means PCs can fight any PCs who are members of enemy groups but not those who are members of their own (or a neutral) group ¹.

PvP and GvG both assume PvE. GvG is a restricted version of PvP that will usually operate within other limits too, for reasons that we shall come to shortly.

The difference between PvE and PvP is central to virtual worlds. In a strictly PvE environment, PCs can do nothing to other PCs. In a strictly PvP environment, PCs can do to other PCs whatever they could do to NPCs or monsters—and potentially a lot more.

Why the distinction? The answer is, fundamentally, consensuality: Players don’t want other players doing things to their character without permission. It’s one thing for a trident-wielding NPC mermaid to attack you, but another thing entirely for a PC to do it. Hard role-playing virtual worlds solve this almost trivially: If a player wants to do something to another player’s character, they ask on an out-of-character channel first. Other virtual worlds don’t have this luxury.

In a pure PvP world, there is no consensuality: People can try to mess you about, whether or not you like it; they may run the risk of being attacked by you, your friends, and perhaps the local police force too as a result, but they can still try. In a PvE world, the world itself prevents anyone from messing you about in the first place (and you from messing them about).

In practice, absolute PvE is untenable. If I want to pick up some treasure lying on the ground and you get to it first, you have interfered with my enjoyment of the virtual world and I am entitled to feel peeved as a result (even more so if you did it deliberately just to annoy me). Weak forms of PvP such as this (looting) are always present in multi-player environments, otherwise there would be no way of knowing they were multi-player. The question is not whether to have PvP at all, but how much PvP to have?

The issue of consent becomes blurred here. Plainly, players have to put up with some degree of interference from other players merely by virtue of sharing the virtual world with them; in this sense, entering the virtual world can be regarded as offering implicit consent that a player is willing to accept the PvP it supports. Unfortunately for designers, players will still complain if they are the victims of what they see as a violation of their character, even if they knew it was a possibility when they signed up.

The problems are compounded by the fact that different players have different ideas of what’s okay and what’s not okay. Some players may think it’s fine that characters can’t walk through each other, because that’s how reality works; others may dislike it because it means access to locations through pinch-points (for example, doors) can be blocked. Then again, some of these people may relax their point of view in cases where 50 characters have just killed a monster then some non-combatant with maxed-out running skills races through the mob and steals whatever reward was dropped. Is a reasonable compromise solution to allow characters to move through other characters but at a greatly reduced speed? A gang could still totally surround someone and continually move around them so that wherever they were headed there would always be people in the way.

Some abuses of PvP can be stopped by giving people switches they can toggle. Generally, players are fine with allowing strangers to speak to them, but not if this is used as an offensive weapon (for example, spamming). Thus, some ability to squelch or gag incoming messages from other characters is often incorporated into virtual worlds. This works, but it’s not always realistic (if your character is asleep, then why would one person’s voice wake them up but not someone else’s?). Stopping people from spamming in the first place is a better answer, but one harder to implement.

This idea can be extended to a system of permission levels. Every character has a disposition to every other character, ranging from “they don’t exist as far as I’m concerned” to “they can mind-control me into doing anything they want.” The default for meeting a complete stranger might perhaps be to allow communication but no physical contact. Permissions can be changed either explicitly (using a “change permission” command) or implicitly (if I offer you my hand to shake, I am implicitly giving you permission to make formal physical contact with me). If someone tries to act beyond the permission level you have set for them, you get the option to turn them down (“Gilead declines to shake your hand”), to allow the individual action, or to raise the permission level (“You shake hands with Gilead”).

The “they don’t exist” permission level raises an interesting question: If virtual hearing can be selectively filtered, what about other virtual senses? One very good way to annoy a player is to follow their character around. In a strongly PvE world, there’s no way to stop it except by running faster; it feels like an invasion of privacy (which of course it is), but the virtual world engine can’t distinguish between friends, stalkers, and strangers who just happen to be heading the same direction. Giving players the ability to squelch the images of other characters would sort the problem out, but again it stresses the fiction.

Fiction is the main enemy of PvE. If you can do something to NPCs, why (in the context of the virtual world) can’t you do it to PCs too? If I create a rogue-class character with high thieving skills, why can I apply them to any NPC from the lowest to the highest, but not to a single PC? The real answer is that NPCs won’t complain but PCs will, but in the virtual world NPCs and PCs are supposed to be the same thing (characters). Although it’s possible to make a formal, fictional distinction between them (for example, NPCs are “normal people” and PCs are somehow “gifted by the gods”), most virtual worlds that impose this kind of restraint don’t bother; it’s justified by its existence. If players moan that this is unrealistic, well, okay, so do they want to hold a vote on whether to allow PvP?

The extent to which PvP is permitted in virtual worlds is historically defined by three factors:

• How much freedom to be pains in the asses the prevailing culture of the majority of players allows.

• What the players say they want.

• What the community management staff can handle.

These are in increasing order of refinement. The real-world culture of players gives them notions of personal liberty in some kind of balance with the rights of other people; players are prepared, in principle, to accept these as the default for the virtual world. Virtual world culture is unlike real world culture, though, because of the anonymity that virtual worlds afford. Some behaviors that would be unwise in the real world because of the repercussions are “who cares?” issues in virtual worlds.

Therefore, the players will suggest changes. If this is the first virtual world they’ve played, they’ll immediately think of the most obvious (but not necessarily the best) solutions; if it’s everyone’s first virtual world, everyone will think of these same solutions and it will become a tidal wave of opinion that the live team will find hard to resist. The changes will be added—some for better, some for worse—and tinkering will take place until some kind of consensus is reached.

The tinkering will stop when the noise from players is down to sufficient levels that the live team can cope with (hopefully before the end of the open beta). Players will occasionally get into conflict, but not so frequently that the community managers can’t sort them out. This degree of PvP becomes the standard for the virtual world; designers of later virtual worlds will consider it as a prototype (or even a paradigm) for the “prevailing culture” of PvP/PvE with which they begin.

This would all be fine except for one thing: Virtual worlds with more PvP are manifestly more exciting than virtual worlds without it. There are two reasons for this: Players are more intelligent than mobiles; players have feelings and mobiles don’t. If you’re in conflict with an intelligent opponent, it makes success and failure more meaningful; if you win or lose to an opponent with emotions, your own emotions are affected.

PvP is “me against you;” PvE is “me against random number generator.” Which is the more thrilling?

Recognizing this, strongly PvE virtual worlds have tried to developed ways to have the good bits of PvP (the excitement) without the bad bits (lack of consensuality). There are basically two ways to do it: arenas and character flags ². With arenas, different locations of the virtual world have different PvP rules. To enter a location where the rules are different from the norm, you must give your consent. This will either be explicit (you won’t get into the gladiator pits unless you confirm you understand what could happen) or implicit (you should expect to be robbed if you walk into the thieves’ guildhouse). Arenas don’t have to be small: Huge swathes of Dark Age of Camelot and Shadowbane are effectively arenas, for example. With character flags, players elect to be PvP-able, and can therefore interact in PvP ways with any other players who have flagged their characters thus.

Neither of these solutions really works. One problem is that there’s no tangible reason to do it: Why accept full-blown PvP when you’re no better off than those who don’t? Sure, you get the extra intangible content that PvP delivers, but while PvPers are taking kit from each other PvEers are getting new kit from monsters with less effort. If everyone else has it easy, why not have it easy yourself?

The big virtual worlds tried to counter this attitude by devoting whole incarnations (shards) to PvP, so every character you met could be PvPed. This didn’t help, though, because players still advanced faster in the strongly PvE incarnations. PvPers felt they were taking on more opponents and being penalized for it. They might have been happier if they were compensated somehow—for example, by a higher level cap (even if it was only one level higher than a PvEer could get); the PvEers wouldn’t have accepted this, though. Besides, a virtual world designed for PvP has different requirements to one designed for PvE; they’re similar, but by no means identical ³.

This is where GvG comes in. The idea here is that the virtual world is divided into sections (the “realms” of RvR), within which “normal” PvE rules apply. There is a further section to which these are connected, however, in which PvP rules are in force (except between characters hailing from the same realm). There is typically a reward for combat success in this area, which means that characters who enter it do get an advantage over those who don’t. Entry into the PvP arena isn’t mandatory—you don’t have to do it—but you’ll benefit in terms of increased stats or whatever if you do. The fact that you can enter with a bunch of guaranteed-trustworthy friends and stay there for only as long as you want makes it far more palatable.

The term RvR comes from Dark Age of Camelot, but it’s not the only virtual world to use this approach; indeed, it’s not the only big, graphical world to do so. Anarchy Online has characters divided into three groups, with people meeting in PvP areas for combat. Lineage has clan-like groups called bloodpledges ⁴, which can conquer castles from one another in (scheduled) sieges; success here has material results, in that owners of castles get tax income they can invest in preparing for the next siege. Shadowbane is trying something similar for western audiences with its guild versus guild mechanic: PCs can build cities, rule countries, lead armies against other countries, and so on.

GvG does add some of the features that PvP has but PvE lacks. It still misses out on an overall point, though, in that in most GvG virtual worlds it’s impossible for any side actually to be wiped out. In DAoC, for example, Midgard is never going to conquer Albion or Hibernia. This is for various reasons, the main two being content and balance.

Preventing defeat to keep content is understandable. If a virtual world has been advertising itself as having three realms, if strategy guides are out in the shops describing the three realms, if each realm has distinct styles of architecture and dress that took months to create, then yes, one realm’s annihilation of another would be a mite inconvenient.

Balance is not a good reason to keep realms from beating on one another too much. Strictly speaking, it doesn’t matter if one side completely conquers another as long as the resulting empire can split through civil war ⁵. If for content reasons you don’t want realms to be obliterated, though, the aim should still be equilibrium rather than balance. It doesn’t matter if one side can win some of another side’s home territory, so long as it becomes increasingly harder to do so the more they advance. It’s like two fencers dueling with bungee ropes tied to their waists: There comes a point when the elastic pulling on the better fencer is so strong that the worse fencer can win and push them back. They don’t even have to have the same strength elastic—equilibrium will eventually be reached.

A problem with GvG combat is what to do if one or more sides give up and don’t get involved in it. Does the virtual world save them, by not allowing enemies to pass a frontier? Does the live team save them, stepping in to avoid putting all that beautiful elf artwork to waste? Or does the opposition sweep through in an orgy of destruction and pillage?

The main problem with all these PvE solutions to PvP is that there are two levels of consensuality at work here: Consensuality in principle, and consensuality in fact. The former is when you accept that something may happen to you, but you’re not necessarily going to like it when it does happen and it had better not happen too often; the latter says that you expect something will happen to you and therefore won’t be too bothered when it does. Part of PvP’s appeal comes from the assumption that a foe has accepted the principle but won’t easily accept the fact unless they win; this is equally true for attackers and defenders—neither will be pleased if they lose, but both will be over the moon if they win. In PvE-based PvP solutions, consensuality is by fact; in full-blown PvP, it’s by principle. Unless players are prepared to accept the principle, they’ll have to make do with the fact.

Unfortunately, they may accept the principle but still have a justifiable grievance. This is best illustrated by what has come to be known as Marian’s tailor problem ⁶. If a rich virtual world offers characters careers in both non-adventuring and adventuring, how are the non-adventurers protected from the adventurers? Marian’s character is a tailor because Marian likes creating clothes for other characters and the opportunities for interaction that this affords. She understands that there are other players who prefer hacking monsters (and each other) to pieces with sharp weapons. She understands and accepts that other players may on occasion attack her. What she doesn’t accept is that they can do so indiscriminately, on a whim. She is completely unable to defend herself, so will always lose. Is her only recourse to take up arms? If so, what’s the point of putting in tailoring as a career if everyone is eventually forced to become a fighter anyway?

The issue of opposition (PvE, PvP, GvG) is wider than combat, but combat nevertheless is the litmus test that decides whether a virtual world “is” PvE or PvP. If you can attack other player characters, it’s PvP; if you can’t, it’s PvE. There may be other PvP elements—for example, casting nasty spells on characters, or nice ones on the mobiles they’re fighting—but combat is the issue that tips the scales. Who you fight, however, is only part of the issue. The arguments really begin in considering what happens if you lose.

Consequences

Different locations have different risks associated with them. A PvP arena, for example, has a greater risk associated with it than a marketplace; a dragon’s lair comes with a greater risk than, raccoon’s.

Players like to know their risks in advance. They like to be able to tell by looking whether or not they can beat a potential foe in combat. Some virtual worlds helpfully color-code PCs and mobiles to make it easier to check them out. Although in real life you might not be able to tell a martial arts expert from anyone else merely by looking, in virtual worlds you often can. If there’s no fiction explaining why you might be able to do so, well, you still probably can.

Risk makes things exciting, but it’s only part of the equation. Equally important are the stake (what you stand to lose) and the prize (what you stand to win).

In virtual world combat, what you stand to win can be anything the designer chooses. There are constraints imposed by other areas of the design (for example, the economy) and by the combat system itself (high-risk or high-stake fights should have high-stake rewards), but this still leaves a lot of leeway.

So, what do you stand to lose?

This is harder, because it must be something that you already have. It should also be commensurate with the risk and what you stand to win. It’s okay to favor the player statistically, because otherwise characters will not advance. In real life, if you stand to win $10 on the toss of a coin, then $5 is a fair stake. In virtual world combat, it’s like you get four tosses of the coin to win $10 for a $5 stake—the odds are stacked so you’re generally going to win. However, (in this example) one time in 16 you’re going to lose. If it’s only $5, who cares? The problem with virtual world combat is the size of the stake.

