Bill Gurley
General Partner, Benchmark Capital

"The Revolutionary Business of Multiplayer Gaming "

In 1996 I begged my way onto Wall Street. I was at CSFB and in the first couple of weeks, I discovered that a lot of people were willing to give out free advice. Now, most of that advice was related to where to get a good cocktail, and that I needed new ties, and that I should never wear those same shoes again. But Michael Mauboussin was kind enough to offer me mentoring advice about what it takes to be a good analyst. I am not sure he was authorized by the administration to do that, but I was happy to take the advice. It had a huge impact on my career.

Michael also introduced me to Bill Miller and his team. This is something else that is difficult to thank someone enough for. It is tremendously rewarding to interact with the Legg Mason organization. I have never encountered an organization so committed to learning. I jump at any chance I get to be near them because I always pick up something new.

Michael called me about four months ago and asked me to speak at his event. When I asked about the theme of the conference, he told me “changing economic landscapes”. I said, “What’s that?” He asked me, “What do you want to talk about?” So hopefully I have morphed this topic to fit the theme of this conference!

I should tell you one more thing before I start my talk: I am a venture capitalist. I invest in early companies, and I try to get three to five years ahead of the curve. Everything that I talk about will be heavily biased by where I sit. This particular topic has moved past the venture stage into a much broader stage. I find this topic fascinating personally, and I believe that it fits with a number of the themes that Michael talked about.

I am going to talk about online multi-player gaming. The acronym most commonly used in the industry is this beautiful one: MMORPG, or Massively Multiplayer Online Role Playing Game. You can see why they shortened it, but you still cannot pronounce the acronym! MMO is becoming a word that more people are using. This refers to the concept of a whole bunch of people entering a single online gaming environment over the Internet and interacting simultaneously. Some of these environments can include hundreds or even thousands of people interacting in a single place from their own computers. For various reasons, this is a lot more intriguing to people than playing a game by themselves on their computer.

To fit the theme, I’ve created a subtitle for my talk: There Are Only So Many Hours in a Day. My belief is that these interactive entertainment mechanisms, broadly speaking, are going to eat into the free time that people have and will therefore become a competitive threat to any company that is vying for that free time.

Interactive gaming began in 1977 and 1978 as MUD’s—Multi user Dungeons. They were text-based fantasy role-playing games, and they began as soon as computers started to get hooked together. One of these was called the Plato system.

From 1977 through the early 1990’s, there were multiple attempts to create online gaming experiences. As soon as modems were invented, someone created games like Zork. These were all text-based games. The computer would keep track of where people were, and you could ask the computer questions. You could also chat with people. None of these systems ever took off—they were more a dream than a reality.

There was a great book in the early 1990’s called Snow Crash by Neal Stephenson that talked about the “metaverse” that people could jack into and run around having adventures. This was a sci-fi thriller in which people had to run around through both the real world and a digital world to save the day. In this world, some people think Neal Stephenson is like a god. There were a lot of attempts in the first Internet bubble, but most of these things failed. Most analysts and investors therefore consider this to be a failed category. Of course, the success of these games is heavily dependent on things like broadband penetration and computing power.

Something finally happened in Seoul, Korea. This happened for several reasons. There is over 80% broadband penetration in Korea, and their DSL speeds are much higher. This environment was much more prone to allowing a game like this to take off.

The first company to make an economically successful game was NCSoft with their game called Lineage. Lineage has had over 4 million registered users over the life of the product. They have been moving outside of Korea and have had an impact on other markets.

One of the other reasons that Korea, and then China, has seen the rise of such games is the presence of Internet cafes. They are wildly popular in those two countries, even though you don’t hear a lot about them here in the U.S. On any given night, every single terminal will be filled in these places. They are empty warehouses with black paint on the walls and a bunch of terminals, but they are very important social gathering places.

About three or four years ago, the most interesting gaming market shifted from Seoul to Shanghai. A company was born called Shanda Interactive. Twelve months ago, Shanda had the largest market cap of any China-based company on the NASDAQ—larger than the portals, travel sites, and job sites. Shanda started later than NCSoft, but they recognized that distribution mattered. They focused on the 200,000+ Internet cafes in China. They put feet on the street and suffocated these cafes with marketing. They built systems into the cafes so that they could take cash, because there was no credit card infrastructure. If you go into an Internet café in China, you will very likely see Shanda paraphernalia all over the room. One of the things that you have to keep reminding yourself about China is that labor is so cheap that sometimes it’s better to hire 500 people rather than trying to automate something.

