Blogging the AGDC: Sulka Haro's Keynote on Open Ended Play

Sulka Haro, lead designer for Habbo Hotel, opened the second day of the Austin Game Developers Conference with a talk on "Fostering Open-Ended Play: Unleashing the  Creative Community." The conference has a heavy MMOG and online roleplaying contingent, but Haro spoke to the benefit and motivations behind creating an open-ended playground (f(players)=f(content)), spider-pigs, and five rules for creating playgrounds. "We’re always working to figure out what the users want next," said Haro. "You have to listen to them. But as a designer, I’m always falling into the trap that I know better. It’s hard to resist the temptation, but you should. I’m almost always wrong."

I’m going to talk about open play and the virtual property model we use. The reason I’m here, I suppose, is that we have something like 7.5 million users per month. We’re peaking at around 100k concurrently in our 30 hotels in 19 markets. We use the free to play, pay for stuff model. Thanks to Daniel James for coming up with that. We made something like $50 million dollars in revenue last year, so it’s a good model.

I’m going to start with our history. We’re really coming from the Web side of things. The people who funded the company really came from the Web, not gaming. We’ve since hired gaming people, because it’s good to have all different perspectives, but we’re really coming from a Web basis.

In 1999, the founders started Disco. In 2000, the guys got snatched by a new media company and started developing all sorts of other products with it. We had a snowball game, and it was the first place with functional items for sale. And we learned a lesson. It backfired. People hate when you can buy functional items.

Six months later, Sulake was founded as a company. The owner of the new media agency figured it would be a good idea to have a separate company. Hotel Goldfish opened six months later, and it had the concept of the room that you could sign up for and get. The retro pixel look started at this point, and it’s served us pretty well. If you look at the graphics, it looks pretty old, but it looks just as old now as it did seven years ago. It works just as well, and the kids playing the game don’t even know what “Retro” means.

The vision of what Habbo would be is pretty simple. WE knew that we wanted to give some kind of tools for people to do their own stuff. The couple of games we did, we realized that. If we did a proper MMO, it would have been much bigger than we could handle. So not doing a plot and just giving users a space was an easy way of getting up quickly.

We didn’t really have a business model yet and thought it would be great if it was around for a couple of months. And then we noticed we had 100,000 users. The business model was selling virtual furniture using premium SMS. It took us about five minutes to set up, and in Finland all the kids had mobile phone, so we could start billing them immediately. And micropayments with content authoring just made a lot of sense.

It kind of backfired again. People realized that if you sent a message “habbo sofa 100x sulake” people would think they could get 100 sofas when it was just a guy named 100x. Always expect the users will figure out everything and will be smarter than you are.

In 2001 we opened the first Habbo Hotel, branded for the UK. We retroactively branded the Finnish hotel. The users hated it, but it didn’t wind up being a problem. At that time, the UK kids didn’t have the mobile phones. So premium SMS didn’t work that well. WE switched from selling virtual furniture to selling virtual currency. WE didn’t realize how big of a change that was.

We’ve been developing it ever sense. We have 100s of payment methods implemented  globally. If you want to go global and get teenagers to buy from you, it’s a massive issue. There’s just no established market. Credit cards and prepaid cards are cool, but they just don’t cut it.

There’s a massive difference between selling furniture and selling currency. I’m not going to spill the beans, but work it out. From the point of sales and the user economy, it becomes more complex. It did add a lot of flexibility. We’re tweaking it now.

I don’t know how you do it in MMO economies where people earn the money, but it’s horribly hard to tweak it. There’s an ecosystem of trading as well. Teenagers get that they own some property and can trade it. Looking at the future, these people who live in virtual properties, eBay is going to be making so much more money than now.

Over the years we introduced rare items that were available for a limited time. People would hoard them, but they were fairly expensive. If you’ve been to GDC virtual property sessions, Daniel James has been talking about how some users might only put in a couple bucks, but others will put in thousands because the game enables them too.

With some of this stuff, speculation starts and there’s trading. Advertising is another revenue model, but we’re shifting away from it. Advertisers are getting that you can communicate with the teens instead of at the teens. We’ve been putting out branded furniture, which is something the teens love even more than non-branded furniture. We don’t know what’s causing it, but it’s working pretty well. And in the trading economy, the branded piece might be worth more than the non-branded piece.

That brings me to the in-game secondary market. We’re peeking in on the economy all the time. And some items are valued at over $2000 each. Some people are trying to monetize this on eBay, but it doesn’t work that well. The Habbos are teenagers, and they just don’t go. The secondary market is around $550,000 per year.