Fights can have three outcomes: victory, inconclusiveness, defeat. Most fights with mobiles end in victory; most fights with characters (where PvP is allowed) end inconclusively; a very small number of fights against whatever opponent end in defeat. Defeat can mean many things—capture, injury, loss of possessions, loss of prestige, loss of time, and so on. For mobiles it means death. Does it mean death for PCs?

As with PvP, character death is not entirely a combat issue: It can occur in other ways, too. In the real world, most people die through disease, old age, or in accidents. The number one cause of peacetime death among British men under the age of 35 is suicide ¹. Death happens to everyone in the real world, combat or no combat.

Virtual worlds, however, are not real ones. Characters don’t “have to die.”

Old age in virtual worlds can traditionally be eliminated without the need for any explanation because only very long-term characters would ever qualify for it. Disease and illness, while okay as a temporary inconvenience ², also don’t have to be anything more than that; this is just as well, because players would seethe with rage if their characters were merely confined to bed with influenza, let alone if they were dying of something incurable.

This leaves death by accident (deliberate or otherwise) and death by being killed.

In virtual worlds, characters are in theory physically capable of doing acts that would, in the real world, result in their death. They can jump off cliffs, stand in bonfires, swallow plutonium, throw a grenade short of its blast radius…. The virtual world can of course be crafted so as to prevent such events from ever occurring, albeit at a cost in realism. If someone in full armor is paddling a coracle across a wide river and they decide to light a fire, they really should expect to drown. Nevertheless, this can be prevented—for example, by not making coracles ignite while people are in them. It can be programmed round.

Death by being killed can’t be programmed round. You stick a mobile full of holes and it dies; a mobile sticks your character full of holes and it also dies—it wouldn’t make sense otherwise. Is it the same kind of death, though?

When a mobile dies, it ceases to exist. It is permanently gone. Some reinitialized version may appear at its spawning site after a period, or when the virtual world resets; it’s not the same mobile, though—it has none of the experience or possessions it had when it died. At best, it’s a clone.

Never-to-return death is called permanent death, or PD ³. It’s the single most controversial subject in virtual worlds.

Permanent Death

PD is not the same as PvP. PvP means that one PC can do things to another PC that the latter didn’t agree to; PD means that if a PC suffers death, it is consigned to oblivion. When PvP and PD are combined the result is player killing (PKing): One PC can directly cause the PD of another PC ¹. Players often use PD and PvP as euphemisms for PKing, but the distinction is there and is very important. Many of the benefits that advocates of PKing cite are primarily due to PD; some of the strongest objections to PKing are due to its PvP element, rather than to PD.

Like PvP, PD has consent issues. These are not, however, with other players; rather, they are with the virtual world itself. As our Chapter 4 analysis of virtual world physics showed, players are normally quite happy for the virtual world to model the real one, because it reinforces their immersive experience. In the real world, though, none of a virtual world’s players have had the experience of dying.

Because the whole aim of virtual worlds is to bring the character and the player together as a persona, players will routinely consider their character to be themselves. On the face of it, then, their character’s PD would thus be like their own death in reality. In practice, players at the persona level of immersion don’t feel anywhere near as bad as players who are approaching it, because they have the duality of player/character sorted out. The character has died, but the player is the character, and the player (clearly) hasn’t died, therefore the character hasn’t either. The character can be reborn in a new manifestation, much as a PD mobile respawns from scratch. It’s not a pleasant experience—it’s like losing three months’ work while trying to back up your computer’s hard drive ²; frustrating, but not the end of the world. Players who are near to total immersion but aren’t quite there will suffer the most, because not only do they lose the work, but they also feel they’ve lost a dear friend. The word “gutted” is frequently used to describe the experience, but even that doesn’t come close to how bad it can feel. Unlike the death of a real person, though, virtual PD grief can be overcome relatively quickly. Players usually can rationalize what has happened and come to terms with it after giving it some thought; the emotional impact is strong, but at an intellectual level all that’s gone is a chunk of over-anthropomorphized computer data. If they can cope with the emotional hit, the intellectual gain can more than make up for the loss. If they can cope with the emotional hit….

There is a quantitative difference between the emotion felt at losing a character to the virtual world and that of losing it to another PC. In the former case, players will be incredibly annoyed but will rapidly conclude that it was either their own fault or the result of incredible bad luck ³. In the latter case, though, they’ll be angry, too ⁴. For long-term PKers (as opposed to newbies), if they started the fight and got killed, then their self-image will fall even lower than it was before. For the victim of a PKer (or, worse, a pack of them), there is the sense of unfairness that the heavily armed attacker will have selected a victim on the basis of their likelihood of winning, and therefore the odds were stacked from the beginning. Unprovoked attacks cause enough indignation to blow an industrial indignationometer.

Consent is an issue. Emotional attachment is an issue. Wasted effort is an issue. None, however, are the issue. The issue is that people don’t like losing. PD amounts to a statement of total, that’s-all-folks loss. There’s no wheedling out of it; it’s final. That’s what people dislike about it. Whether this is merely a passing phase due to virtual worlds’ comparative immaturity or a permanent situation rooted deep in the human psyche we shall examine presently.

For the moment, the important thing to come out with from all this is the sense that PD is not popular. People who fear it are as much—if not more—against it than people to whose characters it has actually happened. Returning to the gambling metaphor, if PD is the potential result of losing a fight, players are wagering a very, very high stake.

Such a stake is constant across all combat, because in a PD world it’s the only thing a player can bet. Because of this, the risks of losing must be small or the rewards for winning must be high (or both). Unfortunately, this can make matters even worse because it means PD actually occurs only very infrequently. Players aren’t expecting it to happen, therefore when it does, there’s the shock to contend with, too.

In real-life gambling, if a risk looks too great you can usually pull out of the bet and lose only a fraction of your stake. In virtual worlds, it should always be possible to cut your losses and flee from a fight. There may be tangible consequences for the partial defeat, but the main loss will probably be your pride. Only if you decided to stay (for example, because your opponent looks like they’re going to die sooner than you are) would you risk losing your whole stake.

PD is not appropriate for all virtual worlds. It would be ridiculous in most non-game contexts, and doesn’t add greatly to games without combat, either. For virtual worlds where non-PCs can (and regularly do) suffer PD, though, it has serious contributions to make.

Approaches to Permanent Death

Let’s assume for the moment that a designer has decided they want permanent death to feature in their virtual world. The reasons why they might (or might not) make such a decision we’ll go into shortly, but for the moment we’ll suppose they’ve done it. What are the various ways it can be incorporated into a design?

It all depends on how you intend to ensure that a death is “fair.” Here are the classic ways to do it (most of which also can be applied to PvP):

• Unmoderated. If your character dies, that’s it. There’s no comeback: It’s gone. PK versions of such worlds will frequently descend into barbaric slashfests, although this is not inevitable: If fleeing is not unlikely to succeed and the player base is too small to support gang warfare, an unmoderated world can thrive (and with a vibrancy it’s hard to match elsewhere).

• Post-moderated. PCs can appeal to administrators to be resurrected if they feel they were harshly treated. The classic reason is because of communications difficulties (lag, line noise, cosmic rays, whatever) that prevented a character from fleeing at a critical moment. The administrators will look at logs to determine whether the player has a case. In worlds with PKing, there may be a justification system in place to ensure that high-level characters don’t wipe out low-level ones indiscriminately, but they can nevertheless avenge intolerable acts (for example, you stole my stuff). Again, administrators would normally enforce this on appeal. This approach also works, but is only sound in small-scale virtual worlds that can log absolutely everything (that is, textual ones).

• World-moderated. The idea here is to implement a justification system by tracking player actions. Unfortunately, it can only ever track tangible actions: If someone insults your ancestry and you hit them, the virtual world will flag you as the aggressor. There are several ways to have a virtual world moderate PvP and PKing, none of which ever really work: Karma systems and NPC justice systems are the most popular. Grudge systems are the most interesting, because they allow slights to be traded: If I hit you, you get a grudge against me that you might sell to someone else who can beat five kinds of Hades out of me. This makes people think twice about performing unprovoked attacks, but has the disadvantage that it’s blatantly unrealistic.

• Player-moderated. These approaches don’t work either, at least not yet. Here, the decision as to whether a character has done wrong is made by players, not by the virtual world (although the world will act on what the players have said—for example, by making NPC shopkeepers charge more to people with low reputations). In player justice systems, there may or may not be NPC enforcers available to make sure that justice is done. The problem with player-moderated systems is that they are too easily undermined by groups of players acting in concert. If 100 players say that some random other player is a thief, that player is going to have a problem clearing their name. There are experiments under way to circumscribe this problem. Sure, it works for eBay, but eBay doesn’t have 10% of its user base trying to screw it over for fun.

• Opt-in. This is the classic response of designers who disabled PD or PvP and then came to regret it. They re-enable it, but only in certain arenas or between consenting duelists. It fails to deliver any of the major benefits of either PD or PvP. If your virtual world has no PvP or PD, you’ve already handed it over to people who don’t want it; adding it is not going to make them happy, nor will it attract people who do want it.

• Opt-out. This is an approach that ought to work but which has yet to find favor. The idea is that by default PD and/or PvP applies (moderated using one of the other approaches) but characters can flick a “pacifist switch” to opt out—some kind of “holy person” fiction covers for it. They pay for this security by having a lower level cap than characters for whom the gloves are off. While this seems fair to non-pacifists, it’s deeply unpopular among pacifists: They want the benefits of not getting killed but don’t agree that they should give up anything to do so. If higher level content is available, they should have access to it.

• Inevitable, but distant. Characters have a shelf-life, after which they die of old age, retire, or otherwise cease to exist except as a memory. This form of PD serves only one purpose: To force people to start afresh instead of staying on with the same, stale character indefinitely. Unfortunately, as mentioned earlier, even at accelerated timescales it would probably take several real years for a character to “die,” and guaranteed impending death would dishearten everyone—even newbies.

The preceding is the main approaches to PD. They all treat dead PCs as “gone forever,” resurrection being limited to external factors such as bugs, rule violations, and the occasional act of gross favoritism by members of the live team.

Most virtual worlds don’t go for PD, however; this is fairly clear from the fact that it’s PD that has the acronym, not “NPD” (which is just “death”). In non-PD virtual worlds, PCs do die in combat (and perhaps in other ways) but it’s not the same kind of death that the cookie-cutter NPCs suffer. What exactly it is instead, however, varies.

Alternatives to Permanent Death

Let’s assume now that a designer ¹ has decided they don’t want permanent death to feature in their virtual world. If there’s no NPC death either (because of world’s genre or application), that’s it, end of story, skip to the next section.

If there is NPC or (in most cases) monster death, the designer has to find some non-PD equivalent for PCs. If your character fights a monster and loses in the worst possible way, what happens to it?

Okay, well the big mistake is to say the character dies. It doesn’t die. If it did die, that would be PD ². One of the absolute, defining features of death is that the dead don’t recover from it ³. Whatever it is that happens to the character, if it can continue its existence afterward, then it hasn’t died.

The words “kill” and “death” have all kinds of powerful emotional connotations that often in no way resemble how the concept is handled in a non-PD world. This isn’t necessarily a problem, but it becomes one when designers transfer too many of those connotations to something that doesn’t merit them. Sadly, this is something they are wont to do when discussing alternatives to PD.

Merely labeling something as “death” doesn’t make it a serious matter. In many cases (particularly in large-scale graphical worlds), the penalties for losing a fight are so lax that even the word “hurt” would overstate them. When I act as a consultant to in-development virtual worlds that don’t have PD, I insist that the designers don’t use the word “death” or “kill” in relation to PCs. This concentrates the mind on the mechanics and not the ideal. If someone says “when a mobile kills a PC,” I’ll correct them based on how they plan on handling it; unhappily, more often than not this means “when a mobile wuss-slaps a PC.” It may be sarcastic, but it leaves them in no doubt as to what they’re actually talking about, rather than what they thought they were talking about.

Okay, so let’s look at the usual ways that non-PD virtual worlds handle what they call “death” but isn’t:

• Resurrection. PCs that die come back to life, perhaps in a worse condition than they were in at the point of death, perhaps not. As with most forms of death evasion, in PvP worlds the exact penalties may vary depending on whether the PC was the aggressor.

• Insurance. This is a form of resurrection that PCs pay for in advance. It’s popular among Science Fiction worlds such as Federation II, but can work in Fantasy ones too. In DragonsGate, for example, players can accumulate “favors” by doing quests “for the gods,” which can be translated into resurrections. Sometimes, insurance works on a “save game” principle, whereby resurrection is only to the state the character was at when it was last insured; this means characters will regularly pay for new insurance, thus giving the economy a healthy cash sink.

• Ghosts and spirits. The fiction here is that PCs’ bodies die, but their spirit lives on. The spirit is usually separated from the body (appearing at some local last-save point or generic shrine) and has to find it to affect a resurrection. In the meantime, other PCs or mobiles can loot the body and steal the “dead” PC’s gear. Losing a fight therefore means a time penalty (you have to find the body) and possible tangible loss (if people loot it).

• Realm of the dead. This is a variation on the ghosts and spirits approach. “Dead” characters appear in some other dimension (that is, they’re teleported to a set of locations disconnected from the main part of the virtual world). They have to find/fight their way out, whereupon their soul returns to their body and they’re back to life again. Their body has taken the opportunity to recover from its massive system failure, too. Again, it’s mainly a time penalty to stop people who have just been “killed” in a battle from immediately wading back into the fray.

