MMOG communities were an interesting (and surprising) addition to today's program at Spring 2006 VON. Robert Moore from the Palo Alto Research Center guided a panel through a discussion of how voice collaboration (via VoIP) is changing online gaming:
- Ron Meiners, online community consultant
- Keith Weiner, DiamondWare
- Nick Yee, the Daedelus Project, Stanford University
Moore laid out a most excellent overview on the dynamics of voice and text communications as used in massively multiplayer online games (MMOGs):
- Voice is the norm in first-person shooter games such as Halo. Like the real world, these games are characterized by transient arenas, successive rounds of interaction, and a quick pace.
- In role-playing games such as World of Warcraft (WoW), text chat is the norm. The world is persistent, and you only interact with characters that are at a level similar to your own. Most RPGs don't have built-in voice, so players will download add-ons that deliver voice capability. Raiding an enemy camp, for example, needs to be tightly coordinated, so voice collaboration is essential. The group activity in this type of attack looks a lot more like first-person shooter interaction.
- Finally, there are creative sociable worlds, such as Second Life or There.com, which also use text chat. (Disclosure note: Omidyar Network is an investor in Second Life.) Socializing is important, and there is a substantial amount of player-created content.
Who plays these games? Yee's research shows that the average player is 26 yeard old, and spending 20 hours/week in the game. And these aren't all kids - 35% of players are married, and 20% have children. Yee then matched MMOG use to VoIP use, by gender and age, and found that VoIP demographics very closely map to MMOGs numbers.
This doesn't mean that everyone playing MMOGs should have integrated VoIP. Moore outlined the tradeoffs between text and voice in MMOGs:
- Input: Manual vs. hands-free. ('Nuff said.)
- Mode: Visual, persistent, asynchronous vs. aural, transient, synchronous. Text is better for multitasking; it allows reduced attention and does not require people to take turns. Voice is better for tight coordination; it requires full attention and forces people to wait for their turn to speak.
- Privacy: Real voice/identity hidden vs. real voice/identity revealed. Text enables a more consistent persona, but it's less personal. Voice is more personal, but it requires more social dexterity and reveals persona discrepancies. (e.g., if you're a small girl playing a tough guy, your cover is blown.)
- Disruption: Low vs. high. You can't very well use a voice-enabled game in the office, or while sitting with others watching TV.
30% of MMOG players use VoIP tools on a regular basis, and Yee has drawn predictive trends on user demographics from his research. Specifically, people focused on achievement and competition are more likely to use VoIP in the game. These players tend to be more aggressive, and aren't as concerned about being immersed in the fantasy world (given that the wrong voice can break the spell). These players don't use the VoIP tools to chat, but rather are using it to give orders.
One of the constraints for VoIP use in games is how often someone works solo as vs. works with a group within the game. Yee's snapshots of game activity reveal how frequently people play together as vs play alone. He's noted that as characters advance levels, they are more likely to join a group; but even so, only 30% of WoW play is with groups. Even though WoW is "massively multiplayer," players actually work alone 70% of the time.
As Yee says, "the ogre voice you can make up in your head is always better than the ogre voice that a 12-year old can come up with." The desire to conceal gender-bending is also a factor; 50% of the female avatars in WoW are played by male players. So, as these players' characters advance, they get into an awkward position -- how can you use voice, and start participating in group raids, when this will reveal that you're not really female?
VoIP can actually be a disabler under these conditions. Primarily, this is because voice introduces the capacity for discrimination in the community. Voice introduces social markers that are easy levers for discrimination. Yee's examples include teenagers leading guilds of adults, and players with an accent of non-fluent English. He also cited the introduction of the "English test," which many players apply to a new player before allowing him or her to participate in a raid. There's a particularly disturbing thread of discrimination against (and negative ethnic slang around) the many Chinese players. And the deaf players that enjoyed being equal in WoW become unequal as voice chat becomes a requirement for participation.
There's also a secondary form of discrimination introduced with voice: the haves and the have-nots. If you don't have voice enabled, then you can't play collaboratively at the higher levels.
All this being said, Meiners still likes voice for MMOGs because of the richness it adds to the experience:
- Identity: gender, age, familiar accents, and a better sense of personality
- Information: voice intonation has a great deal of meaning, but it gets lost in text
- Immersion: voice makes an interactive experience richer (but why?)
- Innovation: voice is enabling creativity, collaboration, and new ways for people to connect online; machinima is one example
The top ten problems for VoIP gaming (as Weiner introduced) boil this discussion down:
- Voice quality sucks
- A prior relationship is a prerequisite for voice
- Too many voices are audible
- "Cover me!" - where is that coming from?
- Multiple simultaneous voices
- You can't exchange the random "hellos" that initiate natural socializing
- You don't sound like an elf, or a man, or a woman...anything that you are not
- Dwarves think that elves should sound evil, but elves think that elves should sound noble
- Military-style 2-way radios
- Team intercoms
The panel outlined solutions that would address most of these issues. 3D positional voice was relevant to most of the items. Voice avatars could be developed and run on the client side, making someone's voice sound plausible for their character. (And hopefully, negating some of the discriminatory issues.) And bundling VoIP directly into the game would mean that players could talk to anyone, not just folks that they connected with outside of the game.
The concept of a "unique voice mix" for each listener was thought-provoking. A unique mix could not only filter out the noise of irrelevant conversation, but also introduce nuances such as tone. (For example, overlaying an evil tone for enemy players.) You could also introduce real-world elements such as the scratchy tone of radio interference, and make it worse for players that are farther apart from each other within game geography.
This panel was one of the more thought-provoking, as it dealt not only with technical challenges, but also real-world interpersonal dynamics.
Tags: christine herron christine.net space jockeys technology voip mmog von world of warcraft