Glitch, the new flash-based, massively multiplayer game from the mind of Flickr founder Stewart Butterfield, has already been called "a neverending feast of imagination" and "an intersection of World of Warcraft and Facebook." And it hasn't even started alpha-testing yet. To get the inside scoop, we spoke with Butterfield about its premise ("it harkens back to aboriginal Australian creation myths"), its possibilities ("when we make our APIs public, we expect to have our minds blown"), and how it pays homage to Sim City, Farmville, Super Mario Brothers, and more. Excerpts below:
There's been a lot of hype surrounding the launch of Glitch. How'd you dream it up?
I remember playing the original Sim City in the mid-'80s, and always wondering what it would be like to play from the perspective of the individuals. The way that game works, you play from the God's eye perspective, looking down on the people driving around in little cars--just a couple pixels each. But if you inverted that, if you played from the bottom up, that's something that's always captured my imagination.
So what's the premise?
Well, it opens a billion years in the future, and everything has worked out totally awesomely--people are peaceful, resources are plentiful, etc. But then lo, one day scientists discover that this world is in danger. And things start Glitching out, or disappearing, hence the name. The solution is to send people back to the past--to make sure the future actually happens.
Yeah. And as everyone knows, the world was originally spun out of the imagination of 11 great giants, wandering secret paths on a barren asteroid and thinking and singing and humming the world into existence. So the game takes place in the minds of the giants--you're a little person helping them grow the perfect world, so that it actually happens. [This premise] actually harkens back to aboriginal Australian creation myths.
I've read that Glitch is less violent, and more puzzle-based.
Most games are about fighting, and usually fighting comes down to getting a whole bunch of stuff ready to go, and then clicking a whole bunch of times. There's certainly nothing wrong with that, but if it's in there, it becomes the focus of the game. It's all about having better weapons or armor or becoming better at fighting so you can defeat other people. There's not really room for other kinds of activity. With Glitch, we've got a whole rich set of possibilities for human interaction.
Kind of like Farmville?
Actually, no. That's much more of a single-player game, where your friends can peek in at what you're doing, and you can peek in at what they're doing. But there's no way for one player to impact the outcome for another player. I can go and fertilize someone else's crops, but if another person who I know is better at growing corn than me, it's not like I can buy corn from her. The interactive comes from being connected through the social network embedded in Facebook. Glitch, on the other hand, is literally one big world. So when you're walking around, you can see other people walking around and you can get together with friends and go exploring and run into strangers. There's an actual supply-and-demand economy--and yes, it includes some farming.
The game's trailer seems pretty psychedelic. Is that the aesthetic you're going for?
We're going to draw from a bunch of different styles. Glitch is 2-D, so it will get hardcore gamers less excited about the graphics. But we've got a huge legacy of 2-D games that we get to both draw on and pay homage to--everything from Super Mario Brothers to Little Big Planet, which we really admire. Our aesthetic will vary from Japanese cutesy hyper-saturated pixel art to line drawings to graphics more evocative of a Tim Burton movie, with dark, weird curly Qs and twisted shadows. We're working with illustrators all over the world to produce some really amazing stuff.
How did you experience at Flickr inform the game's development process?
When I left Flickr, we were doing 35,000 database queries a second, and serving out maybe 10,000 images a second. Managing that level of scalability was a great experience. With Glitch, my engineering team and I are working to do the same thing--that is, deploy code to the site on a daily basis, and usually several times a day. We can deploy tiny incremental changes or big huge changes whenever we're ready. As soon as they're sent out, everyone's game experience is updated in realtime.
Sounds promising. When can we play?
Depends how early you get invited. We're about to start doing short bursts of testing, then gradually opening up to public beta this summer. Our official launch is fall--November, October, the earlier, the better.