One other thing I would see as a difference is, essentially people are shipping complete experiences. Whereas social games evolve and change over time, and you have to be responsive to what's driving the users to stick around, and give them new things. You can't plan so much in advance -- you have to be more reactive.

HS: And that's a good question. Take a look at Magic Land, for example. It's very feature complete. And don't take this word badly -- because it's not complete. It will never be complete. But it's coming to market, considering what Facebook games have been, it's very feature complete. So it's a solid experience with six months of play, for example.

We have calculated how long it should take to play through; a hardcore player should play through in three months without paying, and the usual user is six months or something. So it's very complete in that sense.

But at Wooga, real work starts on the launch, so we don't plan too much ahead. So I would say that... I know things we will do in the next two weeks; after that is up for grabs. I have three or four big features I want to do in the fall, but I don't know in which order. And something might arise that precedes these ones. I don't know yet. I will know in two weeks.

So we do it by week by week. Usually what we do is plan next week, we release, then we -- based on metrics or feedback -- we plan the next week. So we do development cycles in one week. And yeah, usually we know next week, and a little bit of what is the week after that. But the third week is just best guess.

Do you update weekly?

HS: Yeah.

Are all your features only things that can be implemented in a week? Or do you ever take longer?

HS: No, sometimes we take a longer time. I think the longest one I've ever done has been a three week feature. But then each week we have had something smaller. So for example, a content update, because customers, they expect it. And if it feels the game is alive, new content is coming...

For example, what I sometimes do is that even though I had done some assets, I don't launch them; I keep them in storage. So if there comes a week that we can't launch a feature, we can at least launch something. So that's something you can do.

But yeah, you have to have weekly updates. Some companies do it twice a week -- we do it sometimes twice a week. So, it's the way to go. Otherwise users know that, "Okay, this is a dead game. They're not interested; they're not listening to our feedback; there's not enough content." You know, people consume content like crazy. It's like, put in the new house, they're going to have 15 of them tomorrow.

The design of free-to-play gameplay features, there's a lot of discussion of how the game designers have to think about monetization. That's something we can take as true, but a lot of the design of free-to-play features are very basic. Just convenience or energy, or something that you need to spend to play. What are your thoughts on the evolution of the design of the actual free-to-play features?

HS: Well, if you go back one year on Facebook, there was a lot of traction in this, I call it "contract mechanic", or some people call it "appointment gaming". So you have a farm and you plant something, and it will be ready in four minutes, or four hours, or eight hours, or 12 hours. Then, this works -- Millionaire City was a pure contract mechanics game, or appointment gaming game. It didn't have an energy mechanic.

Then, somebody -- I don't know which game was the first to have energy done, but FrontierVille had energy done really well. And we started hearing rumors that "this energy -- it's monetizing like crazy!" So because it's so kind of like, "Oh, I have a hurdle, I would like to do something." I play maybe 20 minutes for free and now I'd like to play more, but I have to either wait one hour, ask friends, or pay 1 U.S. dollar. And then there's a percentage that are not impatient enough.

So we have seen evolution from, first, Jetman -- advertising based, free-to-play -- so you know, they have ads, people clicked on ads, they went to another game, they got 10 cents, 50 cents for the install. So there, the customer didn't have to pay anything; it was an advertising business model.

Then we went to this "appointment gaming", where either you paid for time accelerators that, you know, you don't want to wait four hours; you want to have the benefit right now. For example, sometimes called instant build; we have it also as well in Monster World. You pay for that, so each click 10 cents, and some people pay for that.

And then the next evolution was the energy, so that the energy limits the play. There can be fusion. If you have some parts with contract mechanics, like the farming part in CityVille, and then you have all the rest with energy; that seems to work really well. Basically with like, in Magic Land, now launched, has the same -- it has a little bit of both.

Then, there's a recent trend that there's more resources you can run out of. So a year ago, it was just money. Then came food resources, or items you're missing. And you see that gradually coming more and more -- stuff you are short of. And this gives, obviously, the monetization place that either you grind the game and try to get the item, or pay a little bit.

Going forward in the future, I see this will increase more items you can run out of; Magic Land has a lot of things you can run out of. And then something I'm pretty excited about is; I'm very excited about character development. That's why Magic Land has a character. So, I would like to have abilities or powers to the character.

A player character?

HS: A player character. It's a little bit like improving your character in World of Warcraft; I would like to do it in a casual manner. And this obviously, you set skill points or something else. This can also open up new doors for monetization.