GDC 2006 Round Tables

Like the Economics session this one was standing room only. However, the sessions didn't really come to any far reaching conclusions. We did have one major conceptual breakthrough though. Rather than fighting the trend, designers need to embrace it and design games that put a premium on player skill rather than character skill. The key is that this skill does not have to be the typical reflex based skills as you find in most FPS games.

You can get a hardcopy of the report in Word format here.

MMO Player-to-Player Sales

Overview

Originally, MMO economies were self-contained mini-games. Players created, bought and sold items to each other’s characters within the confines of the game. The industry actively fought players who wanted to buy and sell their items to other players for real cash. This was done to preserve the integrity of the game. Nevertheless, player-to-player sales continued and grew into a minor industry. Recently many companies have decided to allow and even facilitate these transactions between players, getting a slice of the action for themselves in the process.

How does this business decision impact the design of future MMOs? If you are designing a game that will allow player-to-player sales, what type of infrastructure do you need to support these transactions? What changes in the design of game’s adventuring, crafting and advancement systems have to be made? Can hordes of cash farmers exist in the same game environment as “purists”?

The sessions were all well attended, with over 100 attendees, and standing room only in most cases. We had businessmen, academics and designers attend the session, many from Europe and Asia. So there was a wide variety of view points. Unfortunately we had no one who identified themselves as working for a broker like IGE. We did have one attendee who made her living creating items for sale in Second Life.

Originally the roundtable was going to discuss three potential game design strategies to deal with RMT. We assumed that our companies would take one of three strategies; rejection of RMT and active suppression, making RMT safe for the players and finally designing a game that fully embraces RMT.

During the first session (embrace RMT) it became obvious that using those three strategies to frame the discussion was not working. The design we came up with was totally unsatisfying and did not deal with the gameplay issues that RMT poses. A few post-session discussions lead the moderator to reframe the discussion to developing with a functional design that satisfies both the need to provide a fun experience and the business needs of the company looking to deal with a changing business model. We did this by analyzing two games that do not appear to suffer the design issues that RMT brings to games.

Background Information

Each session started with some basic background about the history and the current state of real money trades between players. We also discussed the reasons for and against RMT in general.

History of Real Money Trades

· 1989 – First known character purchase was with a Gem Stones II character in for $2,000.

· 1997 – UO launched. First items offered on Ebay within a few weeks. EA accepts RMT as measure of the success of the game (1999).

· 1999 – Everquest launched. First items offered on Ebay in a few weeks. SOE actively opposes RMT, banning accounts and asks Ebay to remove auctions.

· 2002 –Black Snow sues Mythic Entertainment for Mythic’s attempt to stop sales of Black Snow items. Suit never heard by court since Black Snow stops paying their lawyers.

· 2003 – Journalist earns $24,000 in one year buying and selling virtual items

· 2005 – Anshe Chung earns $150,000 a year operating a business in Second Life.

· 2005 - UO offers players option of buying higher level character for ~$30 (Advanced Character Service)

· 2005 - SOE offers service to players on two EQ II servers for player to player sales

· 2006 - Sony Online takes out full page ad praising PC Gamer’s refusal of accepting those ads.

The phenomena of players purchasing game items rather than “earning” them has been with the industry for almost as long as the industry has existed. Some companies accept it as a natural outgrowth of the success of the game, others actively fight it, and other use it to provide another revenue stream.

Current Status of Real Money Trades

In general Real Money Trades have been with the industry for almost as long as the industry has existed. Until recently it was a small part of the game, but in the last few years it has become a major force in the industry.

· Estimates of world-wide RMT trade runs between $880 million and $1 billion (though the low estimate is $100 million)

· Estimates of South Korea alone is $550 million (recent estimate from private firm survey)

· Estimates of 2,000 gold farming businesses in China employing 200,000 employees.

After we walked through the background materials we explored the various business models that companies where using, to commercialize this phenomena.

· Brokers – in the Broker model, players sell their items, in-game currency, and characters to an intermediary, such as IGE. These brokers consolidate all of these items and offer them for sale back to other players. The brokers make their money on the difference between their purchase price and sale price.

· Secure Exchange – In late 2005 Sony Online Entertainment provided a secure service for EQ II that was designed to allow players to buy and sell currency, items and characters between each other. With this model SOE is trying to replace the services that the brokers and eBay are providing. The advantage from the player perspective is that since this is a sanctioned service, the exchanges are safe, and they don’t have to worry about fraud. From the service perspective they get a cut of the sale price, and they reduce customer service complaints about fraudulent exchanges. Finally, because the service is limited to select servers, the players who engage in these types of sales are segregated from the players who don’t want player-to-player sales destroying the “integrity” of their game.

· Convertible Currencies – A few games now back their in-game currency with real world cash and allow players to create and purchase items from each other in-game. Players also can “cash-out” their in-game currencies for a real world currency, mostly U.S. dollars. Most notable of these type of games are Second Life, There and Project Entropia.

