Dream On

"In the game you can sing to butterflies once you get a certain achievement," Butterfield says, offering one of "dozens" of examples of self-defeating player behaviour. "It takes a big chunk of energy and gives about half of that back as experience points. It's something we imagined you would do once or twice, but you wouldn't do it a lot because it's not a very good return."

"There's also a drink in the game that gives you a buff that means your interactions with animals don't cost energy. These two features were developed at different times; the intention of the 'girly drink buff' was so that you could sing to butterflies, squeeze chickens to get grain, or nibble piggies to get meat without it costing you any energy. It was pretty well thought out and balanced."

In no time at all, players in their droves were gulping girly drinks and clicking on butterflies as quickly as their index fingers could hit the mouse button. The rapid ascent of certain players through the game's levelling system caught Tiny Speck's attention, and after a brief investigation the root of the problem was identified.

"So we took it away," Butterfield continues. "The people that were doing it freaked out. They were like, 'that was the fun part of the game for me.'"

"I can sympathise, because finding those magic combinations can be fun, but at the same time if you create an incentive for people to do something that's that boring - they log into the game and just click on the butterfly for four hours, then stop, and come back the next day and do it for another four hours - at some point they're going to say, 'Why the hell am I doing this. I'm just running a number up and nothing else is happening.'"

"Running a number up and nothing else is happening" - it wasn't Butterfield's intention, but these words could serve as a particularly sceptical description of the experience of playing a social game. Indeed, Glitch has been likened to the games of Zynga, Playfish and Playdom, but the comparison clearly makes Butterfield uneasy.

"We get asked all the time what the demographic is, and we say, 'people who like the game.' Definitely not meaning to be flippant or glib or anything, but it's not the kind of thing where you can say, 'Women, 35 to 50,' or, 'Men, 18 to 24,' or any of those characteristics. There will be all kinds of people for whom it holds appeal. We're designing it on its own terms."

We weren't trying to have the biggest possible audience because then very few people would really truly love it

As Butterfield sees it, there's a uniformity to social games that contradicts Tiny Speck's core values; a degree of commercial consideration to their design that gets in the way of original ideas and unique experiences. Tiny Speck wants Glitch to be "strongly flavoured, like cilantro", even if that means leaving certain players in the cold.

"We weren't trying to have the biggest possible audience because then very few people would really truly love it. A lot of people eat at McDonalds; not a lot of people have such brand affiliation with McDonalds that they would wear a McDonalds T-shirt or have a McDonalds sticker on the back of their car. We want to build something that some people will truly love, which means that other people won't like it. That, right there, kind of implies that the audience is going to be smaller - it's not going to have 100 million people playing it like CityVille."

"At the same time, if people love it they can be more deeply engaged. We should be able to be a good business with a smaller number of players. The intention on the business side is to design something that is sustainable and does slightly better than break even with a couple of hundred-thousand players. It's certainly been done before - again, EVE Online is the canonical example of that."

Part of that "flavour" is Tiny Speck's interpretation of the freemium business model. Early in the development process, the team took the decision not to charge for anything that offers a "game advantage" - a prominent revenue stream for most social games. They didn't want their players to hit roadblocks demanding either a cash payment or a protracted waiting period before the game continued. On the flip side, they didn't want any player to be able to buy their way through the game.

"If I can always be ahead of you by paying more money it will ruin the game to a certain extent," he says. "Obviously, it can work - that's what Mafia wars is, right? It's like a test of who's willing to spend more money. If I can spend more money than you I can beat you every time."