• Left for dead. Although PCs are clever enough to know when a mobile is faking death, mobiles are universally stupid. They think that the moment a PC hits the deck it’s Game Over. However, the PC is only stunned or unconscious or winded or otherwise wuss-slapped, and will be right as rain in no time!

• It wasn’t a fight to the death. This is absolutely fine in PvP situations, where it doesn’t really stretch the imagination too much to envisage duelers or gladiators who honorably fail to deliver the final death blow of a fight. It makes no sense against mobiles though unless the mobiles also accrue the benefits of it.

• Inheritance. Lefty gets killed. He really is PD gone. However, he happened to have a relative to whom ownership of all his kit passes. The relative is very close in looks, skills, and experience to the character who just died. Welcome, Lefty II! This works best in materialistic virtual worlds that define characters by their possessions more than by their attributes. The character may be nominally dead, but the player is no worse off for it. This is perhaps the best of these alternatives to PD because it offers some opportunity for players to start new characters when their old ones are wuss-slapped. Instead of Lefty III, you might go for Leftina instead (not that many players do—they treat it basically as resurrection). The fiction is basically sound, too, although it’s invariably stretched so thin as to be transparent.

• No believable fictional explanation. Characters who “die” bounce back with whatever time, movement, kit, experience points, and skills that the designers decide. There’s no sound rationalization, they just do. Asheron’s Call 2 with its progressive attitude that penalties for death in virtual worlds are “outdated” is the jaw-dropping epitome of this approach.

All these are alternatives to PD. However, the concept of actual PD is surprisingly incomprehensible to many players, who can’t quite bring themselves to believe that there are virtual worlds in which death really does mean death; occasionally, therefore, some of the preceding will be described as being PD even though they’re not. I’ve read reviews that described an insurance system as being PD, presumably on the grounds that your character is actually not there for a short period before rematerializing.

There also seems to be a good deal of denial among designers of virtual worlds. I’ve sat in on meetings where designers agreed among themselves to go with a PD system, then started to think of ways they could get around it (“Maybe we could have like a soul ring, that stores a character’s essence and…”).

As a designer, either you want or you don’t want PD in your virtual world. If you do, bite the bullet and have it. Don’t argue that what you have instead is “as good as” PD, or that it “hurts just as much as” PD; it isn’t and it doesn’t unless it is PD. If you don’t want PD, that’s your call; don’t refer to the effects of a wuss-slap as “death,” though, or you’ll be perceiving consequences that aren’t there.

It may seem from the preceding that I’m a die-hard PD fan unable to move with the times. Actually, it’s not so much that I’m a fan of PD as that I’m an unfan of the alternatives as they are often used—that is, with complete ignorance. In many cases, PD should indeed rightly be off the agenda; I have no problem with that ⁴. What infuriates me is when people put up ill-considered alternatives to PD without addressing the consequences, setting unwise precedents that the equally badly informed designers who follow slavishly copy.

If you want PD, you must understand what it buys you, and what you have to pay for it. Likewise, if you don’t want PD then you also must understand what that buys you, and what you have to pay for it. If the consequences of PD are too hideous to contemplate, fair enough, go for non-PD; however, if you do so you should make sure you can address the consequences of not having PD, too, or you may end up with something equally hideous.

So, let’s take a look at these consequences.

The Unfortunate Consequences of

Permanent Death

People really do not like it when their character dies. That’s really do not like it. Not only will they say they’ll leave if it happens, some of them actually will leave. There’s ample evidence of this from virtual worlds that launched with PD but toned it down in the face of player opposition. You can take steps to minimize this—keep the incidence of PD low but the awareness of its possibility high—but it’ll still happen. Even non-PD PvP was softened in early flagship graphical worlds. People don’t like to lose, they don’t like to suffer, and they especially don’t like other people making them suffer.

It’s easy to dismiss this as a short-term thing, and that once players “mature enough” to see the benefits that PD brings they will be willing to tolerate it; there’s informal evidence to support this position from long-term textual worlds with PD. However, even supposing that it’s correct, this still doesn’t help. Virtual worlds rely on newbies for their survival. Newbies are, by definition, immature (in their appreciation of virtual worlds). Given the choice between a virtual world that they can explore in complete safety and one where it seems a moment’s lapse of concentration could undo weeks of play, well, they’ll take the safe one, thank you very much.

Ah, the argument continues, but in time people will come to realize that such virtual worlds are unsatisfying, and, although they may be newbies to my virtual world, they won’t be newbies to virtual worlds in general. They will be mature enough to give PD a chance. Well, perhaps, yes, but remember that people view all virtual worlds in the context of the virtual world with which they started. If they began with a virtual world that had no PD, they’ll judge your virtual world from that standpoint ¹. It had better be a damned good one if they’re to give it a try.

Even if they are “mature enough” for PD, their attitude is analogous to the way that people in the real world view public transport. I like the idea of public transport, so do you, so does everyone else. I like it because the more that other people travel by bus, the roads will be less congested for me to travel by car. So do you, so does everyone else. We’re all in favor of it, but only for other people. So it is with PD: It’s fine when it happens to you, but not so fine when it happens to me.

Another issue is that PD is regularly confused with PvP. Players of Ultima Online didn’t like it when their characters were attacked by large numbers of PCs operating in well co-ordinated gangs. It was unfair: These PKers were ruining the game. That’s true, it was unfair and they were ruining the game; but it’s a PvP issue, not a PD issue—those “PKers” weren’t actually PKing. Nevertheless, because PvP and PD are so often combined to such disastrous effect, in many players’ minds they’re two sides of the same coin. Most players today haven’t even experienced PD but are categorically opposed to it.

So, PD has an (not entirely unjustified) image problem which means players will probably stay away. Furthermore, if it occurs too often or too unexpectedly, so that players feel it’s either arbitrary or unavoidable, even those players who came may decide to go. This is why developers of virtual worlds are rarely keen on PD.

There is, however, another reason why PD (in combination with PvP) is off-putting: The need for a very strong elder game. Players who manage to survive to “the end” need something to occupy them fully. If the designers don’t provide it, they’ll make their own entertainment. This may well take the form of employing their finely honed combat skills (possibly using a secondary character) to attack other PCs. A single, highly skilled killer can literally empty a small virtual world of other players, and it’s very hard for administrators to stop them. After all, they’re not exactly cheating. The fact that they want to protect the integrity of their rank by ensuring no “undeserving” characters reach it is, to them, incidental. For larger virtual worlds, these people would probably lead mobs rather than work alone, but the effect is the same: Other players quit to avoid the intimidation. As Chapter 3 pointed out, there is often a place in virtual worlds for small numbers of players of this type. Beyond a certain threshold, though, they can rapidly bring a virtual world to its knees. If you let fun PKing become your virtual world’s elder game, prepare for the worst.

Despite all this, PD nevertheless has a lot going for it. Quite how much is best shown by considering what happens when it’s taken away.

The Unfortunate Consequences of Non- Permanent Death

Let’s imagine what the real world would be like if the philosophers of ancient Greece had discovered the secret of eternal life. If you take your daily dose of elixir, you won’t age and will live forever. How would today’s society be different?

Well, it’s almost impossible to say, but one thing is for sure: Alexander the Great would still be in charge.

In virtual worlds, this is called sandboxing—the people who are first to positions of power get to keep them. There is no opportunity for change. Ensigns on the Star Trek spaceship Voyager (which was transported millions of light years from home) did not have a great career path no matter how able they were.

With no permanent death, virtual worlds consist of relentless progress toward either “the end” or some point of equilibrium (where the losses from whatever replaces permanent death balance the gains accrued in between such pseudo-deaths). When players feel they aren’t advancing, they can become very frustrated; most virtual worlds without permanent death therefore endeavor to keep losses lower than gains. This keeps players’ sense of achievement intact until they reach “the end,” although a side effect is that it devalues what that achievement means: If anyone can plod to success, no one successful gets bragging rights.

What happens at “the end” is also problematical. “The end” is when highly experienced (or highly plodding) players run out of content. Ideally, these characters should be retired and new ones started. That, however, is too much like PD for many players to stomach. Remorting is a way of restarting with an existing character (as a different class), but it’s a cop-out that really only delays the inevitable.

This irresistible rise problem is endemic in DikuMUDs. Eventually, high-level players will either join their virtual world’s administrators or find something else to do with their lives. Interestingly, this may involve starting a new DikuMUD of their own. Thus, although it may be bad for an individual DikuMUD that players leave when they run out of content, it could be good for the species of DikuMUDs as a whole.

This isn’t possible for players of large-scale graphical virtual worlds; it would be like a big-shot movie director telling a complaining member of an audience that if they don’t like what they see on the screen they should go away and make their own movie instead. That said, there are many graphical worlds being developed by people who cut their teeth playing EverQuest, so in a sense the same well-worn path is being taken.

Of course, if you’re a business that makes its money running an individual virtual world, you may not want your players to leave and set up ones in competition. Therefore, you need to encourage them to stay. The only solution not involving PD is to create new content, which is (as ever) expensive.

There is theory behind this, which anecdotal evidence from EverQuest suggests has some basis. The idea is that when new high-level content is added, it benefits not only the players for whom it was created, but also those that follow. It’s called the trickledown effect, although it’s perhaps a misnomer in that content doesn’t trickle down so much as characters trickle up to meet it. Nevertheless, the point is that although it may seem that content is being added only for the benefit of a few very powerful players, eventually all players will benefit from it as they reach that level. It’s like building a road as people walk along it: The ones at the back of the line eventually get to use what was created for the people at the front of the line.

It sounds all well and good, but it’s not without flaws:

• To newbies, the finishing post looks very, very distant. It may indeed look so distant to some that they decide not to bother embarking on the journey.

• Newbies (and not-so-newbies) feel they can never catch up. The people in front will always be in front, and there’s no way to overtake them.

• The horizon advances at the speed you approach it. You’re a donkey following a carrot on a stick.

It’s like running a city marathon. You know from the start that it’s a very long way, and not for the unfit or faint-hearted. You won’t beat any of the people who started two hours before you, so you can only judge your performance at a personal level. Oh, and the course keeps lengthening, so you’re not actually ever going to finish anyway.

In a virtual world with no PD, you only get to experience a body of content once. After you’ve consumed it, you rise in power such that the old content is no longer a challenge, so you go on to the next content. Aside from the fact that this is a huge waste of resources (players will repeat content many more times if they can come at it from different angles, for example, with different characters or different strategies), it’s not necessarily even the content as it was originally formulated. When greater monsters are created, greater treasure must also be created as a reward. Any weapons, armor, or other kit that’s better than what a character has will be retained, and their old equipment sold on to characters at a lower level. These, in turn, will sell their old stuff to yet lower-level characters, and so on.

The result of this is that characters often have better gear than they ought to have for their level of experience. They can better tackle the adversities they face, and when victorious the rewards they obtain may be less useful than what they have already. Thus, they’ll sell those on, too. It’s a form of inflation that devalues objects at lower content levels.

The situation is not as bad as it sounds, in that there are more characters at lower levels than at higher ones and therefore they can’t all have overpowered equipment. However, it’s also worse because of deliberate twinking. Formally, twinked characters ¹ are ones that have acquired equipment that they couldn’t ever have obtained through the normal channels; in EQ’s case, this means killing monsters and trading with other characters. Although strictly speaking even trading to buy a better sword than you “should” have is a form of twinking, most people use the term to mean using your own high-level character to obtain goods for your low-level character.

Twinking isn’t necessarily bad for a virtual world’s economy (especially if objects can wear out), but it throws a dark shadow over content. Newbies who are twinked by a high-level stranger are often very impressed; they’ll like the virtual world much more as a result. However, newbies who don’t have a higher-level benefactor are resentful of people who do—especially when those people don’t seem to see anything wrong with it. It’s like a Hollywood actor researching a role as a homeless person, who turns up at a soup kitchen, takes a bowl of soup, then produces a freshly baked baguette and some croutons to eat with it. That soup was meant for a homeless person, not a Hollywood actor; similarly, newbies who see experienced players consuming their content aided by luxuries no one else has will feel resentful. No amount of “I worked hard for this” is going to mollify them. It exposes the newbie experience as worthless.

Even at higher levels, it’s an issue. If you experience content with a twink (or otherwise overpowered character), you have an easier time of it than the designers planned. It’s like taking a calculator into a mental arithmetic test: You’ll finish sooner and get all the answers right, but you won’t learn as much. Players with twinks won’t necessarily see anything wrong with this, of course, because in their eyes they already sat the test once without a calculator, their main character. However, that assumes that only the answers are important; in terms of immersion, how you answer can be more telling than what you answer.

Twinking happens in virtual worlds with PD, but not to the same extent. This is because PD usually involves the total loss of all the kit you had with you when you died. If you get eaten by a dragon, the dragon gets whatever of your stuff survived; if your new character wants it, they’ll have to take it off the dragon. It’s not inherent to PD that object loss on death is implemented, but it usually is because it makes more sense. Virtual worlds without PD have moved slowly in the opposite direction, ensuring that even if you do “die” you won’t lose certain bound items to looters, lessening the impact of “death” even more. In some virtual worlds, you don’t lose anything tangible at all if a monster “kills” you. The trickle-down effect requires that new content be added. The highest-level characters, for whom it is intended after all, will invariably want input on this. It’s very tempting to listen to them. Power gamers ² are an excellent resource for regular computer games, and their opinions on possible changes will regularly be sought. Discussions on web sites devoted to Sid Meier’s Civilization III ³ have led directly to several code modifications released in the official patches. Unfortunately, however, in a virtual world where people can plod to the top without exhibiting the slightest sign of imagination, it is not guaranteed that the players of highest-level characters are power gamers; they could just be people with a lot of time on their hands. If you do listen to your highest-level players in determining content issues, make sure you can distinguish between the ones who have good ideas and the ones who merely think they do.