· Korean Model – the Korean business model is not a pure player-to-player model, but rather a player-to-company model, where the player can purchase items and in-game cash directly from the company running the game. Normally the company will not repurchase the items, though a Vietnamese MMO allows the players to use in-game cash to pay for playtime. Typically the game is “free” but the players can purchase better items, and currency using some sort of hard currency. Some of these games have two currencies, a hard currency purchased from the company and a soft currency earned in the game. In the U.S. the best examples are Puzzle Pirates, Haboo Hotel and Go Pets.

Why?

In all cases these business models exploit a desire by the players to either, purchase their success, start the game further along than the normal starting position or recoup their time investment in the game. So the next question the round table tackled is the question of why does this happen in some games and not others. We identified a number of design features are conducive to the emergence of player-to-player sales.

· Character vs. Player Skills – RMT only occurs in MMOs that have persistent characters, equipment and/or property. However, not all persistent MMOs have active RMT. For example Planetside has no active RMT even though there is character advancement and persistence. The reason that Planetside does not lend itself to RMT is because that success in the game is primarily the result of player skill and not character skill. The First Person Shooter system as well as the game’s PVP nature means that a player must depend on their own skill for success rather than the skills and attributes of their player character. Purchasing a Planetside character only gives a player access to a ranking on a leader board, a ranking that can only be defended by the skill of the player.

Game designers make this trade-off between character and player skill all the time. Certain games types are more likely to be character skill oriented, such as RPGs and PvE games. FPS and PvP games are more dependent on player skill for success.

· Time sinks - The primarily reason for players to engage in RMT is because players value their time differently. Most MMORPGs are designed so that player have to spend a lot of time in-game to improve or advance their characters. Some examples of time sinks are

o Geometric leveling – each level requires more experience points to gain than the previous one. As a result players have to spend more and more time in the game to gain the next level. The old “hell” levels of EQ are an extreme example of this system.

o Low-probability Spawns – to complete a high value quest the players have to wait for NPC that only rarely appears, or appears only once per some long time period. Stringing a number of these quests together, makes completion a time consuming function, with players camping spawns for hours on end.

o Low-probability Reward – A variation on low probability spawns are low probability rewards. In this case the “good” stuff drops from a defeated monster very very rarely. The same holds for rare quest advancement items. Again long hours of camping are the result.

o Transit Time – If it takes 20 minutes to get from town to the area where the player wants to play, or time to organize a group for a raid, then the player isn’t advancing their character.

· Level dependant content – A lot of level based or skill based games either directly or indirectly lock content from the players. Having a level or skill requirement on some area directly locks content. Putting the most desirable content in the most dangerous areas indirectly locks content. Has designers we tend to put the must interesting stuff around the enl of the game.

Many players want to get to the “end-game” and not do the “boring” advancement activities in order to get there. Coupled with significant time sinks, players have a high incentive to pay a premium to get to the end game which has the most interesting content.

The Problem with RMT

The main issue with RMT, from the designer’s perspective, is the impact that it has on player behavior. The sessions came up with 3 major problems.

· Unearned Success – players buy their success rather than earn it. This is more of a Western problem, but never-the-less, it causes resentment between the player’s who achieve their success through game play, and those who use RMT to jump over the “boring” parts of the game. (Bartles called this violating the Achievement Hierarchy.)

· Destroys the Game Integrity – The games are a shared fantasy, where players are entertaining each other in a fictional environment. RMT requires that hordes of pharmers break into this shared fantasy, to harvest gold and quality items. This brings the real world into the magic circle of the game, ruining it for those playing the game according to the intend of the rules.

· Incites Anti-Social Behaviors- This damage to the integrity of the game, as well as pharmers monopolizing the best camping spots, results in many players striking back at those that they perceive as pharmers. And in some cases that activity boarders on racism. The prime example are stories out of South Korea where some RPG players attack anyone who can not response to chat messages in grammatically correct Korean, assuming that those who can not are Chinese Gold pharmers. Similar stories circulate in the US and other counties.

Why is it good?

The question of the value of RMT also came up, and the various sessions came up with a few reasons of why RMT is good.

Lets players start where they want – most of these games have significant time investments, and many players want to play those elements that they enjoy, and do not want to have to advance through time consuming activities, before they get to the part that is fun. RMT gives this opportunity.

· Another incentive to play – MMO are time commitments. If the player feels that they can earn some money while playing, or recoup their time investment once they are finished, then they have another incentive to play. This means more players for the game.

· Avoid scarcity frustrations – some MMO items are restricted in numbers, causing scarcity, and frustrating players. RMT allows players to avoid this frustration, thus keeping them in the game.

· Most profitable model – this is a business reason, but if the company supports or provides RMT functions they can make more profit than if they just simply offer the game for a monthly subscription. The average revenue per player for games using the Korean model has been reported to be $20 to $25 per month.

Legal Environment

The final element discussed was the state of the legal landscape. Given that player-to-player sales result in items having a real world value, what are the tax, liability and other legal issues that a company has to face? A lot of these issues boils down whether virtual items are property and/or who owns this property.