Shanda will generate over $265 million in revenues in 2005, and they have had over 700,000 users simultaneously logged onto their system. They have 47% percent net margins. China has always had problems with software piracy. Shanda is a software company whose value depends not just on the code, but on all of the other users of the software. The value proposition for a multi player game is dependent on the number of people playing the game. You could rip off the entire Shanda code base, and it wouldn’t mean anything. In fact, one of the leading portals partnered with NCSoft to try to launch Lineage in China. This was a game that was known to be popular and compelling, and they got nowhere. This is an increasing-returns phenomenon. This is a great business. It doesn’t matter if someone rips off the code, the labor is incredibly cheap, and they have 47% margins. You don’t see anything like this in the U.S.

These Internet cafés are wildly popular in China. I went into several of them at two in the afternoon, and every room was 80-90% full. These games are so popular that China has just passed a law that if you are under a certain age, say 25, you can only play the games for three hours at a time. In fact, the government is requiring the software companies to deduct points if you play longer than three hours. It will be interesting to see how they enforce that!

Most of the games to date have been fantasy-oriented. People will go into these worlds, like World of Warcraft, to accomplish tasks in teams. Some of these tasks require 50 or 100 people to succeed, so you may have to join with a larger group. There is a lot of teamwork and strategic thinking required to be successful in these games. A real person sitting at a computer somewhere in the world controls every character on the screen.

Shanda went public in the summer of 2004, and became the top performing stock on the NASDAQ. People recognized that this was a pretty cool thing.

Meanwhile, in the U.S., some companies were having some small successes. Electronic Arts introduced Ultima Online in 1997 and attracted about 200,000-300,000 users. Sony offers a product called Everquest that some people call “Evercrack” because of its habitual nature. Everquest was successful—about 500,000 users—but still only appealed to a niche. It was a great product for Sony. It cost $30 million to build and generated $80-90 million in annual business. They expected over $500 million in profits over eight years.

City of Heroes was developed in the U.S. by Cryptic Studios and distributed by NCSoft in 2004. This product started very strong—they quickly attracted 300,000 users paying $15 per month. They were knocked down by the same product that hurt Shanda, which I will talk about in a moment.

At this point in time, however, most people believed that this was an Asian phenomenon. They believed that there were systematic, cultural reasons why these types of games would be successful in Asia, but not in the U.S. The theory that I have heard the most (which I do not believe personally) is that in many Asian societies, people grow up in very uniform apartment buildings, living in a room with three or four other people, watching three television channels instead of hundreds. They do not have very many alternatives for escapism. Because of this, the theory goes, this type of fantasy game would be much more interesting for them. I actually don’t believe that this is an Asian phenomenon. Then, something happened that proved it wasn’t.

Let’s go back to California. In Irvine, California, there is a captive studio of Vivendi called Blizzard Entertainment. Blizzard had put out a series of previous-generation games (not multiplayer) called Warcraft. These were fairly successful products. The team had been working for three years on a multi-player game called World of Warcraft, which they launched in November, 2004. To me, World of Warcraft is the shot heard “round the world” for multiplayer gaming. This game blew away all previous perceptions about this market space.

World of Warcraft launched in late November. By March it had 1.5 million users. The previous record in the U.S. had been 500,000 users over 4-5 years. This is like having a Hollywood blockbuster generate four times more revenue than any other movie ever made. This put to rest the notion that multiplayer gaming is an Asian phenomenon.

I don’t know exactly why World of Warcraft worked while some of the other products did not. I wish I knew the answer! I have heard some theories, but none of them quite answer it. I’ve even been “in world”, but I still have not figured it out.

This game is pretty expensive to play. The CDs cost $50 and the monthly subscription is an additional $15 per month. It is more expensive than any of the console games that came before it. Next, World of Warcraft went international. It went to China—one of the reasons that Shanda’s stock is down is that analysts believe that World of Warcraft is a competitive threat to Shanda’s Chinese market. Everyone thought this was an Asian phenomenon, and suddenly a U.S.-developed game is having an impact on the Chinese market! They now have over four million users worldwide—in less than a year! I estimate their revenues at over $30 million per month, with a first-year revenue of $350-$450 million. This is an asset that does not simply disappear overnight. In fact, sequels are quite popular in the gaming industry so I wouldn’t be surprised if World of Warcraft becomes a multi-billion-dollar product over time. This would compete with the best-grossing films of all time.