Note that these aren’t reasons why your virtual world should have PD; rather, they’re the situations you must tackle if you decide not to have it.

One of the biggest things you’ll need to decide is what to do with the fiction. I listed earlier the more popular ways of explaining how come PCs recover from mortal blows, but the problem doesn’t stop there. In particular, if you want a rational fiction then you’ll need to produce some reason why the mobiles that the characters kill don’t recover when they die and why they themselves don’t go up levels when they “kill” characters ⁴. Note that saying nothing may be better than saying something unconvincing, as at least it leaves open the possibility that a plausible fiction exists; if you pass off some cock-and-bull nonsense instead, you’re stuck with it ⁵.

It is particularly hard to reconcile non-death with historical or licensed worlds. When authenticity is a selling point, the inauthentic can be jarring. For example, setting a virtual world in war-torn France with PCs as members of the resistance could work, but it would rather lose something if the PCs weren’t ever in any danger of being killed. If the real-life French guerillas had been indestructible, the Nazis wouldn’t even have invaded. Non-PD isn’t the only authenticity issue with historical scenarios, of course (the fact that Paris was occupied for only 50 months somewhat limits the lifespan of a resistance scenario); it is, however, the biggest one.

Even in original fictional worlds, effective immortality has major implications. Society would develop in entirely different ways if everyone knew they weren’t going to die. What careers would people have? How would laws be made and enforced? Would science advance faster or slower? Would there be war? If not, how would groups resolve their differences? Would there be religions? Jesus’ resurrection would be no big deal if the same thing routinely happened to everyone. Where would everybody live? What would children do when they grew up?

Most designers take the point of view that these are not questions that need to be answered. The loss of immersion delivered from saying nothing on the subject is minuscule compared to the loss that would be incurred by dumping players into a virtual world that followed through the implications of immortality—it would be seriously alien to players’ real-world experiences. Under the circumstances, this is therefore the most sensible approach.

Pro-PD people may still urge caution, in that systems which work fine in the real world may behave less well when applied in a virtual world with no PD. Example: If no one died it would play havoc with the housing market. It’s often the case that churn will take on the role that death does in the real world, though: Characters do die when their players quit, therefore a house will go back on the market if its owner cancels his or her account.

There are some exceptions (for example, no PC would take out life insurance, therefore insurance companies would not be as rich as in the real world, therefore investment in stocks and shares would be lower) but these are invariably swamped by other factors anyway.

A virtual world without PD can be created to address all these issues, some of which are, frankly, of the “clutching at straws” variety anyway. Most virtual worlds do, in fact—worlds without PD greatly outnumber those with it. You can have a successful non-PD world.

So why is it, then, that time after time designers and long-term players will sense there is something missing from their virtual world, take a good, hard look at it, and conclude that, you know, maybe PD really isn’t all that bad?

Why Permanent Death?

Players are against PD because they don’t want to die. Although some socializers wishing to flaunt their loving and caring credentials might disagree, players in general don’t care if someone else dies—they just don’t want it to happen to them. Because no one wants it to happen to them, it doesn’t happen to anyone. So everyone is happy, right?

For a time, yes. Then, the fact that the whole experience is vacuous begins to nag at them.

Achievement—and with it a sense of personal advancement—can come in two ways in virtual worlds, neither of which occurs without PD or some equivalent. Characters can either advance gradually via a series of small actions done well, or heroically through single, life-changing events. Without PD, “small actions” are steps on a treadmill and “done well” means you move slightly faster than people who have “done badly.”

Heroism is no such thing—it’s just another example of a “small action.” This is the root cause of the dissatisfaction that many players of non-PD virtual worlds develop.

Players play virtual worlds to explore and celebrate their identity. The undertaking of courageous acts, whether for gain or to help others, can make highly significant contributions to personal development. It’s the uneasy realization that actually these acts weren’t as courageous as you thought they were that can cause the problems.

Note that it’s not the immediate thrill that’s important. Fights in EverQuest wouldn’t necessarily be any more exciting if death really meant death. It’s in the rationalization where it differs. When failure at an action means oblivion, you can say more about someone who attempts it than you can when all they’ll suffer is a wuss-slap. The difference is of immense importance, because they’re not only putting their tangible character on the line, they’re putting their nascent identity on the line too.

People may not want to undertake such risks. This is fair enough—bravery doesn’t have to mean stupidity. In a PD world, players assess the chance they’ll escape with their lives: If it’s too low ¹, they’ll pass. In a non-PD world, players don’t want the risk but still want to perform the action. The penalty for failure is therefore lowered to acceptable levels. Even stupid players won’t lose their characters. Failure may be annoying, but it takes more than that to make success heroic.

Acts are only heroic when there is a significant, perceived chance of real loss and no easier alternative. The following acts are not heroic:

• Standing in the way of a runaway cat. (No chance of real loss.)

• Walking across a minefield you didn’t know was there. (No perceived danger.)

• Swimming across a crocodile-infested river next to a bridge. (Brave, but unnecessary.)

Putting PD into a virtual world does not make people heroic. It does, however, ensure that those who talk heroic are heroic. The designers of virtual worlds can make all their players immortal, but they can’t make any of them heroes. Only players can make themselves heroes. If they later come to realize they’re not the heroes they thought they were, disenchantment sets in.

The most compelling reason for having PD is that it makes a virtual world meaningful in its own terms. Players can always construct meanings for activities in their virtual world, but these are not intrinsic to it or (crucially) to the real world. Without PD, a virtual world can still be real-world meaningful, but in external contexts only (for example, it’s a dating agency).

Virtual worlds are about identity; if you can’t lose something, you don’t value it; therefore, you can never value your identity in a non-PD world as much as you can in a PD world. This is less fierce than it sounds, in that immersion to persona level can provide the external context that makes a virtual world meaningful (to the immersed player); it’s just harder to reach that level of immersion in a virtual world that doesn’t have PD.

Meaningfulness is the fuel that powers the engine of identity development. It helps you get where you’re going. Without it, you either have to freewheel (non-PD) or get out and push (role-play).

Pleasure only works if there is pain against which to compare it. For many players, the virtual world is the pleasure and the real world is the pain; these players don’t want to experience pain in the virtual world too, hence the issue of consent is important. However, withdrawing consent also withdraws meaning—not just from the withdrawer, but from everyone else, too. It violates the “no easier alternative” requirement of heroism. Mechanisms to prevent this aren’t hard to propose—a simple level cap for immortal characters would do it—but the problem is that many players won’t stand for it: They want meaning without consenting to what’s required to give it to them. If the virtual world can’t provide it, they have to provide it for themselves. Is it surprising that they often end up dissatisfied?

You can see their point of view, of course. As far as they are concerned, they are role-playing a hero in a virtual world built for heroes ². Heroes are special people who don’t get killed, ergo their character shouldn’t get killed either. Players may be nobodies in real life, but in virtual worlds they’re somebody.

In the Gilbert and Sullivan musical The Gondoliers ³, there is a celebrated song, “There Lived a King.” It tells the tale of a monarch who, determining it to be unfair that he can drink Rhenish wine while other people have to drink toddy, promotes all his subjects to the highest ranks. The kingdom swarms with lord chancellors, bishops, ambassadors, prime ministers, dukes, field marshals, lord lieutenants, admirals, and party leaders.

The result isn’t hard to predict:

In short, whoever you may be, To this conclusion you’ll agree, When every one is somebodee, Then no one’s anybody!

If everyone is a hero, no one is.

This argument is very pessimistic. It suggests that everyone can’t be a hero even if they do heroic things: Everyone does heroic things, everyone becomes a hero, heroism is commonplace, therefore no one is truly a hero. The meaning has been stripped by its ubiquity.

Heroism does have a saving throw, though. It’s true that if everyone were a hero then no one would be. The thing is, everyone can be a hero, just not at the same time.

The Hero’s Journey

In his influential book, The Hero with a Thousand Faces ¹, academic Joseph Campbell asserted that the various heroic myths of all cultures follow the same, basic storyline. His theory was that myths are tied to the human psyche; they are cultural expressions of the universal need to explain the same, fundamental concepts of social, worldly, and other-worldly realities. Not all narratives follow the basic pattern (which is called the hero’s journey ²), but all myths and epics do. Famously, the movie Star Wars follows the hero’s journey almost to the letter.

Many of the steps in the hero’s journey are symbolic or metaphoric, so it may take a student of myth to recognize some of the more subtle ones. Once they have been identified, however, the basic sequence will usually be the same. Occasionally, steps may be omitted or reversed in order, but on the whole the overall structure of the myth will be the same. Heroes are heroes because they complete the hero’s journey.

This is a different perspective on heroism to that understood by players of virtual worlds. The hero’s journey includes heroic acts, but these alone don’t make someone a hero. If you risk life and limb—even in a PD world—to save an innocent, that’s an heroic act. People may regard you as a hero, but it’s a simile: They’re holding you in the same esteem they would hold a hero of ancient myth. If you want to be a hero, though, you have to complete the hero’s journey.

This, you can do. Let’s examine how.

The hero’s journey consists of a number of steps. Some of these are more significant than others, some take longer to occur than others, some have more symbolic connotations than others. They are grouped into three phases: departure, initiation, and return. Here’s a brief description of the steps.

Departure

• The call to adventure. The hero is given an indication that everything is going to change. The call may be symbolic, and the hero doesn’t have to realize what it means.

• Refusal of the call. The hero is required to act, but initially refuses. This can be through fear, duty, or a whole bunch of other things. If the hero holds out, they may be subjected to assault until they take the hint.

• Supernatural aid. Once the hero accepts the call, a supernatural guide is revealed to help them. Now they can enter the world of their quest.

• The crossing of the first threshold. The hero leaves their own world and enters that of their adventure. This is a dangerous place, with unknown rules and unknown limits. The new realm may have a guardian that the hero has to defeat or befriend to get past the threshold.

• The belly of the whale. This is symbolic, representing the final separation of the hero’s old world, old self, and a developing new world, new self. By entering the belly of the whale (or anything womb-like), the hero is committing to rebirth and transformation.

Initiation

• The road of trials. These are a series of (often three) ordeals that the hero must undergo to begin their metamorphosis. They could well fail at least one.

• The meeting with the goddess. This is another highly symbolic step. The goddess represents the totality of knowledge; if the hero is not consumed by exposure to it, they are liberated by it. This is usually described in terms of the hero’s experiencing absolute, unconditional love. The understanding the hero gains leads to a “union of opposites;” the hero’s self-image begins coalesce as the hero gains awareness of the forces of life and death, of mortal and spiritual.

• Woman as the temptress. Woman here is a metaphor for the temptations that may cause the hero to stray from their journey. The hero’s old-world origins are at odds with the new world, but the hero overcomes the urge to return to them.

• Atonement with the father. This is the key point of the journey, in which the hero faces whatever entity has the supreme power in their life. The hero recognizes the darkness in his or herself and discovers the light. The hero’s old self is killed (literally or metaphorically) and the new self steps forth.

• Apotheosis. A period of peace, rest, and bliss that the hero experiences before commencing the return home. Old cultural assumptions are finally broken, and old prejudices forgotten.

• The ultimate boon. Although the hero has already achieved the real aim of their quest, this is where they pick up the object they set out to acquire. The boon is often a token for which the hero is now sufficiently pure to obtain. They must bring it back for the benefit of all.

Return

• Refusal of the return. Life in the new world is so pleasant and carefree that the hero doesn’t see any reason to go home.

• The magic flight. The hero can’t stay and have the boon. This is the exciting escape with it.

• Rescue from without. The hero still doesn’t want to return, is prevented from returning, or is too weak to return. Powerful friends or allies help it happen.

• The crossing of the return threshold. The hero returns to their old world and old life. How will they reconcile the old and the new? How will they apply their newfound wisdom?

• Master of the two worlds. The hero achieves a serene or transcendental balance between the old, material world and the new, spiritual one. They accept their destiny.

• Freedom to live. Due to their mastery, the hero no longer fears death; they can live for the moment, unconcerned with whatever the past has held and whatever the future holds. It’s the journey that’s important, not the destination.

In the context of virtual worlds, the hero’s journey is normally discussed in terms of its relevance (or lack thereof) to story arcs. That’s not how I shall be applying it here.

The hero’s adventurous journey betokens a journey to find the self. It’s the same journey that players of virtual worlds undertake. They play to have fun, but they gain much more than that: They develop as people.

Let’s take a look at the steps as they apply to virtual worlds.

Departure

In this discussion, the “old world” is the real world and the “new world” is the virtual world. The departure action therefore takes place in real world.

• The call to adventure. Something triggers the would-be player’s attention. It may be a specific advertisement, testimonial, shelf unit, cover disk, magazine article, or pushy friend—it doesn’t really matter. The point is that the player now finds the idea of virtual worlds appealing when previously either they didn’t or they hadn’t thought about it.

