WOO II An Exploration in Design - July 2012 - By Stoneghost: Dalaran


WOO II - Life Paths versus classes

Dungeons and Dragons (D&D) used classes as a technique for dividing up the work to be done by a group of questing adventurers. Each class had a specific set of skills, and these skills were to be used cooperatively to solve the puzzles and survive the dilemmas encountered during the adventure.

The evolution of massive multi-player online games (MMOGs) at first blurred, and then redefined, the relationships among these skills. Some have been rendered obsolete; all are enmeshed in new patterns of cooperation.

This is perhaps most easily illustrated by example. Any class will do, so consider the rogue:

So WHat?

The point is not that rogues are dysfunctional ... I enjoy playing rogues ... but rather that the basic game patterns that the"rogue" class was designed to support have gone away. And at the same time, a new game pattern, one that is specific to MMOGs, has evolved.

None of the traditional D&D classes have their original functions in the current game. And new roles have arisen. Dungeon raiders need to be tanks, healers, or dps ... in a pure form ... or they cannot fully contribute to their groups. While it is true that occasional talents, such as totems or auras, may be desirable to a raid, it is still the tank/heal/damage pattern that predominates. And if an aura, for example, turns out to be so desirable that every raid wants one, then the talent suddenly appears, in different formats, in other classes.

The key point is this: in a modern MMOG no single class can be indispensable.

The same is true for PvP. A rated battlegrounds team has one tank and two or three healers. The skills are important to the formation of the group; the classes are nearly irrelevant.

Classes get in the way of this division of labor ... because, as a cluster of abilities that defines a character, a class will always have some number of totally useless attributes. (My lock, for example, can create Soulstones - a wonderful ability for dungeon raiding, but nearly useless in PvP battlegrounds. If I was building a PvP character, there are numerous abilities I would choose before Soulstone.)

A different kind of distortion appears when the game tries to adjust for this. For example, the Healing specifications in most classes that can heal are very light on the ability to do damage. This is fine for healers in PvE, but not so good for healers in PvP where an killing and healing are often both required. The solution has been to add damage related talents to the specification tree ... to be chosen by PvPers and ignored by Raiders.

Examples of attempts to address and balance the problems abound ... but the bottom line is clear: classes have outlived their usefulness.

Choosing Skills and talents

Suppose, then, that the player could start with a blank canvas and can choose skills and talents as he or she wishes. How might that change the game?

Certainly, players would choose the attributes that they think will make them successful in their PvE or PvP endeavors. Novice players may pick randomly, but as they gain experience and want to do well in their groups, they will update their choices to fit.

Secondly, some skills would be widely popular. Self-healing, for example, is useful in any kind of combat. A little stealth goes a long way in certain circumstances. It's nice to be able to conjure up some food.

Thirdly, some skills would be chosen idiosyncratically. Underwater breathing, slow fall, fear ... these are not essential, but potentially useful or fun to have.

In other words, players would probably all pick certain obviously useful attributes. They would next pick attributes that, in their opinions, best fit their style of play. And lastly, they would pick interesting or "fun" attribute. Especially due to style of play and preferences for "fun", there would not be a sudden standardization of the Tank, for example, or the Healer. Each would have certain clearly recognized attributes ... just as they do now ... and other attributes would vary from player to player.

Life Paths

This is the basis of the Life Path ... first, the player chooses a role for the character: PvE, PvP, Crafts ... then the player selects the basic general talents he or she wants the character to have ... then the player selects the character's specific role-based talents.

While the construction method is different, the results need not be. If classes are truly useful, then they should reappear as players make the choices that they think are best for their characters. Most likely, however, will be characters that fit the mold for PvE tank, PvP healer, Crafter farmer and so on.

Within these molds, and given the restriction that the number of attributes that can be chosen is limited, there is quite a bit of room for variation. For example, consider a PvP healer:

To a large extent the answers to these questions depend on the venues the player frequents, the skill level of the player (more abilities requires more skill), the player's preferences (maybe I like to CC but don't like damage ... maybe the opposite), and so on. The result should be a kind of baseline for the "PvP Healer" mold ... but there will be no one single "best" healer format, and the many variations will work fine for the players who use them.

Attribute Clusters

As noted in an early section, some players prefer a much more simplified approach to decision making, and because of this, the game should provide templates for player choices based on the basic roles. In fact these probably should be the default.

That way new players would get to choose: "tank", "healer", "damage", for example, within the Raider Life Path, and the game would present them with a limited set of choices representing good options within that role. Players who wanted to could turn this feature off and have access to the full range of attributes.


Every player cannot choose every attribute. This is a limitation of the game design, and it enforces the diversity that makes for individuality. Players accept this because most players want to be unique, and most realize, even if subconsciously, that if everyone could have every talent, boredom would ensue.

Similarly, a player cannot choose all of a character's talents at step one. Well ... that could be done ... but spreading the choices along a sequence of levels gives players time to learn how to use their attributes, and it makes for a certain sense of anticipation and motivation within the game.

This leads to a " tree" approach to attribute availability. WOO II offers two trees, one for general attributes available to all players, and one for attributes available within a Life Path.

As in the current game, basic attributes come before advanced ones, lower level before higher levels, and so on.

In many cases the lower level attributes may be skills, while the higher level versions of the same attributes may be talents. This allows a player to practice with an ability at a low level and then upgrade it at a higher level in a way appropriate to the character's Life Path.

