MMO Community Enhancement
An in-depth list of community improvements for MMOs, both new and old.
by Trenton Kennedy
January 2009
What is “community”?
Community is that nebulous thing that exists inside and outside of a game. It can show itself on any website, forum, club or even t-shirt. In regards to an MMO, it is the all-encompassing micro-society that has formed surrounding the game.
More importantly, a community inside a game can best be described as a culture. There are player relationships and associations (both huge and miniscule), new language, territories, key figures, memes and even stereotypes.
The goal, of course, is to promote the growth of new and
old relationships between players. Bond them to each other while bonding
them to the game.
Assumptions:
1. The game is good enough. Community will do little to retain or attract players when the central game is mediocre or just bad. There needs to be a reason to play in the first place.
2. Good community extends the life of (specifically multiplayer) games.
3. Players will
always act like mature adults; any system you add will be greeted with
appreciation and result in exactly what you expected. Players
will be players.
Purpose:
1. Explore what past MMOs have done to encourage good community.
2. Improve on past iterations and consider the merit of new enhancements.
3. Suggest improvements that give the most benefit in the least amount of development time. (The most bang for your buck.)
4. Offer (sometimes off-the-wall, potentially infeasible) ideas that have the potential to spur thought and improve the next generation of MMO.
Other Notes:
Many of these ideas depend heavily on the type of game that is in question. What might work perfectly for one, could be the wrong choice for another. I try my best to offer different solutions for different situations. At the very least, I hope to promote thought and exploration of new ideas.
Disclaimer:
I cannot claim all of these thoughts as my own. Although there is a significant amount of original work, much of it is inspired by others or may already be a prominent or well-known idea. I’ve tried my best to give credit where it is due. If you think you deserve to be mentioned or if there is something specific you’d like to add, please get in touch with me!
Some of the pictures listed are from existing games (and I
will mark them as such). Others are basic mockups I have done.
They are merely shown to give the viewer an idea of what I'm talking
about. I do not profess to be an interface
artist!
Table of Contents
The most important thing we can do to enhance community is aid in the formation of new relationships between players. Existing relationships are out of our control. While we should encourage and help players to play with their existing friends (and the friends of their friends), those relationships do not need to be catered to. In fact, from a business standpoint, the more friends a person has from different groups, the more likely they are to stick around. If someone's only core group of friends quits the game, they're likely to follow. By varying and exposing a player to more groups and potential friends, players can form those new relationships and keep playing our games.
Bill Fulton of Ronin User Experience gave a talk at the 2008 ION Conference directly relating to this issue. He has also written a very relevant and insightful article that appeared on Gamasutra which dealt with his experience designing social aspects of Shadowrun (a multiplayer FPS). I encourage you to watch and read both.
In his ION talk, Fulton explains that these new relationships we're trying to cultivate primarily form from "repeated contact with like-minded strangers". Breaking this apart, there are three main problems to address.
To form new relationships, players need to:
1. Meet like-minded individuals
2. Have repeated contact
3. Remember them
Here's one possible solution that addresses all three:
The purpose of this system is to help players remember who they have interacted with. Furthermore, this is a form of social design and will actually encourage new relationships. Having a measurable meter for another player, especially when said measurement is called an "Ally Meter", players will approach the situation with a different mindset. For the sake of creating a cooperative and enriching environment, other players should not be viewed as competition (unless you want them to be, in a certain circumstance).
Interacting with other players in several different ways, increases your ally meter with that person.
All point gains for a specific person are on a diminishing return that decays/resets over time.
Points themselves will decay over a longer period of time. A "high point" marker will be displayed on the Ally Meter showing players who used to have a higher Ally score.
Shout to Allies - Send a short message to all allies within a certain threshold. Message appears in Allies tab. Used for finding a group, etc.
Blackmark - Players can blackmark another person. A small text entry window will appear, allowing the player to make a note of the reason for the blackmark (optional). A blackmarked player will not appear in the Allies tab, nor will the blackmarked player have the person blackmarking them in their Allies tab. On character windows, a blackmark symbol is shown for that player, instead of an Ally Meter (players who have been blackmarked will see the blackmarker with an extremely low Ally meter).
Map Markers - Allies above a certain threshold will appear on the player's map. (Threshold can be adjusted by player.)
Filters - Various filters should be easily usable to allow players to sort and browse Allies. Ally Meter level, class, zone, level, etc.
Friends and Guildmates do not show up on the Allies tab because they have their own tabs; however, both still have Ally Meters (that look noticeably different).
