[Roblox founder David Baszucki tells our own Christian Nutt how the kids' online game naturally grew from observing play, and how it's earned millions of users and a Disney deal as a virtual unknown.]

Unlike most highly commercialized free-to-play kids' virtual worlds, Roblox started as an outgrowth of technology designed to simulate physics. It's a pure physics-based play space; kids arrange the blocks into LEGO-like structures, and others can access these spaces as they wish. Rather than a virtual world, it's a collection of user-generated spaces: in terms of how the site is set up, it's almost like a YouTube of play.

When its creators put it in front of kids as part of an educational package, they quickly noticed how much fun the kids were having with it, and moved to develop it into a product with that audience in mind. Now, Roblox has launched and found an organically growing audience, finally reaching the point where its first promotional deal, with Disney, has gotten off the ground.

We recently sat down with David Baszucki, founder and CEO of Roblox to find out more about the project.

How long ago did you launch this? I guess it was more of like a soft launch.

David Baszucki: Completely soft launch, so no publicity. It was about three years ago when we turned it on. It was very crude and primitive. We really haven't made any hubbub or fanfare about the product since then.

It's primarily aimed at kids. You have a lot of kids playing this game, and you've gained enough of an audience so that things have basically started snowball a little bit for you.

DB: I'd say there's more than enough user-created content to fuel the game in kind of an interesting way. I would also say the focus has to be continuously on trying to acquire the majority of your customers virally as opposed through paying just for scalability, and we're doing that.

How did you come up with this idea?

DB: My prior company was called Knowledge Revolution. We made physics simulators for simulating things like cameras, car suspensions, and torpedo mechanisms, and one of the constraints with traditional simulators has always been, you know, you can only do so much.

So, we kind of got excited about pushing in an architecture where we can start pushing two, five, 10 thousand parts in our worlds all running through the physics simulator. You know, one of the inspirations was can we make a simulator that can handle enough parts to do that kind of stuff.

And it just seemed like a natural application of that would be building just a fun world to mess around in?

DB: My company had really two main product lines, educational simulators and professional simulators. The educational simulators, there's a product called Interactive Physics, and we had lots of kids trying to build fun stuff with that.

If you look back at that product, which was released back in the early '90s, it looks very much like the new wave of Flash-based physics simulators, all the [Crayon Physics] kind of physics simulator products that are out there today.

Now, when you say you've been gaining users "virally", do you do anything to engender the viral acquisition, or are you relying on the fact that the kids like the game?

DB: We're relying on the kids like the game. In the 12-year-old group, there's a lot less of the social fabric that you'd find in a Facebook game in an 18-year-old group. So, for us, it has to be more tell your kid to get on Roblox.com at the schoolyard.

Now, do you rely entirely on user-generated content?

DB: Yes, we do, 100 percent. So, every place in our system is user-generated. When we initially launched, we had some places that we made ourselves, but very quickly the user-generated places became more interesting than our own places.

Is there a rating system? How does the good stuff get to the top? Is it just based on the number of people playing it? Or is there a voting system?

DB: We're experimenting with all kinds of sorts, so, favorites, popularity, and even more complex metrics. So, how you sort the stuff is interesting and important. Different users want, I think, different sorts.

So over the past three years, you've attracted users. Do you find that the players you have high engagement with what you've put on offer so far?

DB: We think they do. So, we think they stick around for a long time. We have a lot of kids visit us every week from all around the country and sometimes all around the world. So, there are a lot of kids who have been playing for a long time.

Recently you had a sponsorship deal with Disney. You know, we've seen this with things like Habbo Hotel, but I'd never really heard of your game until now. Can you talk about how that came about?

DB: Yeah. So, it's the first deal we've done of this type. It was for the Zeke and Luther TV show. In the latest season, they were trying to boost the membership of this show. We worked with them to create user items, parts to assemble user-created skateboards.

We had kind of an automated content system for users to rate each others' skateboard parts, and we generated real heavy engagement both for kids building stuff as well as playing in skateboard parks with a lot of Zeke and Luther branding.

Simultaneously, when playing in these skateboard parks, there was, you know, little blurbs telling them when the Zeke and Luther show was going to be on TV. So, it's a little interesting, but I'd say the key point is the stats on this latest season of the Zeke and Luther show we're significantly better than it was last year.

Your game is content, and that's marketing, and they're kind of meeting in new ways.

DB: I'm bullish on user-generated content. Our job is more like a toolsmith. What kind of tools can we give kids to build interesting stuff?

Looking at your competition out there, is it more virtual worlds where kids can hang out? Or is it other games where kids can do stuff? Is the whole thing kind of nebulous to you?

DB: So, I have a big picture view on competition, and that is we are trying to provide the safe, fun, educational experience to, you know, predominantly boys, predominantly younger boys. And so our competition is very broad and granular. It's any online kind of activity, I think, that you'd find a 12-year-old boy doing.

So, for those that are on Facebook, for those that are on Google Chat, for those that are on Xbox Live, for those that are playing with another game, I really think our competition is wide-ranging and far-flung. I view our market a little more like the movie market or the TV market. There's a lot of things going on out there, and our challenge is more increasing the overall quality of our experience relative to the market as a whole. That's what I would say.

You've said the website looks a little rough, but you're working on improving things. Is that your primary goal?

David Baszucki: I think, you know, there's a lot of things that we're doing, but just quality, raw quality and raw performance, are two big things that we're working on right now.

When you say "educational", do you think it's educational just sort of by its nature as a building system that uses real-world physics, or do you actually guide an educational aspect to it?

DB: So, we have kids who are learning in many different ways on our site. Some are learning by designing and building. Some are learning to Lua script to create more powerful objects in our system. Some kids are learning how to be currency traders in our system because they literally are arbitrating our currency, buying and selling things. Some kids are learning more artistic things. So, I think we have a lot of education going on in a lot of different disciplines.

How much of the creativity do you open up to your audience? You mentioned Lua scripting, which is maybe more than I expected given the age of your audience.

DB: Yeah. I think we're always surprised at what kids can do. And you can't build, I think, a product where mass market, everybody is going to be writing Lua script, but you can build a product where a very small portion of the users can write interesting Lua script if that can be shared in the right way with other players.

And how easy is it for kids to share what they've created for your game?

DB: It's pretty easy now and getting easier. Everything's online, everything is sortable, so it's reasonably easy to share things.

You guys have some stiff competition coming when LEGO Universe launches because it's got that powerful branding behind it. Do you have your eye on that? Or is that anymore important to you than any other competitor?

DB: I think there are many high-quality competitors. I think that's just the tip of the iceberg. There are four or five or six or eight other very high quality user-created experiences coming down the pipe as well, and you know, it's an interesting area, and you need generally high quality to compete against all of that.

I don't mean high production quality. I mean high performance, high gameplay quality, high scalability, high intuitiveness -- that kind of stuff. I think, in general, we've gone pretty far with a more simpler production look. I think it's been working for us.