Coca-Cola is aware of this type of game and the type of impact that it is having in China. They created an advertisement simultaneously with World of Warcraft for the Chinese market. In the ad, three young women are playing World of Warcraft, and an older man tells them that “sexy is what sells”. They tell him no, and in fury, he morphs into an orc from Warcraft. The women then morph into game characters as well, and they fight as a team to defeat the orc. As a prize, they open a chest full of icy Coca-Colas. As they drink, they morph back into their “real” selves, and look over at the defeated man, who is now wearing the outfit that the orc had been wearing in the game, looking ridiculous. We are not seeing ads like this in the U.S. market. China is a culture that is more accepting of this kind of thing, but I would get ready to start seeing ads like this in the U.S.

You still may be thinking to yourself that this is interesting, but not for you. You may never play a fantasy game and join a tribe and slay dragons. But there is something else going on as well. These multiplayer games are becoming more casual and more social. Casual games on the Internet are becoming very popular among a much broader demographic than fantasy games. One group that spends the most time in these casual online games like Tetris or on crosswords is 30-40 year old housewives. We are starting to now see an intersection of multiplayer games with those kinds of casual games.

The leading Instant Messenger company in China is called TenCent. They have about 100 million active monthly users. They have started to build “avatars”—digital representations of yourself—into their IM platform. Believe it or not, you can now buy digital clothes for your avatar. Now I happen to think that that is the best business that I’ve ever heard of in my life! There is infinite supply with no inventory. They play off of scarcity—they will offer only 200 of an item that they will price higher. I had a chance to meet the CEO of TenCent, and he told me about one of their ideas, which is pure genius. He has designed deterioration into the clothes. If you have had clothes for your avatar for over six months, they start to develop holes. This is just brilliant!

Some of the American companies have been watching this in Asia and are starting to take notice. We are starting to see some IM companies building avatar representations into their platforms. I will show you a couple of things going on in Asia and Europe that are ahead of the U.S. today.

One product that is in the U.S. today is called Neopets. If you have a child ages 5-10, there is a good chance they’ve been to a website called Neopets. I am not sure that Neopets would market itself as a virtual world, but it has all of the right components. There is a currency, there is teamwork, and there is education—kids can learn how to code HTML inside of Neopets. Then they have their own web page. It is not a very pretty site, unfortunately. I don’t think that the company has maximized its value the way they should have. They recently sold to Viacom for (I think) around $170 million. They have 25 million children as members. They did not really monetize that very well because they tried to make all of their revenue through advertising. Viacom has now added subscription pricing.

We are investors in a product called Habbo Hotel. In this game, you customize your avatar and hang out in the lobby of this hotel or by the pool. You chat with your friends, and then you get a room in the hotel. You outfit your room with digital goods and try to be popular and hold parties. (I’m telling you, this is a great business model!) Most of these sites work on a model where you enter for free, and then you buy “in-world” coins or credits for real-world cash, and you use the credits to buy digital goods. This company is doing extremely well—they have been doubling every year. They did over $15 million in revenues in 2004. They are moving international, and they are launching a partnership with Disney.

There is a very similar product in Korea call CyWorld. “Cy” means “relationship” in Korean. They have tried to blend Instant Messaging, social networking (like Friendster) and the Habbo Hotel concept into one game. This looks very similar to Habbo Hotel. They even have a high-tech control room for CyWorld, and you need a control room because 25% of Koreans have a CyWorld account, representing over 12 million people. They get over 4 billion page views per day. It is insanely popular in Korea! They generated $54 million in revenue in 2004.

Because you were laughing at some of these ideas, I thought you might enjoy this audio clip. This is an interview between the VP of Business Development for CyWorld and an American journalist about selling digital goods.

Journalist: What is the value of a pixel? Try about $200,000 - $250,000 a day. I cannot believe what I’m staring at on this computer screen. It’s a phenomenon in Korea. It is called CyWorld. “Cy” actually means “relationship” in Korean. We are speaking with the head of business development for this company, Rick Kim. This is going to be so difficult to explain to a North American culture. Your parent company is SK Communications?

Rick Kim: SK Telecom bought this product.

Journalist: This is a virtual world or an identity world where people actually pay money to decorate their own homepage. They pay money for these little trinkets.

Rick Kim: First they have to buy, or receive as gifts, what we call “acorns”. This is the “e-money” currency of this digital world. You cannot actually buy anything with your actual cash. We do not accept actual cash. We will entice you to buy bags of acorns. With these acorns, you can buy gifts for your friends. You can buy wallpapers and other items for them to use to decorate their homepages. When my friends receive a gift, they will generally send a gift back to me as well—this is the gift-giving culture in Korea.