Friends and Guildmates gain Ally Points at a much faster rate.
Potential Problems:
Overabundance - Depending on how much time the player spends in the game, he or she can quickly become "acquainted" with too many people. A significant amount of testing and balancing should be done to ensure different playstyles can benefit from the system. It may be a good idea to put Ally Meters on a bell curve, to ensure a critical mass of acquaintances.
Shared vs Unshared Meter - If at all possible, a "shared" Ally Meter system should be implemented. Meaning, the meter Joe the Warrior sees for Rachel the Rogue is the exact same meter as Rachel the Rogue sees for Joe the Warrior. This is to ensure a player know that a highly rated Ally views the player also as highly rated. (An unshared meter would be more of a "point of view" for that specific player and would have a less desirable effect.)
Bad/dumb Suggestions - An automated system like this is bound to have some problems. It may be worth giving a bit more control to players on who populates their Ally List. A Digg-style up and down voting could help keep players they like rated well and players they do not like rated lower. However, be wary of giving too much control, as it would defeat the purpose of the whole system.
Extensions:
Interaction History - List of activities between you and that player. Used to help players remember what interactions they have had (such as grouping, trading) with others. This can include a (searchable) chat log for /tells and even group chat.
Ally Meter Perks - Depending on the overarching design, it may or may not be worth adding associated "perks" to having a high Ally Meter. Be wary of making perks too good as then you'll practically be asking your players to game the system. Perks (if any) should only be used to build relationships between strangers and acquaintances, not advance or improve a player's performance.
In T.L. Taylor's paper, The Social Design of Virtual Worlds, she states, "Once in the game, design decisions shape the possibilities for the environment through the choice of avatars they provide, how communication can occur, and more generally how the 'world' gets modeled."
Environment directly affects behavior. Therefore, whoever controls or creates that environment will have a significant influence over behavior. This is yet another thing game designers have to keep in mind.
If a specific set of goals are in place for players, naturally most will follow the path with the least resistance. We have no direct control over the path they choose, only the (environmental) resistance and the goals.
When Warhammer Online was released, one of the biggest innovations it brought to the table was "Public Quests". While they have their share of problems (mostly due to population), reviewers and players alike hailed the addition as a fantastic step in the right direction. Tobold, a popular MMO-centered blogger, explains it best:
“So what makes public quests so much fun and so much different from lets say World of Warcraft gameplay?
In WoW in the open world, other players are basically your enemies: The last thing you want if you are on a quest to collect 10 foozle ears in a certain area is to arrive in that area and find already half a dozen other players there hunting foozles. They'll kill "your" foozles, and force you to wait for respawns.
Even if you could persuade them to group with you, you'd end up getting your quest items slower than if you had soloed with no one around.
In a WAR public quest, other players are automatically your allies.” (Emphasis mine)
Having multiple servers has traditionally been a technological necessity. Luckily, this necessity coincides with the goal of having manageable communities. Players can only be familiar with a limited amount of people. It's important to give players a sense of belonging and a community.
Unfortunately, when a player's group of friends quits the game, they're much more likely to follow. Frequently, the player may know of others that play, but that specific group may be on a different server. Catching up to them may be infeasible.
So a balance must be struck between maintaining a manageable sized community and enabling friends to easily play with one another.
Possible Solutions:
All of this, of course, comes with a large caveat. As it says in Raph Koster's Laws of Online World Design:
Ownership is key
You have to give players a sense of ownership in the game. This is
what will make them stay--it is a "barrier to
departure." Social bonds
are not enough, because good social
bonds extend outside the game. Instead, it is context. If they can
build their own buildings, build a character, own possessions, hold
down a job, feel a sense of responsibility to something
that cannot be removed from the game--then you have
ownership.
Whether it's through player-generated content, housing, community events, character development, or something else, it's vital that players feel attached to the game somehow. This should be central to the overarching game design.
Guilds, Player Associations, Allegiances, Corporations, Clans, whatever you call them, they are the micro-communities that spring up in any multiplayer game.
For many players, everything they do is affected by the guild they belong to. When they log on, their first reaction is to check the guild panel and see what everyone is up to. Or, maybe they’re logging on for a scheduled raid or guild activity. With most current MMOs, there are very few in-game tools to help guilds keep track of events, members and their activities, guild progress and goals.
Guilds are just mini-communities and are vital to keeping players interested and involved. There are several tools and improvements that can be implemented to encourage community development.
