Many MMO developers are moving away from large packaged releases and toward the web as a platform for development -- including Raph Koster of Areae, Dan Ogles of Conduit Labs, and Scott Hartsman of Ohai. But why?
The group convened at Austin GDC to discuss. Ohai CEO Susan Wu brought some of the confusion about the evolution of the online marketplace to the forefront when she asked the panelists: "So what's the difference between a social game, a multiplayer casual game, and a social MMO?"
Ogles suggested that "the term 'social game' is largely meaningless," while Hartsman discussed a conversation he'd had with other developers.
"We came up with 10 different attributes that 'casual' means, whether it's budgetary, time of gameplay, ease of gameplay," Hartsman said. "What I am all about is trying to make accessible games, and that doesn't necessarily mean shallow."
Koster pointed out that keeping in mind how web users engage with services is more important, in this space, than defining games in a traditional way. "Those terms all suck, don't they?" He said.
"The big thing is that, on the web, you do see different definitions than you expect to see in the game industry for 'social' ... it's asynchronous... the engagement is there... but it's a different definition of 'social' or 'casual' or 'immersive'."
When asked what web developers can learn from game developers, Koster opined that game developers have more to learn. "I actually think the web guys have learned more from us than we have from them... I tend to think of the need for the information to flow the other way a bit more," he said.
"There is still plenty they can learn in terms of game design... but I see them learning it at a pretty rapid rate." Ogles agreed, suggesting that game developers need to understand web developers' "handle on development cycles and shipping early and often, and embracing the iterative loop and getting feedback from your users."
All the panelists have backgrounds in traditional, big-budget MMO titles -- so what are their biggest challenges in making the migration to the web?
"For me, the biggest challenge is really finding the happy medium that must exist somewhere, between what is pure chaos and hyper-rapid iteration on one end, and five-track parallel development with Excel sheets and Project plans on the other," said Hartsman.
He added, "Whatever production method we do end up with is going to look more like... the chaos end... than taking the big stuff and scaling it down."
This mention of product development styles launched into, predictably, a discussion of Scrum. Said Koster, "We use Agile and Scrum and a board full of sticky notes. Particularly in the rapid iteration phase it was the best way to do it... as the team's grown we've actually had to institute more kinds of process."
Ogles said his team uses Scrum as well. "The tricky thing for us is that there's actually a bit of a learning curve for using Scrum, and it's not a simple recipe... it's going to change depending on the organization... and we're changing it pretty much on a monthly basis."
From that, Wu asked what that changeability means for the design process. For Hartsman, "the old formal way of doing a multiple month pre-production and doing stuff on paper is largely not required anymore."
"In order to really start on any development you need to have a general idea of what your game is and the general idea you want to take it... in the environments we're in right now, we have to leave a lot more to iteration and testing."
Koster described his own move to web-based: "We have, from scratch, started over on the server three times, and we are on scripting language number two. From scratch, we have written the client five times in different languages and thrown away old ones."
"We have rebooted the tool suite from scratch four times -- in two and a half years. The thing is, never in that entire process could you not log in and use the product, from day one, for longer than 24 hours. And every step it has gotten significantly better."
Koster also noted the importance of click-tracking and harvested data for interface improvements, cross-referenced with survey data -- which is very much a web developer tool, rather than a game developer tool.
Agreed Hartsman, "That's something we never spent a lot of time with in the MMO world -- usability of interfaces."
So, Wu then asked, "What's more important -- usability, interface design, or gameplay?"
Koster, a self-styled big Facebook fan, said he tends to summarize all of the social network apps in one of two ways: "They are either profile entries -- to show off the things you like -- or poke. The interface sucks, but the gameplay is rich."
Particularly on the web, he added, "People think the interface is the game, and I think that is kind of backwards. I think the game is the game, and we should be thinking what are the many interfaces to it... you touch Twitter in many ways, you touch Facebook in many ways."
Many developers, like Ogles, use Adobe Flash as their primary platform. While it has some advantages, it also has many disadvantages -- one of which is that, despite its prevalence as a medium for games, Adobe does not see game development as its primary use (and improves the tool for its uses in video and rich web experiences.)
However, Ogles pointed to one advantage: "You can take an experience and break it out onto other sites... for us, we're music-oriented, so we're thinking about what a band will want to put on their MySpace page, what a user will want to put on their profile page."
Social networking, is in a way, the opposite of MMOs, Koster said. MMOs attract (primarily) lone users who form social relationships with other players, but social networks are pre-existing networks of potential players that have to be lured to play together.
"Working on traditional MMOs... we never had to think about 'Hey our game is fun, but how will it be more fun if you get your real life friend involved?'" said Hartsman, "but it's a huge part of what we do now."
However, Koster pointed out that the current state of the web is less than optimal: "A lot of web games suck now. Gaming on the web is a far less creative and more derivative industry... if you think we have no right to bitch, look at the web."
Wu posited that one of the biggest differences between MMOs and social networking games is that games on social networks are entirely designed for asynchronous play. How do you get users who are used to that style to move to a synchronous style of gameplay?
Hartsman said, "We need to make patterns of gameplay that fit their expectations... gameplay that's synchronous for a little atom of time, like one minute, or five."
On the other hand, Koster feels it's important that the entire game experience can, if the users want, be had asynchronously. "You need to fit into the way people use the web," he said.
"We've seen people who come in and start building a virtual world -- to them, multiplayer is 'a neat feature', not the point. I think many of these people will eventually discover their book club or knitting club is better with synchronous stuff, but you can't drag them kicking and screaming into it."
Finally, why did all of these developers leave their jobs at major MMO studios for the world of the web?
Koster's response seemed to sum up a lot of the inspiration among developers who have come to love the web as a platform.
"It isn't so much about the size of the market, though that's important," he said. "And it's not even about the six year long [traditional MMO development cycle]. To me, it's about the greater capacity for personal expression. There's a great capacity to touch people. it was about empowering everybody. Working on the web now reminds me of working on the internet before the web was around, where it was much more community-driven and participatory and exciting and end users have their say. It feels like you can accomplish everything."