Journalist: Not to make it sound bad, but you are essentially minting your own money, aren’t you? But you’re being kind, too. People want to buy these items, and when they don’t want them any longer, they can sell them back to you for… what is it? Seventy percent of their original value? This depreciation happens, of course, because this virtual object has been used. Maybe it’s just me, but how did you pull that off?!?

[Audience laughter]

Rick Kim: It was what the users wanted! It is what they asked for.

Journalist: I think more American companies need to move to Korea! That’s unreal!

Rick Kim: When people buy these digital items, sometimes they get sick of them. They don’t want to necessarily hang onto wallpaper with a picture of falling leaves when it is now July. They asked us if they could exchange items for something new or if they could sell items back to us. We thought it was a really fun concept to buy back “used” digital items.

Journalist: Fun concept or scam?

Rick Kim: Actually, they like it!

Journalist [laughing]: I understand that physical items sustain damage and scratching. That’s fine. But these pixels are in the exact same condition they were when you first sold them! And I think it’s smart to make money on the up-sell. But how did you convince…? What do you drink? What’s going on over there that we’re missing in the U.S.?

Rick Kim: Of course, I think that users would prefer that we give them one hundred percent of the value of their items back, but come on! It’s been used!

Journalist: I can’t believe this. You have got the dream. Lots of people have been chasing this, but you have already figured it out! I am actually going to be very curious to see how it will play out when you enter the American market. [Switch to a later part of the interview]

Rick Kim: That is my job, by the way. I am here to talk with other companies to see if they are willing to grab hands with us. Some have been very open, and some have been basically laughing in my face.

Journalist: If you pitched this to anyone in the U.S., they would laugh you out of the office!

Rick Kim: In our version of San Francisco, you see the half-inch versions of Mini-Me dancing in the street, and there is a cable car. We have included many of the famous sights in the city, and the American flag, of course…

Journalist: You are missing the puddle of urine that the city of San Francisco is known for. Is that icon available? And if so, then at what cost?

Rick Kim: It is available actually, and there is an icon for a guy puking in the corner.

Journalist: So let me get this straight. Not only are you selling nothing, you’re selling nothing by-product!

Rick Kim: We are also selling a guy puking in the corner. We are selling a buddy who is patting his back. And sure, we can sell you puddles of urine, if you want. We’ll sell it to you.

Journalist: I am not speaking on behalf of all Americans here, but you guys are nuts! But hey, man, more power to you if you can pull this thing off! I think we’ve come full circle—I think you’re nuts, and you use acorns! I’m borderline. [End of Interview]

So CyWorld is taking off and going gang-busters. They are highly profitable. This is something you should get ready for. You will see a lot more of this kind of thing in the U.S. in the next two or three years, including CyWorld bringing their own product.

If you think that’s crazy, just wait. What if I told you that there was a very popular game in Korea where you go online, customize your own go-cart, and race it against other people? KartRider is a product that is about five times the size of CyWorld right now. This is the most popular thing going on in Korea right now. There are even professional KartRider racers who play eight hours a day. They win prizes and make money. Coke sponsored a $50,000 prize at a recent KartRider event. The racers are sponsored—so their carts have decals from leading brands that are advertising. As if that weren’t enough, there are TV channels that cover KartRider races in Korea. This is a huge phenomenon.

KartRider uses the same business model. You buy the in-world coins, and you spruce up your cart to make it look cool and go faster. About 25% of Koreans have had at least one KartRider race. They estimate $250 million in revenue this year for this product, and it is obviously extremely profitable.

The American companies are not standing still. I mentioned that Vivendi has entered this space with World of Warcraft and that Viacom has bought Neopets. Disney has launched its own interactive game called ToonTown. It does not have the same types of numbers as the Korean products, but these companies are paying attention. Disney also partnered with a Korean company to launch Disney’s Virtual Magic Kingdom—this product looks similar to the Habbo Hotel product.

Now I will talk about a company that is near and dear to my heart because I invested in it and I am on the board. That is my warning to you that I am biased to this company. Second Life is a company based in San Francisco that has grown from 15,000 to 50,000 users over the last year. The key difference to Second Life is that the founders of this company decided that they would not build out the virtual environment for their users to interact with. Instead, they developed the tools and allowed the community to build out their own environment.

As a group familiar with the concept of increasing returns, you will understand why this is so exciting for me. None of the actual physical things that you will find in this game were built by any employee of Second Life. We built the toolset, and people are running businesses inside of Second Life. Second Life’s revenues come from two key sources: land sales and land tax. We charge tax to people to build environments inside this world. We are a platform, if you will, for this kind of environment. Inside the world, we have developers and consumers. Over 2,000 of our users are netting a profit in their Second Life businesses after the fees that they pay to us. There are over 1,000 users making over $100 per year. There are about 500 users making enough to pay their rent. We estimate about 40 users are working full-time inside of Second Life as their primary job.