• Refusal of the call. There are lots of reasons not to enter a virtual world: expense, time, fears of inadequacy, a social life. However, the appeal of the virtual world grows. The player really wants to try one.

• Supernatural aid. This step may be skipped if the player concedes to their desire. Where it isn’t, it’s likely that the player will know someone else who already plays and is able to help, or is a member of a group of friends who decide to try the virtual world together.

• The crossing of the first threshold. The player installs the software on their computer and powers up. They enter the virtual world, which is a dangerous place with unknown rules and unknown limits. The quality of the interface determines whether or not it counts as a guardian.

• The belly of the whale. The character generation system. The player creates a new self for the new world. It’s a formal rebirth; the player is undertaking to become someone else.

Initiation

This stage takes place in the virtual world, and is therefore the one that is most important to virtual world designers (more important than they may realize!). You may want to look back at the player development tracks shown in Figure 3.5 before you read this. You also might like to refer to the review of Hedron’s Six Circles, because this (at last!) is why I mentioned it ¹.

• The road of trials. The player finds their feet. This is the opportunist/griefer step, where players meet and respond to small challenges to determine the extent of their abilities. Their success or failure at these will establish their course of future action.

• The meeting with the goddess. The player seeks knowledge, either through experimentation (as a scientist) or from others (as a networker). The more that players learn about the virtual world and its inhabitants, the more they learn about themselves. If they can handle what they learn, they are stronger for it. They begin the process of immersion—that is, of coalescing their real-life self with their virtual self.

• Woman as the temptress. Once players have learned enough to become accomplished, they are at last informed enough to judge whether they want to attempt to complete it. Are they in it for the long haul, or is their academic curiosity sufficiently satisfied that they don’t feel they need to continue? This step covers the transition from discovery to application, from learning to doing.

• Atonement with the father. This is the step that players spend most of their time taking, putting the skills they have acquired into practice. It’s where the planner-type achievers and politician-type socializers strive to improve and impress. The virtual world’s achievement metric (for example, character level) charts the players’ progress to their goal. When they reach the end as the virtual world defines it, they “win.” It is the key moment in their virtual existence. The “father” who they are confronting is the virtual world’s (lead) designer, whose will is expressed by the virtual world and who wields its ultimate power. When the designer (through the implementation of the virtual world) accepts them, then they have the closure they need to move on to the final, serene state of total immersion.

• Apotheosis. Players wind up here as friend-type socializers or hacker-type explorers. They understand the virtual world, its people, and themselves; they are at peace with all. Challenges from the virtual world, when they arise, are no longer important.

• The ultimate boon. Oh dear. Virtual worlds don’t have an ultimate boon. Players have no token of their achievement. They have skills and wisdom that they can bring back with them into the real world, but nothing formal to symbolize it.

Return

This final stage accounts for players’ return from the virtual world to reality (or, perhaps more tellingly, the acceptance of players that the virtual world is part of reality).

• Refusal of the return. The player has power, respect, friends, and peace. Why would they want to return to the real world?

• The magic flight. The live team will usually not want a player to leave, and may provide a compelling elder game to try to ensnare them. They don’t, however, want their boon back (or wouldn’t, if they had one in the first place) and the barriers they erect are of the temptation variety rather than the physical or emotional of common myth.

• Rescue from without. Your parents, your workmates, your significant other—“Stop spending so much time on that computer!”

• The crossing of the return threshold. The player finally separates from the virtual world. Contrary to what many players think, this rarely involves a “burn all your bridges” deletion of characters (that will typically happen in the atonement step). If it involves an account cancellation, it’s for practical rather than symbolic reasons. The player stops playing simply because they don’t need to play any more.

• Master of the two worlds. The player’s virtual and real self are the same. The player can return to the virtual world, but it’s a place like any other. It has lost its mythical significance (to that player).

• Freedom to live. Players can finally be themselves.

Apart from the problem with the boon, this all looks very neat and tidy. The very general nature of the hero’s journey, however, with its mythical and symbolic components, means that with a little imagination it can be applied to stylized sequences of action where perhaps it isn’t really appropriate ¹. To what extent is this true for its use concerning player development tracks?

Analysis

Conjecture: Playing virtual worlds is a kind of hill-climbing activity through identity space; the hero’s journey is an algorithm for finding a very good local maximum, if not necessarily a global one.

I believe that the hero’s journey does directly apply to player development in virtual worlds. However, the fit is not exact. Given that the hero’s journey has evolved independently in all cultures through the millennia, should some effort be made to try and make the fit better? Or is the fit not exact because in truth it isn’t a fit at all?

I’ll go through what I perceive to be the main issues. This is all speculative; see what you think:

• The belly of the whale. Although it’s fine to identify the character creation screen as “womb-like” in its effect, normally a myth would have something that was suggestive of a womb in appearance, too. A cave, a pit, a log cabin, a belly of a whale…. Would making character creation take place in such an environment speak more deeply to players’ subconsciousnesses? Or does the fact it’s called “character creation” do away with the need for such symbolism as it’s pretty obvious what’s happening anyway?

• The meeting with the goddess. It’s very tempting to equate the addiction that many players feel about virtual worlds with the “unconditional love” metaphor here, but sadly it doesn’t work. Not only does addiction share none of the qualities that unconditional love is meant to symbolize, but it doesn’t really bite until the next stage anyway, where the player gets to use their new-found powers. That said, then, what is the virtual world’s equivalent of “unconditional love?” The goddess could be a player’s guide or a web site or some other fount of knowledge, but where does love come in? It’s a very common metaphor in myth.

• Woman as temptress. This is a stage that all players go through, but it’s not really any more special than the transition from griefer/opportunist to networker/ scientist.

If we lost lots of players here who decided that now they know the rules they don’t want to play, fair enough; but we don’t. Why is this step important for the hero’s journey but less important (although still present) in player development terms?

• Atonement with the father (I). The mere existence of this step suggests that the player’s presence in the virtual world must come to an end; players need the closure, they can’t go on forever. This is not something with which the developers of virtual worlds would necessarily concur. They don’t want players to leave ¹, so they continually add new content—thereby denying atonement and initiation. In such circumstances, players have huge difficulties satisfactorily completing their hero’s journey: They can only do so by rejecting the father (who has rejected them), which is not pleasant.

• Atonement with the father (II). Atonement implies that acceptance isn’t automatic: The player must “deserve” it. This is where a PD world delivers and a non-PD one doesn’t. If success is guaranteed with time, acceptance isn’t in question.

• Atonement with the father (III). Virtual worlds can only measure tangible things. For planner-type achievers this is easy enough (experience levels, skill levels, possessions, and so on), but for politician-type socializers, how do they gain final acceptance by the designer?

• Atonement with the father (IV). What happens when control of the virtual world is wrested from the design team and handed to, say, the marketing or the community service group? An individual can act as a foster-father figure, but they didn’t actually create the virtual world. Has the father abandoned the hero?

• The ultimate boon. The reason that most virtual worlds don’t have this is because developers don’t want people to leave. Although some players may accept the inevitability of their departure, most developers don’t. It wouldn’t be hard to give players recognition either within the virtual world or without it; indeed, some virtual worlds do this ². It doesn’t have to be the player’s ticket home, but it should be something special that players can only ever obtain once.

• The magic flight (I). Without a boon, this still sort of goes to plan: Players try to leave against the wishes of the guardians of the virtual world. With a boon, the magic flight would perhaps work counter to this. It’s easy to see how forcing a player to retire would lead to attempts by that player to continue to play. Secondary accounts would be opened then closed, attempts to masquerade as a new player would be made then detected, and so on until eventually the player would get the message and stay away. This is not how magical flights normally work! Heroes will usually have to break out of the new world to reach the old, rather than be stuck in the old because they fail to break back into the new.

• The magic flight (II). If players don’t leave, does not their continuing presence interfere with the progress of other players? Can heaven become full? Doesn’t the sheer number of heroes still wandering around debase the currency of heroism ³? Perhaps trying to keep players involved indefinitely is actually bad for a virtual world in the long term?

• Master of the two worlds. The keepers of the virtual world can’t give anything to the player that is of material use or value in the real world because people will cheat to get it (imagine what would happen if there were a substantial cash prize for reaching the level cap, for example). Unlike the new world of myth, people really can enter virtual worlds; what’s more, they can visit it with their buddies and bring out boons by the truckload. Players take from the virtual world the confidence, self-esteem, friendships, and life skills they picked up there, but that’s all that’s of any use. It’s up to them what they do with these.

• General. Players can achieve immersion without following all these steps. If there is no formal end-point, for example, that doesn’t stop people from reaching full immersion; it might be harder or more frustrating for them, but they can still do it. This may be a template for a hero’s journey, but you can become a hero by following your own path, too.

You don’t have to leave a virtual world if you don’t want to (they’re just places, after all), but for some people it’s a necessary closure. Is denying an exit—through the design of the world—therefore immoral? As you’ve probably come to expect now, this is discussed later in Chapter 8.

Not everyone can complete the hero’s journey; not everyone can be—or wants to be—a hero. People can and do stop beforehand; some may even be heroes before they start. Everyone can try to be a hero, though, and a well-designed virtual world gives them a better chance of succeeding than almost anything else available to ordinary people.

The critical thing is, although you can do heroic deeds in the virtual world, you have to leave for them to count. You visit the virtual world to become a hero, but you can only be a hero in the real one. The virtual world is a part of the real world.

Warriors are only heroes when the battle is over. While the battle is raging, they’re just trying to stay alive.

Attitudes to Permanent Death

Will attitudes to permanent death change?

If some way could be found to retain the desirable consequences of PD while losing the undesirable ones, attitudes wouldn’t need to change. With our current knowledge of PD, however, this seems unlikely.

Change will have to occur, however, if virtual worlds are to remain faithful to their essence, at least in combat-oriented worlds. Dilution of the virtual world “product” will leave people unsatisfied. Will they stop playing all virtual worlds, or continually drift from one to the next, forever unable to find something that satisfies their long-term aims without violating their short-term criteria?

Existing virtual world culture is anti-PD. Upcoming virtual worlds are therefore also likely to be anti-PD. This is primarily due to imperfect early implementations ¹ and bad customer service decisions; nevertheless, the legacy is there. Trying to stop it is like trying to prevent a herd of cattle stampeding toward an oasis in the desert: You know they’re going to drink it dry and then they won’t be able to make it to the distant but flowing river you want them to go to. But if you stand in their way you’re only going to get trampled.

Players, fortunately, are not cattle. Those who think about the subject—what they want and what they don’t want—may change their views and (reluctantly) support PD. Then again, they may become even stronger in their opposition to it. Change would amount to a paradigm shift, however, which is very hard to effect. That said, we’ve already had one the other way (early MUDs had PD) so it’s not impossible.

Many designers feel that the reason PD fails is because it isn’t “done right,” at least on the large scale. They could be correct. However, they also believe that the way they plan on implementing it is right. When it turns out to be wrong, they set back the cause of PD even more. The more that designers know about PD and non-PD, the more they can base their decisions on experience rather than hope. With any luck, this section has gone some way toward helping inform such decisions, whichever way they eventually go.

I have one last thing to say before I leave the subject, which I present here without comment. It’s a statistical observation from a well-known, large-scale graphical virtual world:

A change from PD to non-PD has no effect on the number of customer service complaints.

Draw your own conclusions.

Crafting

When characters are not fighting one another, what are they doing? Socializing, exploring, and being pains, yes, but in gameplay terms what do they have to occupy them? Except in the elder game (which I’ll come to later), they’ll typically be creating something (for example, digging iron, making arrows) or providing a service (transporting iron, selling arrows). Because in most virtual worlds the service side is either trivial or automated, it’s normally lumped together with manufacturing under the general heading of crafting.

Much of this was covered earlier in the Chapter 4 section on economics, but there are a few more points that can be raised in the light of what has been discussed since.

Manufacture

Here’s how a simple real-world manufacturing process works:

• Raw materials are grown, dug from the ground, or captured. Examples: wheat, ore, cows.

• The raw materials are refined to get rid of the useless bits. Examples: grain, copper, milk.

• The refined materials are made into components or finished goods. Examples: bread, wire, cheese.

• Components are combined with other components to make new components or finished goods. Examples: sandwiches, circuit boards, pizzas.

• Finished goods are sold to consumers, who use them to destruction.

• Destroyed goods are recycled into refined materials.

In virtual worlds, it’s usually even simpler. Except for animals, raw materials typically emerge in refined form. Components don’t exist: Raw/refined materials are combined (using an abstract formula known as a recipe) to make finished goods. Recycling does not occur, except perhaps when encapsulated in some grand closed-economy scheme.

Needless to say, the excitement derived from crafting approaches 0% in intensity.

There are exceptions. The virtual world A Tale in the Desert ¹, which is set in ancient Egypt, has a full-blown crafting system at its core. Players can collect slate from riverbeds, that can be used to pay for tuition in stone blade fabrication; they can then make a stone blade from slate, which will pay for tuition in the carpentry skill; using wood obtained from trees, they can manufacture a wood plane and create boards; also using wood from trees, they can pay for the tuition necessary to make a brick rack; they can then make a brick rack; grass can be harvested and dried to make straw; they can obtain sand from the desert and mud from riverbeds; using their brick rack, they can finally manufacture bricks.