Generally, MMOs limit a player to one guild. While it’s good to encourage association between players (they would form regardless of if there was a system in place or not), it can be limiting for a player who has friends in multiple guilds. Often, players are forced to choose between staying with their friends in a casual oriented guild, or joining a competitive raiding guild. Either choice has merit, but ideally, the player should be able to do both.
Furthermore, there is an untapped potential for more player-created mini-communities, such as crafter or class specific associations (i.e. The Warrior’s Guild, or The Tailors Union).
Problems to solve:
Issues to watch out for:
Possible solutions:
- Simple “Dual” System
Players can choose a primary and a secondary guild. The primary takes precedence over the secondary. Simple.
- Guild Archetypes
Several categories of player associations exist. Players can be in one “guild” per category. At anytime, players can choose to change their tag/tabard to one of their guilds. Each guild and rank the player is in is listed on their public character sheet.
Possible categories would depend heavily on the game setting, some might be:
- Emissary System
Players have a primary guild they belong to, similar to existing systems.
Players also can be recruited into up to <arbitrary #> other guilds as “Emissaries” or “Ambassadors”. What the emissary gains access to (which chat channels, guild bank, calendar, etc.) is determined by the guild leadership and may change from guild to guild. This allows the player to have a primary guild but also keep in touch with other guilds.
Scheduling events or raids for guilds hasn’t always been easy. Most of the time guilds have had to rely on their website for coordination. The downside is, not everyone checks the website or forums on a regular basis. Adding tools to support activities can greatly enhance guild player involvement.
Some of the most popular player-made addons for World of Warcraft have been calendars and guild organization tools.
GroupCalendar is one such addon.
Fortunately, many MMOs have been adding support for in-game guild calendars (or have had it for awhile!). This is a step in the right direction, but more can be done.
Basic Philosophy: Look at what has been done and take the things that work. Google Calendar, Facebook events, past Addons, can all be sources of ideas.
Some Suggestions:
- Facebook-style invitations
- Invite players (offline and online) to events
- “Who’s Attending” list shows who is definitely attending, who “may attend” and who is not attending
- Open, private and hidden events
- Event scheduling support
- Ability to associate events with a Raid Group or Chat Channel
- History of player activity in raid events (time joined, time left, attendance, etc.)
- Support for reoccurring events (quick inviting for past attendees)
- Class/level/item-based limits
- A maximum of 5 Druids/Warriors/etc. can attend
- Only levels 65-68 may attend
- Only those who have “Key to Evilville” can attend
- Group invite assistance
- Automatically invite attendees to your group
- “Ready” button for attendees to show that they’re ready for an invite
- Open group implementation, attendees can add themselves to the group (See Open-Group section)
- Out-of-game event support
- Creating/editing/joining events on website outside of game
- See Outside-Interaction section
In present day MMOs, player progression is often directly correlated with guild progression. While this in itself is a design choice, it cannot be denied that having larger goals that require more than one player to participate can do great things for the community. Players should always have a purpose.
Many games have attempted to add different types of guild progression, with varying results.
EverQuest 2 has guild levels that can be achieved by doing certain quests and activities (such as killing boss NPCs). Mounts, housing items and titles are just some of the special rewards gained by having a higher guild level.
Having a viewable history might be another way to reinforce guilds being seen as living, breathing entities and give players a way to show their achievements.
Character customization and development are central to any MMO, but depend heavily on art direction and design choices. Some (such as City of Heroes) open up visual customization to everyone, regardless of class or development choice. Others, like Warhammer Online, being a PVP-centric game, chose to limit visual uniqueness between players of the same class for the sake of easy identification.
Regardless of the limitations that might be present, players want and need to feel as if they own their character. It’s important to remember that development should not be purely stat based, how others and players themselves view their character is just as vital. Give your players a chance to define themselves.
City of Heroes allows players to decide every detail of their appearance, regardless of class choice.
Star Wars Galaxies
Examples:
- Hometown
- Inspirations
- Fears
- Weaknesses
- Virtues or Vices
- Goals
- Weapons of Choice
- Relations/Family
- Personality
Many popular MMOs feature character title systems. Some grant
new titles based off of a rare, significant achievement; others, you
might not be able to walk more than a few feet without being granted
a new one. Titles themselves range from humorous (the Meat
Shield), to serious (the Brutal Gladiator), to descriptive (the
Master Blacksmith), to silly (the Extra Squishy).
Warhammer Online grants titles based off of achievements.
Titles are a reasonably inexpensive way to add a bit more
customization and personalization to player characters. Not
only can they be a fun addition, they can help a player further
define their character. Furthermore, titles are just one more
thing that can be used as an incentive or reward.