How do these people get cash out of the game? There is a site called Gaming Open Market, where you can trade Linden dollars for U.S. dollars. There is a full book, just like you would see on Wall Street, and they match trades. About $500,000 per month back and forth between Linden dollars and U.S. dollars. That is how our developers make money. They also make money by getting people to enjoy the content that they built in-world.

The complexity is very compelling to me. One gentleman has built a motorcycle shop in-world. He uses the toolset to develop a motorcycle that he will sell to someone in-world. We allow digital ownership rights, so we assign the ownership to the developer, not to us. They can set DRM on their code—they can block their code so that no one can copy it. This allows them to create unique items that they can then sell. Some people build out areas that look and feel very space-age. There is a whole community that is consistent with the Star Trek theme. Other areas feel much more like the Old West. The detail on the objects that people have been able to build is truly astounding. Some people have built environments that feel much more like the real world.

There is a full physics engine in the world that allows developers to create airplanes, boats, jet skis and other objects. Fishing became wildly popular about three months ago. This was totally emergent—we didn’t plan any of it. Someone figured out how to make fishing work. Then rare species of fish emerged, and then a taxidermist popped up. There were suddenly tackle shops with different types of fishing rods and lures that you could buy. It is awesome to watch this emergent complexity – we do nothing and all kinds of things pop up from the interactions among the users.

A game popped up in-world called Tringo—it is a combination of Tetris and Bingo. It is a game that groups of people sit around and play. It immediately became addictive inside the game—users were complaining on their blogs that they were spending too much time playing Tringo and not enough time developing. A professional developer investor came into the game, found the person who invented Tringo, licensed it, and last week the game launched as a console game. This is a piece of original IP that was developed in-world.

There is a lot of socializing going on in the game. Many of the environments are very social, like bars and night clubs. Dancing is pretty popular. Games are also emerging within the game. I told the CEO in passing one day that I couldn’t wait until someone builds 11-on-11 soccer inside of Second Life, where every player is independently controlled. When I was getting ready for this speech, I was looking on the Second Life forums for photos and I discovered this photo of a Second Life soccer match. I called the CEO and asked why he hadn’t called me to tell me there was soccer inside of Second Life. He told me that he hadn’t known it had happened! 15,000 acres are being built up every week inside of Second Life. There are over 20,000 acres already developed. There is so much going on that the CEO cannot even keep track of what is being developed by the people in-world.

There is a lot of retail business inside of Second Life. People design clothes for their avatars, as well as lots of other products. We are starting to see themes emerge. One new theme is a whole section of the world devoted to pirates.

We even have land barons. Our biggest land baroness is a Chinese woman who lives in Germany. She buys huge sections of land from us. In fact, she just built a huge German world. She buys the raw land from us, builds a German theme on it, subdivides it, and resells it back into the community. We love this because she is improving the overall product. She spends as much as $5,000 or $7,000 per month with us, and we think she is making about $150,000 per year working inside of Second Life. We are still early yet— those numbers should go way up from here.

You can go to Flickr.com and search for Second Life. You will see thousands of pictures of all of the different things going on inside of Second Life. One of the interesting features of Second Life is that the users create all of the content. Even more interesting, some of the users sent us an advertisement for Second Life without our even asking for it! It is of such high quality that it is better than anything we could have done on our own.

This is one of the most exciting companies that I have ever been able to work with, just because of all of the emergent stuff going on inside the game. We are really just trying to hold on because it is not in our control at this point.

As a venture capitalist, I like these businesses for a number of reasons. These are subscription-oriented businesses with recurring revenue streams. There are competitive “moats” – once users are playing in your “world”, they are very unlikely to switch. The Second Life product is especially interesting because the developers have spent a lot of time and effort to learn the toolset. Now that they are making money, there are huge switching costs for them to move to another platform.

Increasing returns and network effects take place in these businesses—the more users there are of a game, the more fun and valuable the game is for everyone. There is “real” competition—competing against another human is ten times more enjoyable for consumers than competing against the computer. These environments appeal to our need to socialize. These are habitual games that users become very passionate about—I know people who have had to take World of Warcraft away from their kids because their grades have been hurt.

The top 10% of users of Second Life spend over 40 hours per week in-world. It is not uncommon for people to spend 20 hours a week in these games. Over the lifetime of a console game, a user plays for an average of 50 hours. Playing these MMORPGs, users spend 50 hours in two or three weeks. People are very passionate about these games.

