DICE 09: Media Molecule On 'The Broad Church' Of User-Generated Content

Fresh off LittleBigPlanet's success winning eight awards at last night's Academy of Interactive Arts and Sciences award show, Media Molecule's co-founder Alex Evans talks about his studio's social, design, and production experiments with the PlayStation 3 game. Among the topics covered in his DICE presentation were the "the broad church" of user-generated content and the LittleBigPlanet team's struggles to make sure it was "not just making shit." "We had the super arrogance to think we could do this," said Evans, who began his talk with an amusing video of an early LittleBigPlanet prototype that served as a proof of concept for the procedural animation the founders had previously explored in Rag Doll Kung Fu. Though the graphics were incredibly basic, the prototype demonstrates the fundamental nature of the game's underlying control system. Selling An Unusual Experience "I still find this game hard to describe to people who haven't played it," Evans admitted. "With [Sony], we were trying to figure out how to sell this game in a 15-second ad spot." He showed a few examples of television commercials for the game, including a fairly inscrutable Japanese ad that was mainly live-action, and a clever European ad that featured a disgruntled man's voiceover over his "Janice Planet" LittleBigPlanet level, a visual metaphor for his recently-ended relationship with ex-girlfriend Janice. Reinventing The Wheel "We reinvent the wheel too much," said Evans of the game industry at large. "Why do we do that? I still don't know." One reason, he offered, is that "code is much easier to write than it is to read." Programmers have a hard enough time re-reading their own code, let alone someone else's -- so frequently, it ends up seeming more straightforward to just write internal code and, in effect, reinvent the wheel. But this isn't necessarily bad in all cases, Evans added. That mentality can end up giving team members strong investment in the game, thus increasing morale and a sense of ownership. Media Molecule looked at Valve's "cabal" process, which organizes the team into small groups of developers working on various specific parts of the game. However, when attempting to do that, the team ended up simply cross-pollinating into one big blob in which everybody tried to work on everything. That's fine up to a point, Evans said, but it falls down with overall studio growth. "That kind of chaotic process works up to 30 people, but beyond that it doesn't," he noted. The Broad Church Of User-Generated Content Media Molecule sees user-generated content as a "broad church," a component with many definitions and, contrary to some discussion, a long history, going all the way back to early arcade cabinets featuring user-inputted high score initials. It was important, he stressed, to avoid getting exclusively fixated on the tools element of the game, even though so much of its focus is on user-driven creation with tools. At the end of the day, a strong core experience is crucial to getting players invested enough to want to create content. "You have to remember that you're making a fun game," Evans said. "You have to approach it with a playful attitude." That philosophy should extend to the studio at large, from hiring practices to working practices -- and that mentality ended up being reflected in LittleBigPlanet itself, he explained. "You have to love what you're doing to make sure other people enjoy what you do." Keeping Focus "I don't want to paint a picture that we're just a bunch of hippies in a room," Evans added. "How do you make sure you're not just making shit?" he asked, saying that as a studio director he considered that question many times. He recalled Media Molecule going through many, many iterations of its core user creation tools, many of which were only practical for creating "melty lumps of shit." The user tools ended up needing to be powerful enough to create production-quality content. "We are just power users," Evans explained -- the PC-based tools the team uses are simply more efficient versions of the same tools that ship with the game on PlayStation 3. The decision to pursue that focus with the tools was one that required great effort and caused many headaches, he admitted, but it "paid off in huge spades." Why User-Generated Content? "What does user-generated content give you in the end?" he asked. "Well, it gives you epic pain. [But] why is it an interesting thing to put in the game? Basically, it's the investment players make when they have creative input into your game." That investment pays off in real-world terms. "When they've made something, it doesn't matter if that thing is good or bad," he noted. "They become evangelists, not just for the thing they've created, but for the thing with which they've created it." With a particular nod to the executive audience at DICE, Evans warned that staying too cleanly within the lines of what is already known to work can be dangerous: "You need people willing to take a massive freaking risk for no other reason than you feel it in your gut."