Much more can be done to innovate on existing systems. While
funny/silly titles have their place, there are other opportunities
that should be explored as well. Here's one such example that
can greatly improve on existing systems.
Rather than award players with preset titles, award them with "parts" of a title.
Now, Bob has choices to make. Using drop down menus, he can
choose each piece of the title.
Using the syntax:
"The"/"A"/None + [Adjective] + [Noun]
Bob can now select:
This allows players to mix and match different words, to create
their own titles.
The other option, of course, is to entirely open up the process of title creation. Players are already allowed to name their characters and guilds, so why not titles?
This might be inherently terrifying to the little customer service representative inside all of us, but the benefits could debatably outweigh the risks. And again, there's already a system for naming guilds in place.
A small cost or cooldown would discourage players from changing the title too often. Normal format and profanity filters could be applied, similar to guild names.
When Warhammer Online: Age of Reckoning was released, there was
a relatively minor but prevalent complaint about the lack of a
public or "general" chat. While there was in fact regional
chat (similar to WoW's /1 channel), it was rarely used.
Coriel, author of the blog Blessing of Kings, had a
pretty good
explanation of why things seemed so "empty". The
main reason seems to be an overload of information.
Nick
Andrew Quagliara , a graduate student at Indiana
University, did his Master's Thesis
on Chatting
in MMOs. Many of his results coincide with the
complaints heard about WAR and others.
For the sake of compilation, I'm pasting these below, but I very much encourage you to read his article on GameCareerGuide. Again, all credit to Nick Quagliara.
Certain tasks carried out through the chat communication interface are repetitive or require no decision making as they are always carried out. Rather than placing the burden on the user to carry out these tasks, allow the user the option to automate certain tasks.
Typically, chat communication windows relay an abundance of information to the user. Often this information seems unnecessary. In order to reduce the barrage of information presented to users, reduce information to only that which has significant meaning for the user.
The chat communication interfaces in MMOs appear to be either working off of the IRC paradigm or counting on the users to have previous familiarity with MMOs. This can often lead to confusing interactions with the chat communication interface.
Look to the mod community.
The proliferations of certain mods points to the needs that users have that are not being properly met by the first-party interfaces. Mods that are widely embraced by an MMO community should be incorporated as part of the first-party interface.
Simplify when possible.
Although this is easier said than done, try to simplify the chat communication interfaces. Perhaps some of the functionality of these interfaces is unnecessary. The navigation of submenus can be restructured.
Alert the user to messages.
Some messages are lost because the user is occupied with other tasks. Some messages are lost because messages are scrolling too quickly. Allow users to set up a system for being alerted to specific types of messages.
Gradually present some functionality.
New users are expected to learn complex interfaces in a short period of time. In order to alleviate some of this burden, gradually present functionality. Certain aspects of chat communication functionality are not integral to initially learning a new MMO.
Give users access to information relevant to their play.
Users often have to turn to public chat in order to find information that is relevant to participating in an MMO. This adds to the number of messages that appear in the chat communication window. Users also have to minimize their MMO windows in order to check online forums. Allow players access to this information while in game.
Keep consistency with interactions.
If right clicking a particular piece of an interface produces result A, then right clicking on a similar piece of the interface should produce a similar result. Although this seems axiomatic, problems of this nature occur and can lead to frustration for the user.
Provide different functionality for different users.
Not every user desires the same functionality. This can be seen in the mod community where some mods are specific to certain character classes. Allow the interface to be optimized for different play styles as well as different avatar types.
Consider icons in place of text commands.
Icon-based interactions are the standard for users who are used to the desktop metaphor. Allow users to carry out certain actions through the use of icons.
Give the user the chance to opt out.
Some players are not going to embrace changes to either current or future chat communication interfaces. Allow users to opt out of changes to MMOs that they already participate in. Consider options for future MMOs such as a "classic" interface setup for users who are ingrained in the current system.
As Scott Jennings aptly says, "MMOs are special. MMO communities are special. They require a special, deft hybrid form of public relations, rapid response, and disinterested ombudsman."
Interaction can be just as important for the community outside
of the game as it is inside.
You will never find a more wretched hive of scum and
villainy. We must be cautious.
A forum can be a source of information and community building for your players, and it can be a giant headache filled with whining, abuse and ASCII art.
