Originally published in Kotaku.
By now you've probably heard of Tringo, the Tetris-meets-Bingo game created in the user-created online world Second Life-- it's the casual game that became so popular in-world, its creator sold the rights for a Web spinoff, a GBA port, and now, incredibly, for a TV game show. This is possible because SL's Terms of Service allow subscribers to retain the IP rights to anything they create in-world. That feature, plus a potential audience of a quarter million SL subscribers and growing, plus the ability to sell the in-world currency you earn for real US$, and you'd think Second Life would be an indy game developer's dream platform.
But Tringo hit the world in early 2005, and since then, no SL game has come anywhere close to its success. When I was still on staff with Linden Lab, in addition to my work as the world's embedded journalist, I helped design and run Linden's annual Game Developer Contest, promoting game creation in SL. I brought in veteran designers like Harvey Smith (Deus Ex) and Doug Church (Thief: The Dark Project) to help judge the entrants; like me, they were impressed by the ambition and talent that went into them. But even with their imprimatur, none of the winning games went on to become the next Tringo.
After co-creating three such contests, and watching other SL games come and go, I've managed to synthesize my observations on what works and what doesn't into a handful of tips.
Keep It Simple, SLer
With so many events going on every hour, your window of opportunity to win over fans is tiny. They'll give you maybe 60 seconds to understand basic gameplay, and start having fun-- and if they don't, they're liable to go looking for it elsewhere. Much like Web-based games, the sensation-immersed Second Life audience demands instant entertainment, and aren't particularly interested in spending much time learning how to play your game. (Mastering SL's cumbersome user interface takes enough time as it is.) There are exceptions to this rule, mostly for games from recognizable genres with an established fanbase--"Dark Life", for example, a mini-MMORPG in SL that's been around for nearly 3 years, and has a decent (though not Tringo-big) following.
Develop for Lag
Since all content in Second Life is streamed onto users' computer, your game has to be designed on the assumption that there will be at least a little lag during play. Real time games that require high frame-rate, in other words, are right out. This was a painful learning experience for the creators of U:SL, an Unreal-style FPS created in Second Life last year. It was beautifully designed, with graphics almost on par with Half Life 2 or Halo, and elegantly coded--but when it opened to the public and the bullets started flying, lag promptly made it seem like it was underwater. This isn't to say no real time action is possible--the winner of the last SL game competition, for example, is Tech Warfare, a very cool, Total Annihilation-style RTS where hundreds of battle 'bots end up fighting at the same time. (But making the code lean enough for that to be possible was a painstaking task for its creator.)
Design for Sell
Creating a distribution and marketing campaign for your game is as important as the game itself. Many SL developers assume that a good game is all they need to succeed, but that hasn't been the case for the dozens of genuinely cool games lost in the ocean of user-made content already in Second Life. Hold regular matches, post them in SL's in-world Events listing, promote them with "evangelists" and even advertisements--this is just as key. A major reason for the success of Tringo is that its creator, Kermitt Quirk, created a franchise system for selling it. Land owners bought his game for the L$ equivalent of about $60, meaning they had to promote and host regular Tringo matches to make their money back. The result: Tringo games played throughout SL, 24/7.
Take a Gamble
Though many game developers are wont to look down their noses at casino games, the fact is that most of the successful games in SL are virtual recreations of the kind of automated games you'd play in Vegas. One resident named Games Prototype, for example, created and runs a franchise of hugely popular SL casinos and by his estimate, clears $2,000-3,000 monthly for about ten hours of weekly work. With so much content to choose from, giving residents the chance to risk and win virtual currency is a seductive proposition-- something that's even more important because there are no traditional "leveling" metrics for players to judge their success by. (Even Tringo has an important gambling element-- players in a match ante L$ into a pool, with the highest scorer taking all.)
None of this is to say that a game with wagering has to be casual or casino. The aforementioned Tech Warfare RTS comes with a betting pool so the competing teams can place a wager on the outcome. One developer even used the gambling craze to create a game that's sort of a morbid satire on it-- multiplayer Russian roulette, where you wager to be the last player alive.
Create a Community
No solo-play game in SL has done exceptionally well. Not a surprise, because despite all the talk about Second Life being an online world development platform, the fact remains that most of its residents enjoy it as a social gaming space. People are there to share content and enjoy events together, with each other. Besides designing multiplayer games, smart developers should create one or more SL groups that can function as clans, teams, or guilds, or at the very least, a communication channel for instant messaging fans of the game. They should consider launching a tournament system or other structures that give gamers a way of competing, teaming up, and most crucial, keep playing your game. (The killer app of Second Life games will be an Xbox Live-style matching system which makes it easier to find available games and other people online looking to play. The person who creates this will, in my estimation, truly pwn game development in SL.)
Even solo games can benefit from a community aspect. For example, the casual word game Boogie Board (
2nd 3rd place in the 2006 game competition),
can be enjoyed in single-player mode, but there's still a scoring
system so people can match their score against others. The ability to
export data from SL to the Web helps greatly with this, and suggests
another mechanism for building community: a website with top scores,
game recaps, and so on.
The final word for pwning game development should go to Eckhart Dillon, lead creator of Tech War, winner of this year's SL Game Developer Contest, which took in the L$ equivalent of nearly $2500 during the two months of its run. Eckhart's top recommendation is to find talented residents to collaborate with:
"I did all of my own scripting," he says, "but there is no way I could do it all myself, other people made builds and bots, textures." Above all, he adds, "Be patient. It took many months to make it good, I tweaked the hell out of it, I listened to what people said would be fun. The wonderful thing about SL is that you are RIGHT THERE with the consumer-- they can instantly tell you your product sucks," he chuckles, "or is good."