For example, consider "invisibility":

As with the current game, a certain number of new skills and talents become available at each level, and, a certain number of each can be chosen in toto.





Still, why go to all this effort? What's the payoff?

Fundamentally, WOO II is about entertainment. Players need enough structure to make game play predictable, but they also need enough variation to keep it interesting. The advantage of Life Paths is that there are three primary, and more or less equally robust, ways to play the game ... four, if Questing is treated as a separate game path. Each of the four pathways has a structure ... a set of rules that guides Player decisions and sets up their play.

The current approach is based on "classes", a D&D concept that made sense in the adventure/story milieu of its time, but which does not work very well in the combat/venue milieu of the current MMOG. But there is nothing sacred about classes, they are just a way of arranging character attributes. Life Paths enable the transition from a story based game to a venue based game - they change the method of arranging attributes to fit the way that characters now actually behave.

To play the full game today, a player needs to create three characters ... one to fit into a dungeon raiding team, one to engage in PvP combat, and one to handle farming and crafting. The Life Path approach simply acknowledges what has already happened, and reorganizes the process to simplify choices and add variation, making it more fun for players, and easier for the game designers to manage.

Foreseeable Problems

Conceptually, this is a very large change. Here are a few objections that might arise. (No doubt there would be others.)

The Class Is an Integral Part of the Game and Should Be Preserved

That boat has sailed. D&D classes were created to have specific weaknesses that would be balanced by specific strengths of other team members. A team always had one of each class and their combined efforts were required to deal with the situations that arose in the game.

The current game has already demoted classes to a subcategory under role. A Tank, for example, has strengths that the raid team needs if it is to succeed ... and the Tank's weaknesses are balanced by strengths in the team's Healers and DPS. The Tank's class, however, may vary ... Paladin, Warrior, Death Knight, Druid, Brewmaster ... any of these classes may serve as main tank in a raid. Similarly, the mage's abilities to cast damaging spells have been distributed among other classes, including Priest, Shaman and Warlock.

What About Multiple Specifications?

Technically, a single character could have two Life Paths, or all three ... but it's hard to find a reason for implementing things that way. Physically, within the game, the characters would always be in different places doing different things. Since characters share items and bank accounts, and since switching characters can be done on the fly, it makes a lot more sense just to create two, or three, characters, one in each Life Path.

There is no reason why each character could not have two or more "specs" within his or her Life Path. This would allow one character to shift from Farmer to Factor, for example, or from Flag Carrier to Damage. The Life Path term for these is "occupations", and a character with multiple occupation specifications could play the entire Path.

However, it would be easier, and more effective to introduce a "cloning" option. This would let a player generate a copy of any existing character on his or her account. The character would be created at the current highest level of the existing character, the player would chose an Occupation for the clone, and the copy would be given a standard set of gear appropriate to that Occupation. Cloning would allow players to add another character in the same Life Path but with a different Occupation ... essentially the same process as creating a specification.

The advantage to this is that it makes life much easier on the Player. If a change in battleground choice necessitates a switch from flag carrier to dps, then all that is needed is to switch in the dps character. This eliminates the "respecing" activities such as equipping new armor and rebuffing.

The Invisible Plate Wearing Melee Mage

Won't players all gravitate to one immensely powerful set of activities?

No, because:

Players have at least two pressing constraints on character construction. The character has to do something in the game, and the player has a favored playing style. Regardless of the attributes trees, the character still has to tank, heal or dps in a dungeon, or fight PvP battles, or craft items and sell them. Players may try to optimize a character for a particular role, but there are a variety of ways to do this.

Further, there are a variety of ways that any particular role can be accomplished. For example, a PvP healer may want to emphasize mitigation for survivability ... or damage with less mitigation ... or damage and mitigation with less healing ... there are many possibilities. Players tend to create characters that fit their playing styles. This is hard to qualify, because there are so many players, but while certain attributes are "no brainers", many will be chosen because the player wants to do things in a certain way.

Finally, most players want to be unique. While some will no doubt try to find the single "best" way to play a role and choose that particular set of attributes, the majority will include a few attributes to make their character different. (And, with a wide variety of talents and skills available, the definition of "best" is going to be arguable.)

Total Specialization

What if a player builds a character that is over-balanced in one dimension. Imagine for example, a character that chooses only offensive damage actions with no mitigation or utility actions at all.

If the game rewards that, why not? A combat player with no defense is not going to survive long in either PvP or PvE, though. And the more damage attributes a player chooses, the more buttons he or she has to push to play the game. There are a number of similar trade-offs in play.

Probably, the place where that is most likely to happen is in Crafting, where a player might decide to focus 100% on buying and selling, for example, and omit farming and making as much as possible. It's hard to see why this would be a problem ... though ... players should design characters based on what they want the characters to achieve. It's up to the game designers to make the game world reward balance and penalize obsession, if that's what they want to achieve.

Won't Players Get Confused?

Probably, some will. Expanding the variety of options also increases the possibility of making bad choices by those who do not fully understand what they are doing.

The solution to this is for the game to offer a selection of "template" choices: Healer, Tank, Damage, Utility, for example, in Raiding ... or Healer, Damage, Flag Carrier in Fighting ... or Farmer, Maker, Factor in Crafting.

Thus, the player could choose a basic template at level 1 and have the game present a limited number of choices, based on that template, each time new points become available. Players who wanted to make their own choices could turn the templates default option off.