In Korea, there have been real-world retaliations. In-world, a horde will beat up another horde, and in the real world, they will track down their adversaries and beat them up for real. Korean police have been brought in for in-world theft users and they have actually prosecuted people for in-world theft. A man recently died in Korea after playing for 50 hours to win a competition. People are very passionate.

And people are actually making money, and not just in KartRider. In World of Warcraft or Everquest, you can spend many hours building up a character, and then you can sell it on eBay. You can make a living playing games. There is unlimited complexity in these games any time you are interacting with other humans; there is an unlimited universe of possibilities. This space is also very appealing because, as of last year at least, it was not very well understood. I yearn to be contrarian in the VC market. I have put down some bets in this space when it was not well-known, but I believe that it would be more difficult to do so today because more people are paying attention. It is a high risk, high reward proposition at this point, but that’s the kind of business that I’m in. The payoffs are big.

How does this compare to other forms of entertainment? A World of Warcraft user will pay $50 for the CDs and $15 per month. In their first year of use, the WOW user will pay $230. That is not as much as cable TV ($600) and it’s more than someone will pay for HBO ($120?). I try to compare this class of business to other media types because they are going to compete for people’s share of free time and share of wallet. People are spending a lot of time engaged in these games.

Many KartRider players spend 2-3 hours per day playing. Over 8% of Second Life users spend over 40 hours per week in-world. Let’s think about household penetration: a greater percentage of Koreans have played KartRider than U.S. citizens watched the last episode of Friends. The first-year revenues of World of Warcraft ($400 million) are very impressive compared to other media. This is close to the net revenue for EA’s Madden game (after the license fees paid to the NFL), the top game in their portfolio. Grand Theft Auto made about $500 million. It would rank as the 25 th best movie of all time, if you compared it to Hollywood releases.

Remember, though, that this game will have a very long tail—this is only its revenues for the first year! These games are also starting to reach broader demographics, which will make them more competitive to other forms of entertainment—35% of the users of KartRider, Second Life and CyWorld are women. The global market capitalization of this industry is approaching $10 billion.

We are talking about changing economic landscapes. As we think about companies that compete for people’s entertainment time (ABC, Vivendi, Yahoo!, Disney, Fox, etc.), they will have to pay attention to this emerging business and make bets about whether or not they will participate. Vivendi is sitting in a very strong position with World of Warcraft, and I imagine that they will do more based on the success of that product. Viacom bought Neopets and they’re exploring other options as well. Disney is obviously paying attention. Yahoo! is doing a lot of things in this space and is a leader in the casual gaming space in the U.S. You are going to see more and more activity in this space from these types of companies going forward.

Question: Is there gambling in Second Life?

Gurley: There is some gambling in Second Life. We tell our users not to, but they do it anyway. In any of these environments, you will see that kind of thing going on despite the prohibitions of the companies. In some of the games targeted at younger audiences, they spend a ton of time and effort blocking inappropriate behavior. If you are younger than 12, you need to send a fax from your parents before you can get full access rights to Neopets. You have to be 18 years old to sign up for Second Life.

Question: What kind of data do you have about the demographics of the different users in different countries?

Gurley: I don’t have any data with me, but that data is available. You can probably find a Shanda report online. I would not be surprised if this is still a mostly male phenomenon, but the percentage of women is growing in the fantasy genre.

Question: You mentioned the Chinese law that limits the number of consecutive hours that a young person can play these games. You have to wonder what is going to happen to the Chinese youth if they spend all of their time playing these games.

Gurley: You need to read the book that Michael mentioned—Why Everything Bad Is Good for You!

Question: Will the Chinese government clamp down on the amount of time users can spend playing these games?

Gurley: The article that I read about these moratoriums came out in the last few weeks. This is still being played out. There are a number of cultural controls that the Chinese government is using in regards to information dissemination and the ownership of media companies. It would seem that these efforts will limit their long-term success, and they seem counter to what the rest of the world is doing. So, yes, it appears they will continue going in that direction, but global pressure might cause them to back off a little bit. It is also very difficult to implement controls like the Chinese have legislated – how do you keep track of whether or not players are getting points docked for playing too long? It seems kind of crazy to me, but they’re going to try.

Question: It is like the kings in the fifteenth or sixteenth centuries trying to stop all of the farmers from playing golf – it was consuming too much of their time.

Gurley: The other issue is that users can hack around just about any restrictions that you put in place.

Question: Are there a certain number of users in these games who are spectators and not active participants?