Who’s playing. IT’s about even boys and girls. They’re teenagers, 13-16. But if you go to Japan, there are the kids playing, but also lots of older users and housewives, but 13-16 is the predominant group. It’s the same age when teenagers are building their identities and becoming interested in the social aspects of life. As they grow up, the tipping point in the US might be when they get their drivers licenses.

It’s actually a little scary to think about the teenagers. If they’re 14 now, they were born in 1994, which is after the Web came out. They’ve been online their whole lives. The values in the way they’re using it is vastly different from any way we’ve been using it. The only thing they really care about is that their parents don’t find out what they’re doing online. We had a scammer we wanted to make an example of and tell the cops about it, but the only thing he was pleading for was to not tell his parents.

People are playing the Internet now. It’s not just use.

We’re doing a lot of research with a yearly global youth study. Last year we interviewed 44,000 youth to get to know them. There are some rebels, 52% boys, all age groups. They want to be different. There are the creatives, making things in Habbo differently. There are achievers, 56% boys, self confident and ambitious. There are loners, mostly over 17, who are coming in because they don’t have as many friends in the real world. Then there are the traditionals, mostly under 12 and girls. They’re conservative in the sense that they just like to do things the way they’ve always done them.

In Japan, there’s a massive loner segment. In the US, there’s a large achiever segment. And in Finland there’s a large traditional segment. So people act differently depenging on the country.

When we’re localizing, we try to take that into account. Going aborad has been tricky. One figure that popped up was that 70% of Habbos wanted a foreign friend. When we went to Japan, Finns invaded the hotel and wanted to know what they were like. But globally only 44% of teenagers had positive attitudes toward foreigners. So it was a total catastrophe. The Japanese closed their doors and wouldn’t let anyone in without a Japanese name.

So how do we describe Habbo. It’s a virtual world where people can hang out. It’s very casual in that you can hop in without an installation. But it’s also nto casual. There are people spending a lot more time here than the WoW players.

A lot of people call it Web 2.0 or Game 3.0, but I think that’s silly. We’ve been doing it for seven years. Long before those names came out. My personal favorite, from my dad, is “Gameless game.” It’s a game without gameplay. And it’s non-violent because of that. It makes me proud that we’ve been able to affect such a large group of people without any violence.

So what’s it look like? [slides of Habbo] You get the basic figures, and there are something like 8 billion different combinations you can make. We use isometric projection so you can click where you want to go. There are public areas operated by bots. Teenagers can’t get alcohol at the bar, but they can get bubble juice. And you can sit in the booths there, and only people sitting in the booths will hear what you say. We’re also putting in an instant messenger, There are a couple of games for people who are less into the open play part. Those are great for people who are just getting to know the world.

The real beef is the rooms. You can create a room, but then it’s empty. And there’s a lot of social pressure to create and fill it up. Then you go to the Habbo catalogue, and you can choose from different types of walls and colors and furniture. Some are just for vanity, but the vast majority, you can actually interact with. Here’s a winter-themed room. There’s no gameplay, but people were running around pretending they were snowboarding. [slide of crowded room[ This is actually a bit of an abuse of the way it was meant to work. If you put all the furniture on top of each other, it looks like it floats.

We introduced last year home pages for users. We had a Web platform, but no Web content, so now you can create a homepage. This is a regular user, and she has 30,000 comments in her page. And there is virtual furniture for this, but it’s purely vanity.

This is the latest craze for the dot com. Someone figured out that if you create a group and get a homepage, you get a little badge that’s a little green guy that follows you around. And so people have been creating all these little guys to follow them around and they’ve run out of colors.

We have a mixer that just sounds good out of the box. So we’ve changed the hotel from being something silent, as it was for the first six years, into someplace where the cool rooms have music. We were thinking about free uploads, but it wouldn’t be something the users had created. Now it’s all mixed by users. And you can burn the music to a virtual CD and trade that. WE’re trying to emulate the real world. They have to pay about $.50, but that’s about what they’re used to.