Needless to say, the excitement from crafting using this far more sophisticated system still approaches 0% in intensity. Fortunately, because it’s so tedious players will manufacture components that they sell to each other; this introduces an element of interaction that makes for some fun (at least it does if people will sell you what you need). There are some gameplay bugs to be ironed out, but designers have high hopes for A Tale in the Desert as a proof of concept if nothing else.

If a virtual world has a craft system of any substance, there are three golden rules to obey:

• All crafts should have gameplay value.

• The range of craftable objects should be great.

• The depth of crafting skills should be deep.

I’ve covered these elsewhere already, but I’ll quickly reprise the reasons for them here. Crafts should have gameplay value because this adds to immersion and interaction. This applies even to basically intangible objects such as “pretty” jewelry: Anything that impresses PCs should also impress NPCs, even if only crudely so by comparison.

The range of craftable objects should be great or people will end up making the same things and flooding the economy with more goods of one type than it needs.

The depth of craftable skills should be great or players will not consider crafting as a career. If crafting is something that adventurers do in their spare time, this is enough; if you want to cater to people who would rather create objects than destroy mobiles, however, they’ll need skill sets potentially as deep as adventurers’.

There may still be a problem when player expectations come up against virtual world economics. The “I’m role-playing a hero and I don’t see why I should be killed” adventurer has an equivalent in the “I’m role-playing a swordsmith and I don’t see why I shouldn’t sell my swords’” crafter. Sorry, guys, but no one is buying the identical swords that NPC swordsmiths are making, they’re not going to buy yours either unless you drop the price.

Recipes

My house is within five miles of four large supermarkets. The bread in the first one is too doughy; the bread in the second is too crusty; the bread in the third is too soft; the bread in the fourth is perfect (a shame the rest of the stuff they sell is so rubbish). Bread has been manufactured for thousands of years, but even today bakers use different recipes.

To make crafting more palatable to non-socializers, some virtual worlds have introduced explorative or competitive elements to it. One favored technique is to have multiple recipes for the same basic object (or, commonly, spell—this can be applied to more than just manufacture), some of which will endow the resulting creation with different properties than others. A recipe typically consists of a set of ingredients; these will often have to be explicitly arranged in some (usually linear) sequence, but in some implementations the “use recipe” action will do it implicitly.

This is one of those suggestions that sounds reasonable until you think about it. Unfortunately, it is frequently implemented without being thought about.

The idea is that players should be rewarded for experimentation. If you sit around mixing reagents and come up with a recipe for a steel that is more hard-wearing than regular steel without being any more brittle, you should be able to profit from your experiment. This is fair enough: If people discover through use that your swords are better than Joe NPC’s, they’ll come to you next time they need a new one. This is, of course, until some helpful soul puts a ton of sword recipes on their web site with a description of the precise effect of each one.

To counter this possibility, designers will usually try one of two approaches. The first of these is to allow characters to patent a recipe if they discover it first; subsequent use of this recipe by any other crafter will result in a royalty for the patent holder. This is fine for those who start playing the virtual world early, but it’s no use to those who come along six months later.

The second solution is to individualize recipes, so that the proportions of ingredients that you need to make the perfect sword are different to the ones that I need. This is unrealistic when applied to mundane object creation, but works for things like magical artifacts. Asheron’s Call allows for spell experimentation through different ordered combinations of “spell components,” and incorporated a random taper to make these unique to every character. These yielded easily to brute-force cracking methods, though. AC did have a back-up mechanism to discourage spell recipes from becoming commonly known, in that the more often a spell was used the weaker it grew. Mages who knew the recipe were thus disinclined to tell others what it was. Unfortunately, it only takes one recipe-holder to break ranks (for example, an explorer who doesn’t care if a spell loses its potency) and anyone who wants to find out can.

The root of the problem is that players don’t like having personal recipes. They want recipes to be global, so they don’t have to figure them out in advance. Designers may want to convince them otherwise, but it’s not going to happen. Like any other discovery in virtual worlds, once a recipe is unearthed it rapidly becomes common knowledge.

Ironically, although recipes are typically conceived to help crafters, it’s adventurers who are most likely to end up using them. There is some scope for using recipes in directly competitive situations—that is, as combat scripts. There, the efficacy of your recipe is directly dependent on the recipe your opponent is using. Co-operative recipes, where I do this while you do that, are also possible; however, they’re not going to make manufacture any more fun.

Beyond the Virtual World

So, I want to bake a loaf of bread. I get the ingredients, I mix them up, I stick them in the oven, then—I wait? What if I’m not baking bread but am manufacturing swords? I get the metal together, heat it, beat it, rapidly cool it, heat it, beat it, rapidly cool it, heat it—for how long? Do I actually have to do this? My character has the skills, not me—why should I have to be present while it happens?

The sheer monotony of repeatedly clicking on the same icon (or, for really advanced systems, the same three icons) for long, long periods has led designers to look at ways to alleviate the boredom. The most promising one opens up other possibilities, too: offline action.

Although many players may find this hard to believe, they cannot spend 24 hours a day playing in their virtual world. There are real-world demands that have to be met, most notably sleep. Some of the time, players are disconnected from their characters. What do the characters do during this offline period?

Well, the answer is that they can do the stuff that the player didn’t want them to do while being played. Thus, the character can sleep, eat, visit the bathroom, and so on. For reasons of safety ¹ and confusion ², they will usually not be visibly present in the virtual world while they do this (that is, they’re not temporarily NPCs); however, they are there in the abstract ³.

It seems obvious, therefore, that if players want their characters to do something mindlessly boring, this would be the time to do it. You set your character digging, you log off, go to school, come back half a day later, log back on and your character has deposited six truckloads of ore in your depot.

Virtual worlds always took some account of characters’ continued existence while their players are offline. MUD1, for example, would gradually heal characters’ wounds while no one was playing them. These weren’t actually offline activities, though; you didn’t really have a choice about what your character did while you weren’t there.

The first steps in that direction concerned trade. You hired an NPC to buy and sell stuff for you (usually the latter rather than the former) while you went away adventuring. The NPC remained in place when you finished your session, therefore when you next returned you might have found that someone had bought something from you or sold you the materials you wanted. There are some issues as to how price changes work in this situation, but in general it works well. Offline manufacture extends the same principle to the PC, rather than some NPC hireling, and simulates what the PC was doing rather than actually doing it (as it does with the NPC).

Skill learning fits this system well. Learning how to make a pot isn’t something that occurs the instant you hand over your training fee; to make sense, it ought to take time. By making it an offline occurrence, this passage of time can be suggested without drying the brains of the players. Furthermore, it makes them eager to come back next session to try out their new skills, and hey, it benefits adventurers, too!

There are other applications for offline action of which travel is perhaps the most interesting. Virtual worlds are large, but players don’t really get a feel for how large because of the (economy-wrecking) instant travel solutions that are available to them. No one wants to make players march their characters across a desert, but how can you convey a sense of remoteness if you let them press the magic button and materialize instantly? Well, offline action lets you do it. You set off walking, riding, flying, or whatever, then disconnect. When you reconnect the next day, there you are. If you were to reconnect prematurely, you would appear in the desert, your exact location interpolated from your starting position, destination, the elapsed time, and your mode of transport.

There are two ways to handle offline activities, whatever you use them for.

The first way is to give orders in advance. Just before you log off, you instruct the virtual world as to what your character will be doing while you’re away. Some of these tasks will take longer than others, but you can create a queue of orders so it doesn’t really matter. When you next log in, the character will have worked through some or all of the queue, and you can pick up from there. The advantages of this approach are that you can be informed (for example, by email or phone message) when a task is complete, and that you can set orders without using the virtual world’s client software (for example, via a standard web browser from work).

The second way is a more quantum mechanics approach. Your character accumulates activity time while you are offline, but you don’t have to decide how it’s spent until you log in and collapse the quantum wave. This means you can take into account events that have occurred while you were offline—for example, if gold has been discovered in them thar hills in the direction opposite to where you were intending to travel. This retroactive approach doesn’t allow for interrupts when actions are complete, but it’s handy for meeting up with friends who have wandered off since you last played.

There’s no reason why a virtual world couldn’t use both of these approaches, of course: When the action queue is empty, subsequent offline time is banked for when the player next logs in.

Offline actions aren’t a panacea for boring activities. In practice, there will usually be some limit to the amount of offline time that can be spent, if only to stop players from setting up programmed mules to act as craft factory cash cows. Some people object to it in principle, as it can break the bond between player and character (“How can I be in the virtual world and at work in the office at the same time?”). Nevertheless, this idea does have potential and the advantages seem to outweigh the disadvantages.

The Elder Game

There comes a point when players have advanced so far that they feel they have achieved everything that they set out to achieve. They are no longer interested in the activities that used to occupy their time: They feel they have “made it.” The question then arises: What can they do instead?

Who Plays the Elder Game?

The content that occupies players in advanced states of immersion is known as the elder game ¹. It deals with four types of player, each with differing needs. In hero’s journey terms, these are those who are in the following states:

• Pre-apotheosis. Players whose characters have advanced quicker than they themselves have. This may be because the players have had too much help on the way, or they didn’t recognize/understand the atonement (that is, final achievement) process. These players will want more of the content that they were consuming before they reached the elder game, but which they have now exhausted.

• Apotheosis. The players for whom the elder game is primarily intended. They have succeeded in their goal, and are enjoying resting on their laurels prior to returning to the real world.

• Refusal of return. They don’t want to stop playing, and the live team usually doesn’t want them to go either. However, they’ve either used all the elder game content or are self-generating it.

• Post-crossing of the return threshold. These players can return happily to the virtual world at any level. They have nothing to prove, and can therefore play how and when they want for the sheer joy of it.

Let’s take these in order.

Looking back to Figure 3.5’s player development tracks, the main types of player in a pre-apotheosis state will be politicians and planners. Politics has long been promoted as one of the two great elder games for large-scale virtual worlds ², because it self-generates content: Players do things that impact one another, which leads to new courses of action with different impacts, indefinitely. Self-generated content is inexpensive and long-lasting. Evidence from textual worlds suggests that politics ³ is indeed a good elder game for players with conveyor-belt characters that traveled at the wrong speed; it’s also a useful delaying pursuit for players at the refusal to return step. It doesn’t really help those in apotheosis, though.

The other pre-apotheosis player type likely to have characters in advance of their degree of immersion are planners. These are the achiever types who went at it too mechanistically and matured their characters before they matured themselves. They’re still looking to achieve. For these players, remorting is probably the best solution, as it at least holds out the prospect that they’ll reach a sense of fulfillment at some point; however, the existence of remorting as a possibility may lead other players not to feel they have “finished” when perhaps before they might, therefore they’ll head off on another dispiriting achievement ladder that will do them no favors at all.

For some virtual worlds, the reason that planner-type achievers reach the elder game is because the virtual world itself has no formal indication that they have reached “the end.” In this situation, the human thing to do would be to add a recognition that the player has undergone a rite of initiation. The inhuman thing to do would be to add more content to keep the players achieving. Sadly, the latter is not uncommon, particularly in commercial virtual worlds.

Players who are in a state of apotheosis will be either hacker-type explorers or friends-type socializers. The former can usually be catered for by allowing building privileges (assuming that they didn’t have them already; if they did, building is part of “the game” rather than “the elder game,” so won’t be of much use here). Building may be only a limited option in a commercial virtual world, where real-world copyright laws could be enforced to devastating effect. Legal solutions are likely to be found, however, which will remedy this situation.

Friends-type socializers will want to sit around chatting to one another, and are therefore easy to deal with as long as they have something in-context to chat about. Historically, chat content has been generated by the snoop command, which allows one player to see what another player sees without that player’s knowledge. Again, although this works for hobbyist virtual worlds, there may be legal problems if it were tried in commercial ones; people value their privacy, even if they’ve signed an end-user agreement that says they don’t.

As mentioned earlier, politics works well for players who are refusing to return. It gives them something vaguely meaningful to do, which they can practice almost indefinitely (even when, as a party leader, they lose an election and are rejected by their party—at least it gives them a goal to reclaim what they’ve lost). Unlike the situation for politician-type socializers, it is unlikely that post-apotheosis players will actually enjoy what they are doing, however, and resentment could well build up. The end, when it comes, could be acrimonious.

Players who have crossed the threshold are the easiest to deal with as they require no new content; their elder game, if it can be called that, is to do whatever they like doing the most. Responses vary. At one extreme, they may never feel the need to enter the virtual world ever again; at the other, they might take up residence and spend their whole time exploring or helping newbies or politicking or whatever (hopefully not slaughtering newbies). Some may become members of the live team.

Player Interest and the Elder Game

I mentioned in Chapter 3, while deriving a third dimension for the player interest graph, that MUD2’ s wizzes are griefers, politicians, networkers, or friends: They are not the world-oriented types of planners, opportunists, hackers, and scientists. Why is this? Shouldn’t they be hacker-type explorers or friends-type socializers?

MUD2 has rules about what wizzes can and cannot do. To prevent abuses of their (considerable) powers, they are not allowed to use them to play “normally” as a non-wiz (that is, a mortal). In other words, if a wiz wants to play in the mortal world as a mortal, they can’t use their special wiz abilities; they have to use a different, non-wiz character. In the discussion from which the original player types graph was derived, wizzes therefore considered what they did as a wiz to be separate from what they did as a mortal. Some of them spent most of their time as a wiz; some spent most of it as a mortal. The rules forbade them from doing achiever or explorer things as wizzes, which is why they didn’t show up in the analysis of what wizzes did. In fact, these players did all these activities, just not using their wiz personae ¹.