Gurley: Yes, but it is different for each company and each business model. In KartRider, where they have TV shows, then obviously there are a lot of spectators. In most of the environments where it is free to enter but you have to then pay for things, people are trying to attract people to watch their events – it makes them feel important. There is quite a bit of lurking in these games because people are being courted to go visit a party in someone’s room. I don’t know if that’s equivalent to watching, but that is the kind of dynamic that’s going on.

In Second Life, we are seeing more and more spectators. People will hold events or performances – we are seeing more and more of this. People are holding an “open mic night” at a bar in Second Life. They are actually singing in their house, but their avatar is on a stage in a bar. The room will fill up with people who go to watch. Our CEO goes to in-world chats at a venue like a stadium. Avatars will file in and talk with him about what is going on.

Question: Other than by buying Viacom or Vivendi, how might Legg Mason Capital Managment participate in this kind of thing? How might other companies like IBM or Proctor & Gamble utilize some of these concepts?

Gurley: I do believe that Legg Mason owns some Netease, which is a company that’s had some resurgence around their multiplayer gaming product. The funny thing about stocks is that you have to understand what assumptions are baked into the price of the stock. I am not in that business, so I will plead ignorance. I have not studied Vivendi, so I do not know whether World of Warcraft is big enough to move the needle for them or not.

IBM is well aware of this, and they are chasing a number of multiplayer gaming companies to have them build on top of IBM’s hardware and architecture. I know that because they have called on a number of the companies that I am involved with. You might be surprised to learn that Wells Fargo has built several physical properties inside of Second Life of their own volition. They thought this would be a good place to attract and educate new customers. I don’t know how many people are going to them, but they are offering finance education courses in-world. Coke, obviously, is well aware of this phenomenon—they did the ad in China and sponsored KartRider in Korea. We are starting to see more and more name brand companies participating in these games in various ways.

Question: What is Yahoo!’s perspective on these games?

Sue Decker: Bill is diving much more deeply into this topic than I will, but we concur with most of the themes that Bill has talked about, and we have looked at some of the properties that have come on the market, like Neopets and others. We started in the casual gaming market, which requires less investment up front and offers less of a blockbuster payoff in the end. We use more of an advertising business model than a subscription-based model. We have done pretty well. We are rebuilding our gaming platform now, and we agree that this is an area that will get more and more attention and time spent in the U.S. Part of the reason for the delay in the U.S. is that broadband developed more slowly here than in Korea and Japan. The development costs of the multiplayer games compared to the potential market was not that attractive for us. We believe that these worlds are moving together – you will see casual gaming and multiplayer gaming converging. Especially as self-publishing increases, the distinctions between casual and multiplayer gaming will become less and less defined. We are moving into that area. The most important theme for us is the role of community in the success of these games. We want to build communities around every one of our sites, whether it is sharing music with friends or gaming. We believe that communities will provide a very high barrier to entry.

We are all very familiar with web searches, but in Korea now you can do a “knowledge search”. You type in a question and hundreds or thousands or millions of people will come up with an answer to your query. This is far less capital-intensive than using web-crawlers, and the answers are often better. This is a very sticky service—the more users there are, the better the answers you will get. We are quickly launching this all around the world as well. There are a lot of concepts that Bill talked about that are very applicable to all of our verticals.

Bill Gurley: The investment required for fantasy games is moving towards the level of investment required for major Hollywood movies. Everquest cost $30 million to develop. I do not know exactly what it cost to develop World of Warcraft, but it was probably a similar number. Companies have to make a very big bet on these games, and there are companies out there doing it. Someone is working on a Dungeons & Dragons game, and there is a Lord of the Rings in development. These games might be $30 million zeros—if they don’t tick up quickly, then they will be dead coming out of the gate.

Question: EA estimated that it cost $100 million to develop World of Warcraft, not including promotion and advertising. You need to be very confident that these products are going to work. As a venture capitalist, what are the key qualities that you think these games have to have in order to be successful?

Gurley: As a venture capitalist, I don’t want to bet that kind of money on a single product, so I look for something that can be leveraged. Thirty minutes into my meeting with Second Life, I was trying to close the deal because of the platform nature of their product. We did not have to be the ones to invest in the development of the environment – the users did it for us! Without the portfolio benefits of a large media company, I do not think that we as venture capitalists would make that kind of investment in a single game. I looked at a company called Turbine in Boston that is the kind of company you are talking about – $60 million of venture capital to develop two titles. I cannot play that game, but a media company can.

Question: Is one of the keys for you that the community develops the environment?