Rather than attaching a karma rating to forum accounts (which can lead to griefing), allow users who are logged in (and subscribed) to vote up or vote down specific posts in a thread. Posts with a low enough rating will have a negative tint (red shade, etc.) to them and will eventually be collapsed (user defined, à la Digg) if low enough. Higher rated replies and posts will be tinted positively (green, blue, etc.).
It might be a good idea to only use this sort of system on the more volatile boards, such as General Discussion (at least to first see how the community responds to it). The tech support forums might benefit from such a system as well.
Moderators would also benefit from this system. Not only
would there already be an inherent community-based moderation,
but official moderators could be given the power to raise or
lower post scores. This would allow them another level of
moderation, one that is invisible to users. For example, a
particularly good post could be "boosted" up by a moderator or a
troublemaker's post could be modded down; this would be
indistinguishable from community voting (no one would be able to
tell the difference).
In the article linked above, Blizzard co-founder Frank Pearce speaks directly about the potential for allowing players to "hook" into their character's guild chat from a cell phone. The same might be possible for a community site.
Even if real time chat is infeasible, a Twitter-style, asynchronous type "chat" might be a better alternative. Tie it directly into the game (through a guild/group panel with notification options) and you have yourself a new connection to the outside world (or, more importantly, your players have a connection back into your world).
Another popular suggestion is adding the ability for players to view and possibly even control their characters' inventories or auctions through a website. Whereas it may or may not be worth doing it for a game like World of Warcraft, others, such as Eve Online, that have larger macro-economic levels of resource exchanges, might benefit more from having outside control over auctions.
Many players enjoy monitoring and actively using the Auction
House as sort of a metagame. They constantly attempt to outbid
or outmaneuver their competition. Having access to this
information on a website or phone might be the next logical
step.
With the casual game market booming, tie-ins and spin-offs might be inevitable. Luckily, this is another opportunity to connect players back to our virtual worlds.
There's room for significant growth with simple IP-related (or connected) web and mobile games:
Furthermore, adding a way for these sorts of mini-games to affect players' in-game characters would greatly improve value (and fun).
Small rewards, such as temporary potions or lower-value tradeskill items, could be given in these types of casual games.
A risk vs reward could also be added through gambling or token-based games. Allow players to buy a special currency (tokens) in game that can then be gambled with other players on the community website, or through single-player (web-enabled) mobile games.
Give me a way to play Texas Hold'em with my characters' currency and I'll be a happy, happy (unfortunately unproductive) person. Add a computer controlled dealer who plays and you can even have an additional currency sink.
I'm going to assume we agree that metrics play an important role in development (especially with MMOs). If there's any contention, go talk to Darius or Sara.
Besides the core uses of gameplay metrics, they can also be used to aid in community development.
Knowing and keeping track of social groupings can have great applications for community building. As Darius Kazemi explains, it's actually quite easy to show which players are connected socially to each other by using the friends list. This can be extremely helpful in spreading live event information through word-of-mouth. It might even be possible to automate a system that would greatly enhance the potential for server-wide events, without needing GMs to do it manually.
They can also be used to show how successful certain systems (such as the Ally Meter!) are. Looking-for-group systems could tap into this information to give preference to a player closer in the social network, or even to someone who's far away in the network (to attempt to knit the community closer together).
These social networks can be analyzed closely to aid in player retention. What effect does more friends have on the likeliness of a player to quit? Keeping track of where, when and how players meet for the first time can be used later on to see which correlate to lasting relationships. Are there certain situations that are more likely to aid in new player relationships developing? Knowing these, how can we improve on other areas to have the same effect? These types of questions can be valuable in prioritizing development time, but can generally only be answered by having effective metrics.
And, of course, gameplay metrics can be used to catch players who are up to no good. Darius shows in another article how to catch a gold farmer. Furthermore, certain exploits such as gold-duping are bound to show up quickly if there are systems in place to look.
Sara Jensen Schubert explains that it can be helpful to "augment that hard data [metrics] with more subjective evaluations", those being: surveys.
Some MMOs have patch surveys that launch while the patch is
being installed. Pirates of the Burning Sea takes
occasional
surveys to see how players like a new patch. Lord of
the Rings Online has
offered
surveys in their launcher for things like PvP balance.
Good community features will not make a bad game good. It will, however, push a good game into greatness. It can be a significantly meaningful experience to see players come together, forging new relationships with others of all walks of life, centered in your game. Even from a business and monetary standpoint, it makes sense to encourage community building as it can vastly extend the life of a game. I hope some of this has at least encouraged thought and discussion on the subject.
I welcome any comments or feedback. I love discussing and analyzing this stuff, so feel free to contact me anytime.