The real goal is the activities. It’s not the furniture, but what you do with other users. For example, Harry Potter is really big, obviously. I tried to join and play, but only Hufflepuff had a room at this Hogwarts, so I went to another that was very similar. And there I met a group based on being emo--only 18 members. Spider-pig is really popular right now.They actually figured out a way to do a spider-pig badge. This original is from Australia, but others have surfed to their community and then brought it home. And there are people that promote ecology, like this Be Green group. A lot of people express their real world values even though they’re anonymous. Here’s a McDOnalds. They’re roleplaying a minimum wage job. They’re getting paid maybe $.50 an hour to be there, but it’s fun. They’re inventing new burgers and emotion what they’re doing. This is a trax store. With music, a lot of people are wanting places where they can go and trade trax. Here’s a dragon room, and each of these is about $5, so someone spent about $200 to get this really fairly boring room.

There are lots of armies and mafias. I guess that says something about teenagers. They can’t do anything. They just stand there and emote how they’re punching people, I guess. But everyone else can just ignore them. It’s pure roleplaying. This is a moving furniture room, that, similar to the early days of UO, you could catch people and PK them if you could trap them. [more and more rooms]

Habbo is really about open play. It’s the players coming up with all the stuff that they’re doing. We try to stay away from defining the experience. There are no pre-defined feedback loops, at all. You don’t get rewarded from what you do unless the other users do.

So if you don’t get rewarded, why do it? Obviously there’s fame gathering. But there’s alos just a love to play. You can see that in children. Some teenagers probably have that, but they’re no longer in a place where that’s acceptable. So we’re providing a space where they can rediscover that.

What’s the difference between this and a traditional MMO? The players always know what’s hot before you do. And players can update that immediately. So they don’t have to wait for us to provide the spider-pig.

F(players)=f(content)
It’s always scaling. WE don’t have to put into more content just to provide something to do. The users are always providing it.

Which brings us to the rant of UGC. I hate the term. It makes users sound like bots or machines. IT’s also not as applicable for us. Maybe player created activity, or just stuff.

There are 1000s of fansites. And whenever we change the look of our site, the fan sites scramble to keep up. IT goes farther. There’s a couple that met five years ago in HAbbo and now they’re getting married. There are Habbo Raids on YouTube. I guess these are probably older users who used to think it was cool, but now are way too cool to come play. But they just want to get back inside. They’d probably be pissed about that. There’s a phenomenon around it, “Pool’s Closed.” You can get a tshirt. It’s spread to other virtual worlds. And most people probably don’t know what it means, but it looks cool. I might buy one.

Community management is important. The community managers in a lot of other worlds are the superstars in the world. WE’re trying to avoid that. Otherwise, the players lose the idea that they’re the most important.

Releasing new content is important. Players get bored if there’s not new content. So we’re releasing a new build pretty much every month. We switched to scrum last year, and it’s been hard, but also helpful.

When scrum is done wrong, it’s a bad waste of time for designers. If you’re switching, you need to support your team and really live with the team at all times.

We’re always working to figure out what the users want next. You have to listen to them. But as a designer, I’m always falling into the trap that I know better. It’s hard to resist the temptation, but you should. I’m almost always wrong.

There are 5 prerequisites to creating a playground.
1. You need to create something to play with. Legos are a great example. We had the pieces of furniture, and we made them so simple so people could just jump in.
2.You need intuitive interaction. Kill the UI. If the users are noticing it, it’s not working. Everything should be happening in real time, and you can’t assume that most people know how to free model or program. Most people can’t even draw.
3.Set up a mood for play. In the real world, it’s increasingly hard to play. For me as a 30-year-old guy, the only options are really when I’m playing with kids or the dog. Otherwise there’s a box around you that says you can’t just play. There are social games, like you can play hard to get at the bar, but there are so many rules around those games. You need to put a lot of energy into figuring out how to do that. People shouldn’t laugh when you do something stupid.
4. You should support user created goals. The Leather Ball in WoW was great. It completely changed the way people played the game. Instead of trying to get the most XP, the loser was the guy who wound up with the ball at the end. And there are all sorts of different activities for just passing the ball.
5. The last thing is the shared social setting. Even when people are authoring the content, you could still walk into the room and see it. If you want to play, you need to figure out how to play.
Bonus: Safety. You need to make the users feel comfy, even if it’s just a matter of letting them have privacy. Even the virtual character should feel safe. If you construct the game in the way that people can’t affect the content that other people offer, it makes it simpler.

So is it easy? No. The hotel is big enough that people understand you can have a room and still go out. You need a setting that communicates that. I’m also starting to sound like half the skills of the people in the games industry aren’t needed. Like you don’t need writers because the users are creating it all. But nothing could be further from the truth. You need to weave the stories in together. So it’s harder than the old style linear stories.