Gurley: It is a key for me as a venture capitalist because it is more capital efficient. I think everyone in Silicon Valley is thinking along those lines. Wikipedia has everyone excited about user-generated content. I think you will see more people focusing on how to get the community to develop the actual content.

No one has asked, but I will say it anyway. Shockingly, it feels like we are moving into a bubble-like atmosphere again in Silicon Valley – a mere six years after everyone got their head handed to them. The disadvantage of this market place is that the prices of these opportunities are increasing, but the liquidity events are not. This is the worst of both worlds from my perspective. There is quite a bit of frothy investment going on on Sand Hill Road right now.

Question: How sticky are users?

Gurley: I have a firm belief that the end users investing in the product—to develop friends or to develop a property that they are building—are the stickiest kind of users you can have. I have wondered whether there is a difference of increasing returns – what is the incremental value of the Nth user. In a game of hearts or spades, for example, once you have 400-500 people in the room, the experience is pretty much the same whether you add or remove a couple of people. If you have a game like Second Life where 100,000 people can play at the same time, then I believe you are more subject to increasing returns. This could add an even higher level of stickiness.

You have probably heard a lot about a game called Party Poker. You might say that this is a game with a low coefficient of increasing returns because of the spades example. What PartyPoker.com did differently was to overlay tournaments that start every five minutes. They took a game that might have been less scalable or subject to increasing returns, and made it far more scalable by adding tournaments. If you want to play a $25 tournament before you go to bed at night, you are most likely to get that experience at PartyPoker.com.

Question: Right now these games have to be played at home or in a café. What will mobility mean for these games?

Gurley: I could talk for an hour and a half on that subject! As I mentioned earlier, broadband rates have a direct impact on the use of multiplayer gaming. This is a direct correlation. Even in the mobile world if you graph multiplayer gaming against per-bit pricing, you see the same correlation. Companies that sell unlimited bit packs see far more multiplayer gaming than companies that have bit limits. There are also huge latency and twitch problems with the current cellular network that make a lot of these games impossible to play today. I also happen to sit on the board of Jamdat, the leading maker of games for the mobile market. They have multiplayer games, and about 10-15% of the usage today is multiplayer, but this is heavily impacted by latency, extra subscriber charges, etc.

There is a whole other theme going on that I can only touch on briefly. Wi-fi offers much broader speeds and services. Playstation Portable has Wi-fi built into it. Something could happen in this direction that could cause a dislocation and dramatically accelerate this development. Currently, though, you are left to innovate around the complexities of the cellular network, which is not optimal. The kinds of games you end up with are turn-based games. Lots of people are working on this.

Question: Are there health risks or is there a risk of lawsuits related to health risks associated with playing these games for so many hours?

Gurley: I try to make neither moral nor legal judgments when I am on stage! I am highly confident in this litigious society that we have built in America that you will definitely have a lawsuit related to this. I have no idea whether or not it will be warranted. The book Why Everything Bad Is Good for You was written with this in mind. I think that there are obviously things that can happen in these environments that are bad for people, especially young children. I know of kids that were A students who missed their grades after World of Warcraft was introduced. Moral judgments aside, there is no way that this trend can be put back in the box. We will have to find a way to be better parents and to institute new laws where necessary to deal with it. But this trend will not be stopped for health reasons.

Question: This is obviously a growing industry, but it is still very young. There are going to be a few winners, but also a lot of losers. What factors should we be looking at to pick the winners? The budget? The research department? The ideas themselves?

Gurley: If things pick up in the investment environment, then we will see a lot of failures. There are evolutionary parallels that we can draw. The “primordial soup” has been energized with excess capital. We will see a lot more innovation simultaneously with a lot more death. That is my perspective. Of course, I am the one who has to swallow it – this is not the kind of investing that most of you need to worry about! I have noticed in the venture environment that sometimes we get so excited about a sector that we burn the whole playing field. The competitors start doing things that are unnatural, and then nobody wins. In the end, large companies can sift through the ashes picking up nuggets of value for almost nothing.

The easiest thing to look for is an opportunity with early momentum in which increasing returns will play a role, and pay too much for the asset relative to what someone else will pay. I will pay 10X for a deal with momentum relative to a “series A” deal in this environment. The opportunity to get momentum and increasing returns is so rare today. Who would have guessed that KartRider would have been so successful? But the trick is to find an early KartRider and do everything you can to invest in it.

The comments, opinions and any forward predictions presented about any particular security, the economy and "the market" are based on the analysis of the speaker. These are not necessarily the opinion of, and should not be construed as a recommendation on the part of Legg Mason Capital Managment or any of its affiliates.