The reason that wizzes weren’t exclusively hackers or friends was because many of them had crossed the threshold and were no longer in apotheosis; they were playing merely because they enjoyed playing. Some things they could do using wiz powers; others they couldn’t, so they did them as mortals. This is why Figure 3.3 doesn’t include any world-oriented player types—it only lists what wiz-level players were allowed to do using their wiz-level characters.

MUD2’s wizzes were automatically co-opted to the live team, a duty they could dispatch to whatever degree they cared. They could only use their wiz characters, because wiz-owned mortals had to remain incognito so as not to compromise them ². Thus, there were only four ways to behave as wizzes. Here they are:

• Griefers. Griefers with power will either use it responsibly ³ or abuse it and be banned. In a well-designed virtual world, griefers who are unable to move on will rarely reach the elder game anyway; those griefers-at-heart that do will understand that they can act explicitly on other players in a benevolent fashion, even if the delights of malevolence ultimately prove too tempting. In MUD2, for example, griefer wizzes would try to make the lives of mortals more interesting by tinkering with the virtual world itself in subtle, undetectable ways. Done well, it was like being guided through trials by your own, personal angel; done badly, it was like being tormented in perpetuity by your own, personal devil.

• Politicians. These people see themselves as good guys, organizing fun activities like leaders at a children’s holiday camp. These activities can be invaluable as bonding exercises, but they also can be immersion-busting and incredibly irritating to those who don’t want to take part. Politicians tend not to get on with other politicians, whose events (and even presence) clash with their own endeavors.

• Networkers. These are oracle-like players, who are happy to help others by sharing what they have learned, yet rarely in a boastful kind of way. Unlike explorer types, for whom the acquisition of knowledge is itself intrinsically fun, these people give it away for free. Although their help and advice is limited to low-level activities that help people get involved with a virtual world, it’s generally fine. If they spout less well-known information, this will annoy achiever types (who see people getting an easy ride) and some explorers (who see the dissemination of information as taking away the potential fun of other explorers).

• Friends. They snoop on mortals, they chat, they laugh, they complain about how things aren’t the same as they were when they were a mortal. They like each other’s company, and don’t like it when other wizzes fall out.

To this list, we should add hackers and scientists. Hackers use their knowledge to create things, but aren’t all that bothered in exploiting it. Whereas politicians will build a whole new area and dump players in it to have fun, hackers do it because the creativity itself is the interesting part—they don’t care if players actually use it, except maybe so they can test its functionality.

Scientists are like friends-type socializers, in that they snoop a lot. However, this is because they can find out more about the virtual world, rather than finding out about the players. Scientists use snooping like it was a documentary; friends use it like it was a soap opera.

So the wizzes did, in fact, spend their time doing more than four things. If players have mastery of both worlds, they can live how they see fit. That means they have passed beyond the confines of player development tracks (which stop after apotheosis), and can be themselves. They can be any of the eight player types—it’s just that these didn’t show up in the analysis. When playing as achievers, they looked just like any other achiever; when playing as hackers, they were mistaken for politicians; when playing as scientists they were mixed in with socializer-type friends. So the analysis that led to the inclusion of a third player types dimension was, while not exactly flawed, a little incomplete.

Well, it’s complete (for) now.

Player-Created Content

Most content in virtual worlds is created by the development team (to the specifications of the designers). Some is created by the live team ¹. Some is created as a community statement because the players will it ². Some is created by the players themselves.

Players “add content” whenever they play by their in-context actions. Anything they do that affects the virtual world or its inhabitants is adding content. There’s always a positive effect to this, even if the action as a whole is detrimental to the virtual world: In the same way that any publicity is good publicity, it can be said that any content is good content (because it’s better than no content).

Content that arises from the natural actions of players within the design framework is said to be emergent or self-generating. Although designers recognize its importance, normally when they talk about player-created content they mean content explicitly created by players; players as designers and developers, in other words.

Player-created content is extremely sticky, at least for those who do the creating ³. This makes it prime elder game material. Some virtual worlds have player-created content from the very beginning, of course; these can’t really use content creation as an elder game, unless it’s a different kind of creation.

From a designer’s perspective (especially one working on a commercial virtual world), allowing only experienced players to create content lessens the chances of its consisting of the low-quality, inconsistent, illiterate, out-of-context, gratuitously profane, sporadic, incomplete, atmosphere-snapping, unimaginative nonsense that experience has shown is inevitable if people can (effectively) self-publish their creations. In practice, using only experienced players does indeed raise the overall quality of what’s created; at least they’re not going to try to subvert, mock, or otherwise deliberately undermine the source material. Nevertheless, their creations are still very hit and miss; some are bad, some are good, but without training ⁴ even the excellent ones are prone to making the same errors that professional designers spend all their time trying to avoid (particularly selective depth).

In one sense, it doesn’t matter. If people like doing this, why stop them? A single virtual world could act as a content aggregator for thousands of individual demi-worlds in various states of permanency. Then again, experience with TinyMUDs has shown that virtual worlds that couple unrestricted building with unrestricted access have tended to have short (albeit occasionally glorious) lives. Restricting building to only the most experienced players will remove a lot of the in-fighting that dogs content addition, but it won’t guarantee quality.

It’s the quality issue that bites hardest. Players are harsh judges. It doesn’t matter to them that a fellow player spent eight weeks creating a castle in the air: If they think it sucks they’ll either say so or keep away ⁵. In virtual worlds where players have been cosseted up until this point, such a reaction can be something of a shock. There’s no point in complaining, either: “I’m not playing if a character I spent two months building up can die!” may galvanize a live team, but “I’m not playing if an area I spent two months building isn’t played” won’t.

The answer might be to have all content vetted for quality before it is accepted into a virtual world. Most books have one author, but some have two, three, or more. When there are several authors, there will usually be an editor involved to ensure that the writing is consistent, complete, and correct. Even books with just one author will have a publisher-appointed editor to make sure everything fits the house style. Even this book has such an editor (hi, Stephanie!). If content for virtual worlds were subjected to the rigors of an editorial process, the result would be much higher-quality material.

Unfortunately, the central reason why developers like player-created content is not because it’s sticky, or because it works as an elder game; they like it because it’s free. If it stops being free (for example, because an editor is spending time looking at it), it becomes less attractive. Why employ someone to look for diamonds in the rough when you could employ someone else to make those diamonds for you ⁶?

Power to the Player

Chapter 1 stated that players can create content in a number of ways, which can usefully be categorized as being either in-context (you put a minotaur on retainer and hire a bunch of NPC peasants to construct your labyrinth) or out-of-context (you manifest a minotaur and a labyrinth from nothing). People with the ability to make out-of-context changes are “builders.” An analysis of who can build and how long their creations last determines who “owns” a virtual world.

Although the switch between in-context and out-of-context creation is a step change, on either side of the discontinuity the power to change is more continuous. Again, although designers do consider content creation using in-context methods to be important—especially for the elder game—they don’t mean that when they talk about player-created content. Player-created content means building, and there is an escalating scale of what can and can’t be done using it. In describing this, the important concepts are

• Objects. Tangible virtual world entities, including mobiles and geography.

• Functionality. Code that changes the properties of objects.

• Building blocks. Prefabricated components that are combined to make a whole.

• Free-form. New components can be constructed from nothing. In increasing power of expression, these can be combined as follows:

• Objects from predefined components. A musket is a heavy, antique, two-handed firearm. Its default weight, value, and accuracy have these new default values: mass=8000, value=125, ammo=ball, and so on.

• Functionality from predefined components. To load a musket, you must put in the gunpowder, the wadding, and the ball, in that order. Only then can you fire it.

• Free-form functionality. When you fire a musket, there’s a chance it could explode and cause a new kind of damage to your hands, arms, torso, and face. The noise could deafen people close by.

• Free-form objects. There’s this entirely new object, a grenade. When you release the trigger, you have 10 seconds before it explodes. It has entirely new properties for blast radius and the distance it can be thrown, with these default values: mass=250, delay=10, blastradius=50, and so on.

The latter two can perhaps be combined, because for functionality to be truly free-form it would include the ability to create objects anyway. Free-form object creation appears last because it cannot occur without free-form functionality.

In-context object creation occurs by combining components within the (possibly loose) constraints of a recipe, therefore it’s not hard to extend to out-of-context creation. You may get some fairly bizarre constructions, but they’re not going to cause the virtual world to crash.

Functionality additions are more problematical. If they’re from components, again your virtual world should be fairly safe. So long as you ensure that there’s no danger of recursion (for example, your new “load musket” function contains a call to “load musket” ¹), all should be well.

Free-form functionality is dangerous. People can do things that cause crashes (for example, dividing by zero) or hangs (for example, non-terminating loops). Crashes can be trapped, but the state of the virtual world after a function has aborted mid-way through will not necessarily be consistent. Loops can’t be detected, only assumed: The virtual world’s engine can stop a player-created loop after a million iterations only to screw up everything because the critical million-and-oneth iteration never got to execute. Resource hogging also can be a problem: Players really will create 99 red balloons ² without a moment’s thought, and if that works they’ll go for 1,001 Arabian Knights (that’s humor for you).

TinyMUDs and their ilk have evolved mechanisms to prevent these issues from becoming too problematical, using approaches similar to those adopted in the old days of mainframe computer timesharing systems. Crashes are handled by the operating system’s interrupt mechanism. Hangs and processor-hogging are handled by interrupting functions after they have exceeded some guaranteed minimum amount of processor time. Disk and memory usage is limited by the number of objects (and, optionally, properties and functions) that an individual can create in total—you have to delete old stuff to make room for new stuff once your quota is reached. Some virtual worlds have allocation committees to decide who gets what resources.

The more powerful player-created content forms are prevalent only in textual worlds. This is for practical reasons more than anything: Everyone who plays the virtual world can by definition type, therefore they can all create word-based content. For graphical virtual worlds, few of the players can draw (and, unlike the situation with writing, it’s usually apparent even to them ³ that they can’t). This isn’t to say that their ideas are no good; it merely means that they are unable to put them into practice. Given that most people can’t even take professional-quality photographs, this situation would likely prevail even were the best graphics-creation tools to be made available to players.

For the time being, graphical worlds that want to allow some degree of building are therefore pretty well stuck with using predefined components or restricting creation to artistically blessed players. That’s for object creation; new functionality would need to be hugely controlled if it were to escape being employed as a tool for abuse—so much so that it hardly seems worth bothering implementing.

The first 3D graphical virtual world to take player-created content seriously is World Fusion’s Atriarch ⁴. This Science Fiction world has modifiable terrain and allows characters to purchase components for the construction of buildings. Construction is open to all who can obtain the components ⁵, therefore it’s neither an end-game thing nor even out-of-context. Still, it’s a start ⁶.

Another approach is to push the problem one level back. This is the tactic adopted by Bioware with Neverwinter Nights ⁷. The idea is to make the necessary development tools for virtual worlds available to players so they can create their own independent (in this case) games. Bioware makes its money from selling the boxed sets and from hosting secure servers for characters (so they can move from one mini-world to another ⁸). It’s a business model that has potential, but each virtual world is limited to 64 simultaneous players, so it’s hardly going to challenge the likes of EverQuest on that basis ⁹.

Also, as I said, it pushes the problem one level back. Bioware are the developers of the game engine and they maintain network operations for players, but that’s all. The same design and customer service issues that apply to virtual worlds still apply, just to the (from Bioware’s perspective) players. It’s like the way that Dungeons & Dragons defines a gaming system but the players themselves design and (within the confines of that system) develop the individual game worlds; this is perhaps not surprising, given that NWN is a D&D license and implements the 3 rd Edition D&D rule set.

This tool-oriented attitude is set to become increasingly important in the development of virtual worlds. If cheap engines are available to run the virtual worlds and enough customizable predefined artwork exists to satisfy most needs, at a stroke the two most expensive development costs are removed. Hobbyist designers will be able to create their own graphical virtual worlds in the same way that for years they have been creating textual ones. A start has been made with the use of virtual reality VR viewing software (Active Worlds ¹⁰, for example, allows people to create custom, shared virtual spaces, albeit not ones you can do much in except look around). If actual game engines are made open source, we’ll see the same codebase branching occurring that drives much textual world development, too. There is some movement in that direction, most notably with Nevrax’s ¹¹ decision to make its NeL (Nevrax Library) development toolkit available under the GNU General Public License ¹² (GPL). They stop short of giving their server source code away, however.

A way to introduce some element of freeform player creation while maintaining the integrity of the virtual world is to allow link-outs to the general Internet. Depending on the degree to which developers are willing to sacrifice design control and immersiveness, there are many possible scenarios. Some examples:

• Clicking on a character extracts data from a file on the player’s own web site and presents it as a biographical description of that character.

• Clicking on a character’s shop opens a browser window showing the prices of the goods available for sale.

• Clicking on a player’s castle reads the definition of its interior layout from files on its web site.

• Clicking on a magic portal teleports your character to another virtual world entirely.

• Clicking on a web site queries the virtual world and produces data that can be incorporated into the web site (for example, prices, character name lists, remote views).

There is, obviously, great scope for abuse here. Instead of prices, that browser window might contain pornography, for example. In small, tight, role-playing virtual worlds, misbehavior can be controlled relatively easily because players know that there is a line of people waiting who would be only too happy to replace someone who was spoiling things for others. In large-scale commercial worlds, this measure of restraint is simply not going to happen except perhaps when players have a lot to lose (for example, in the elder game).

The Content Conundrum

Player-generated content is popular with those who design it, aspirational for those who see it being designed, inexpensive for developers, and it satisfies a perceived need (that is, a constant flow on new content to prevent a virtual world from becoming stale). Arrayed against it are concerns about quality, which, although legitimate, can to some extent be alleviated by the provision of design tools, documentation, testing facilities, and a person to act as content editor. Given the advantages, why aren’t more virtual worlds looking at this as an option?

It’s because of intellectual property (IP). People who create content can claim certain rights over it. For in-context creations (for example, planting differently-colored virtual flowers in such a way that they grow to show a picture) this isn’t a problem ¹ because all works can be considered to be transient; you can no more complain that someone dug up your virtual flowers than a pavement artist can complain that the rain washed away their chalk masterpiece. If the context of the virtual world allows it, then by creating within that context you have no redress.

It gets much, much trickier when out-of-context creation is considered. Different laws apply in different countries and different states, but basically:

• If work is done honestly and for hire (that is, they were contracted to do it) then you get the most rights. They can ask that their name be taken off the credits, but that’s about it. The downside is that individuals must be paid whatever minimum wages apply for work to be considered for hire, and “a free account” is below this level.

• If work is bought when complete (as you might buy a painting from a gallery), the artist retains more rights. They could, for example, object if you used it in such a way as to affect their reputation adversely. A photographer might have grounds for complaint were you to display his or her amusing images of clothed farm animals in an exhibition about bestiality.

• Paid-for licensed content (for example, a Star Wars franchise) will typically state the boundaries within which content must operate. It may also give the license-holder a veto over aspects of the virtual world’s design (for example, no one gets to kill Luke Skywalker), but as long as it operates within these formal constraints the developer still has a reasonably high degree of freedom.

• If work is donated for free, the artists may have even more rights. You might have to ask their permission before you make any changes affecting it; in extreme cases, this could mean all patches to the virtual world, on the grounds that they alter the context of the work and therefore its meaning. Also, just because they gave it to you for free, that doesn’t mean it’s not an asset—you can still be taxed on it.

• Content with sufficient creative input from the developer can be considered to have joint copyright. This means you ought to be able to do whatever you like with it, whether the other copyright holder objects or not. However, so can they. Disrespect them and you could come heavily to regret it.

• Content created without any formal signing over of ownership gives you very few rights. You can throw it out if you don’t like it, but provided that you invited its creation there’s little else you can do. If I put up a sign saying that passers-by can create sculptures in my garden, I don’t get to alter—or even necessarily move—them. All I can do is demand that the sculptor take their work away. A sledgehammer is not a legal option.

• Virtual worlds that are paid to show content have least control over it; this is effectively sponsorship. If the Brown Fizzy Drink Company pays you to put cans of virtual Brown Fizzy Drink in your virtual world, it can claim copyright abuse when the players buy those cans and arrange them to look like cartoons of suggestive parts of the human anatomy.

For non-commercial virtual worlds, none of this is normally much of an issue (basically because they have no money and are therefore not worth beating on ²). For commercial virtual worlds, many of whose players could well be lawyers themselves, it’s frightening. The legal situation is by no means clear, and nobody wants to be the one who has to empty the bank paying lawyers’ fees to determine the precedents. Consequently, developers are very keen not to dilute their IP with anyone else’s ³.

Chapter 6 has more on the new legal issues that virtual worlds are raising. Chapter 7 looks at some of the effects of player-created content on storytelling within virtual worlds.

The Whole Picture

So as to make it easier to describe, I’ve partitioned the virtual world design process—however, it’s actually holistic. You can’t treat the individual aspects of design in isolation: They impact on one another in ways that may not be immediately obvious.

Example: Crafters complain that their towns are lifeless. You determine that this is because crafting (unlike combat) is a solo activity. You introduce manufacturing processes that entail several characters’ working together. Players build houses in close proximity to one another so they can get to and from the crafting centers quickly. Towns become more vibrant as a result. However, now that people spend more time in the towns than in the wilderness, adventurers complain that the wilderness is lifeless.

If you spot the problems in advance, you can tackle them as a whole. Taking them one at a time may merely push them about. It’s like trying to put out a burning hoop by dousing it a section at a time—as soon as you move from one bit to the next it catches fire again. You have to extinguish it all in one go: You can’t solve the problem by partitioning it. Unfortunately, you’ll always get some partitioning if the design team consists of more than one person; balancing the various issues in this situation is therefore even harder.

Problems due to the interaction of subsystems tend mainly to be of the consistency, compatibility, and depth varieties; these are, fortunately, tractable. Hard decisions have to be made, however, when a design calls for two elements that are in direct opposition to one another: You can have permanent death or some form of resurrection, but you can’t have both. Indirect contradictions can be almost as difficult to resolve while being less easy to spot. Players want to be able to sell the goods they manufacture at fixed prices to guarantee an income, but that will cause lots of people to manufacture those goods, which will raise the price of raw materials, which will reduce the income players get from their goods.

Solutions are hard to come by. The difficulties that arise by partitioning can be addressed by sound development practices—communication, regular design meetings, internal web sites, lead designer sign-off, and so on. For problems with the design itself, the team must strive for robustness: They can tackle whatever they notice beforehand, but once the virtual world goes live they have to rely on the checks and balances they build in to ensure that nothing goes horribly wrong ¹. Any new design for the economy in particular is almost certain not to work straight away, so damage-limitation code will have to be in place ready for the inevitable tribulations.

Under- and Over-Design

Under-design is a frequent problem with virtual worlds that have been written from scratch. Developers spend so long creating and building the world that they forget to add any gameplay. There’s not a lot for players to do, but lots of things for them not to do it to. Alternatively, there is a lot to do but only to the same things, over and over again. Either way, there simply isn’t enough. The virtual world may be immense, beautiful, and finely crafted, but just a shell. What players want, there’s no way they can have: It’s like going on a shopping expedition and finding that all the shops are closed ¹.

Over-design is also a problem. In this scenario, designers go into such detail that it overwhelms all activity. What players want to do is lost in the noise of what they have to do to facilitate doing it. There is a place in virtual worlds for a modicum of annoying detail, and some players will find particular details more irritating than do others; when they feel it’s irrelevant, though, they’ll rebel. I don’t want to have to specify the barb style, shaft length, and flight composition/configuration of the arrows that I buy, I just want something I can shoot in my bow!

Both over-design and under-design are the result of the same thing: misdirected focus. The designers look at something in too much detail: For under-design, it’s the infrastructure of the virtual world; for over-design it’s the structure built on this foundation.

They both illustrate another aspect of selective depth: Too much detail in one place gives an impression of shallowness elsewhere—even if looked at objectively the detail elsewhere is just right.

Participatory Design

When a bunch of musicians get together and start jamming, they’re engaging in a form of participatory design. Everyone has ideas, which they offer up for approval to everyone else. Some ideas work, some don’t work, and others suggest better ideas. Computer programmers hacking together work the same way: Someone types, other people make suggestions, changes are made in the light of these suggestions, new suggestions are made. It’s an editing process undertaken (ideally) by members of a spiritual community ¹.

Participatory design is good because everyone gets involved, everyone can contribute, and everyone’s contribution is acknowledged. It helps build a sense of community, and gives all participants an investment in the final product. It’s ideal for virtual worlds, then?

At the design team level, yes. Final approval must formally be made by one (lead) designer, but this is essentially a rubber-stamping of the consensus opinion ². Brainstorming works.

Extending participatory design beyond the design team, however, is another matter entirely. Designers have always been open to ideas from players once their virtual world has entered beta testing, but what about before then? It can be very difficult to manage; in particular, inviting prospective players of a virtual world to participate in design increases the number of contributors by several orders of magnitude. The design team itself can either act in concert as a single unit, or split along individual lines. It’ll usually be the former, because otherwise players will try to play off designers against one another. However, that’s not the only reason they may do it. I’ll explain….

Why would designers want to get players involved in design? There are several laudable grounds for doing so:

• It allows players to feel they have an investment in the virtual world.

• Ten thousand minds are better than ten.

• Designers can act as culture seeds.

• Players feel influential—they really can make a difference.

• If players understand the design process more, they won’t make knee-jerk objections to things that superficially annoy them.

• It draws players into a community. There also are some less laudable reasons:

• Players can be manipulated into accepting suggestions that they may otherwise have found unpalatable.

• Designers can accept “new” suggestions they had already thought of, tricking players into thinking they’re participating when they’re not.

• Designers get to pick the choicest fruits of players’ unpaid endeavors.

The cynical view (and players can be very cynical) is that the whole effort is primarily a public relations exercise. Nothing important is ever discussed, unless it’s to get players to suggest a preferred solution in order that criticism will be deflected when it proves unpopular once implemented. After all, although some players’ ideas will be excellent, most will be unusable (jaded, inconsistent, off-the-wall, dull, unimplementable, copyright someone else, and so on); it’s a simple matter when posting a question to wait for the “correct” response to appear and then seize on it as a great, original idea. Players participating in design are at best sounding boards and at worst betrayers of the players yet to come. In this context, it’s easier for designers to stage-manage participatory design if they work as a unit than if they don’t.

This scornful view of participatory design has, fortunately, yet to gain a strong hold of the player population. Participatory design does work, and can bring substantial benefits. It will be used extensively for Star Wars Galaxies, where it will probably work well. On one great test—players’ coming to understand that newbies can’t be allowed to start with a Jedi character—its advantages are particularly clear ³.

Its long-term future is cloudier, however. As with player content creation, there are legal issues concerning the “rights” to aspects of a design. If a scriptwriter pitched a movie concept to a director, they would be calling in the lawyers were that concept to appear as a movie five years later without even a credit (let alone payment) ⁴. Why should a player of a virtual world who pitches a design concept receive no recognition (professional or monetary) when it’s used?

The response that it’s “advice freely given” only goes so far. It may, for example, be material that already exists in some other virtual world to which the designers of that world hold copyright. Besides, even though the assumption that the inclusion of your idea in a design is reward enough holds sway at the moment, this will not necessarily always be the case. Were a new, high-profile virtual world to pay for any player-suggested ideas it used, the situation could well change. Under such circumstances, full participatory design could be hard to apply effectively.

Testing a Design

Large-scale virtual worlds are designed before they are implemented ¹. After implementation, play-testing can be used to pick up the many design bugs (as well as programming bugs) that will have crept in. However, fixing them at this stage is expensive as it involves code changes (potentially quite large ones). The more flaws in a design that can be fixed before implementation, the better.

How, though, do you test a design when there’s no world to test it in?

Some facets can be tested by writing quick-and-dirty programs or macros. An economy becomes a spreadsheet; combat becomes a few screens of C. This helps get the numbers right, but doesn’t say anything about the gameplay. You could end up with a well-balanced virtual world that is an absolute chore to play.

Surprising progress can be made by simulating the virtual world, tabletop style. You draw out a rough map of an area, put some tokens on it to represent PCs and monsters, then move them around and do things like you might expect the finished product to work. This does, however, require a discipline that is beyond most designers; although you should really take notes and discuss what happened after the session, most designers are unable to prevent themselves from launching into a debate the instant the first opportunity arises; the result is that such pencil-and-paper simulations are not usually as productive as they might be.

Two things I normally recommend to design teams on new virtual worlds address these problems: dummy logs and mock diaries. Both start from the premise of imagining that it’s some time in the future when the virtual world has been implemented and people are actually playing it. What will those players be doing?

Dummy logs work better for textual worlds than graphical ones. The idea is to write an imaginary transcript of a session as it might be recorded by a stereotypical player. It forces the designer to face up to concrete consequences of their decisions: the interface, room descriptions, movement, communication, the general atmosphere, and (crucially) what the player will be doing. This exercise will typically expose the greatest strengths and weaknesses of a design: The designers can then play up to the strengths and root out the weaknesses.

For graphical worlds, dummy logs are more like dummy storyboards, and they don’t work so well ². That said, there is still a lot of mileage in asking the enthusiastic, wannabe designers of new virtual worlds what they think their players will actually be doing once the doors are open.

Mock diaries ³ are like dummy logs, but they address longer-term issues. The idea is that you write a series of pretend diary entries for a player covering several imaginary sessions, outlining how the player’s character(s) got on and what they did. If the entries often lack excitement or fun, or they start to look the same, you may have discovered a potential problem.

Dummy logs and mock diaries can be constructed for a variety of player demographics at different stages in their career. Your first session will be different to your sessions after a week (or it should be!), which in turn will be different to your sessions after a month, six months, two years, and so on. By writing dummy logs or mock diaries for these periods, you can reassure yourself that there will be sufficient gameplay to sustain long-term play, and that this applies to all the main player groups you’re targeting.

Dummy logs and mock diaries have their variations. Jessica Mulligan recommends using mock letters rather than diaries, for example; these help the designer consider why things are fun, instead of merely assuming that they are fun.

Whatever, designs must be tested. In the same way that every time a programmer finishes a new section of code they test it, so a designer should consider the effects of every new addition to the virtual world. You don’t have to write out 20 new mock diary entries every time, I hasten to add, but you should at least consider whether it will work in the context of the rest of the design. Although occasionally you really do have to say, “Oh, let’s just put it in and see how it goes,” most of the time you don’t. Think what the players, good and bad, will make of it, and you’ll have your answer. If an idea is so original that you can’t do this, okay, then you can put it into the design document and hope—but be sure that if it does fail there’s nothing important relying on its working.