The question of who “owns” a virtual item or if it has a real world value is one that has not been decided by the courts. The most companies use the End User License Agreement (EULA) to define the rights of the company and the player. However, restrictive EULAs have been challenged. In the US the first attempt was the Black Snow case. Unfortunately, the case did not make it to court. The only other case that actually got tried was in China. In the Chinese case a lower court reportedly agreed that virtual items had real world value, and that the company could not use the EULA to avoid compensating a player who items were stolen by a hacker. The compensation was the return of those items rather than cash, but the court used the value of the items on the secondary market to show that the items did have real world value.

Session 1 – Exploiting the Phenomena

After the background discussion we started to design an MMO. The participants were told that for the moment that they were promoted to management (which meant that they didn’t have a clue but had the vote). They were asked which game strategy to pursue. They had three options, first to design a game that defends the “magic circle” reducing or eliminating RMT. The second option was to try and mitigate the impact of RMT on gameplay. The final strategy was to fully monetize the phenomena.

Design Assumptions

The group decided to go for the money and design a game that fully monetized phenomena. The following assumptions where made.

· Business Model – Some form of the Korean model would be used, with the players being able to purchase items from the company directly. The assumption was that not all items could be purchased some had to be earned though game play. We would provide a safe and secure means of allowing players to engage in direct RMT. There would be no conversion back to real currency.

· Player-to-Player trading – the game would use some form of sanctioned auction house, where items could be sold between players in a secure manner. The company would take a slice of the transactions.

· Legal – The legal climate is one that causes the most concern for all games that utilize RMT. The question of “ownership” of the virtual items is one that has not been decided by the courts. The participants started to discuss the legal ramifications of supporting real money trade (liabilities, law suits over “loss wages” because of nerfs, or down time, etc.) and its impact on game design. However, since there is no case law to guide us the group decided to ignore the legal issues and assume that the design will conform to whatever laws come about.

Design Features

· Persistent World – the game would have persistent, both in characters and in housing and the like.

o Traditional RPG progressive experience curve, with monster bashing rather than questing as the main way of gaining experience.

· Consumable items – nothing in the game is permanent. Everything decays or is used up in some manner, either requiring expensive repair or out and out replacement.

· Robust End Game – Most of the content focuses on the end game. In this case the “coolest” adventuring areas would be for high-level characters. Requiring the characters to complete a complex quest would further restrict some of the best areas.

· Sell items, character and Power leveling – we plan to offer just about every type of “product” that the current brokers offer. In addition to items, we would offer complete characters, and power leveling services as well.

The session end before we could discuss how both a player-to-player RMT model and a Korean based model could co-exist. This requires a skillful division of items offered by the company and those offered by the players.

Many of the participants in the session expressed dismay at the design. In general it was one that commoditized “fun”. The magic circle of play was being violated and the game was not something that they wanted to play. It also used design features that the current generation of MMOs are trying to avoid, primarily time sinks. All-in-all the session did not come up with an enjoyable game.

Second and Third Day – Expanding the Magic Circle

There were a number of post-session conversations that tried to reconcile the need to preserve the magic circle of play with the economic forces that are forcing RMT onto these games.

The Magic Circle

The most interesting conversation that the moderator had was with Raph Koster. In a previous tutorial session Raph’s panel got into the question of defending the magic circle. One of the panelist said that with the increase in free time, the magic circle is expanding taking over more and more of life. In other worlds the magic circle doesn’t need defending, its advancing and taking over more of the world.

The problem for this session was that we were thinking that the magic circle, or the game as a game, needed defending. That mind set means that as designers we tend to mitigate, marginalize, ghettoize or just stop RMT. The design from the first day reflected that tendency, as we just threw up our hands and let the business model dictate the design.

That reversal of the conventional wisdom was then coupled with a comment from a session member. He had an example of a game that had RMT but did not exhibit any of the problems that we had identified, Eve Online. So the session decided to concentrate on discovering why Eve Online design allowed for RMT without its negative effects.

Eve has all of the standard elements of an MMO. Heavy time investment to gain items and resources, level based advancement (though the character does gain exp while off-line), and heavily character skill based for success in combat. But it has one key element that other games do not.

Real game success comes from the fact that the game is a massive corporate simulation, with players forming companies and fulfilling key roles in those companies. The successful players are those with superior interpersonal skills. They are the ones able to work together with many other players to organize and run large corporations.

Selling a high level character in Eve only gives you a leg up for a short while. If the player does not have the necessary personal skills to keep their character successful, then the character will quickly start to lose its wealth, and standing in the game. Thus all the player is really buying is a starting position.

The third day was spent doing a similar analysis of Second Life. Here the company is actively supporting RMT. They provide their players with the tools necessary for them to create items, and activities for other players to enjoy. Effectively they are the ultimate sandbox game providing various game primitives, for the players to use in emergent game play.

Here again player skills are paramount. But instead of the player having to have superior eye-hand coordination, it’s the player’s creative abilities that provide success. Musicians, artists, and even game designers can use Second Life to provide them with entertainment.

Like in Eve, purchasing a Second Life character only gives the player a leg up on the rest of the players. It is more like purchasing a pre-existing business. The new owner is not guaranteed success, they can only grow or sustain the business though their on skills.

Conclusion

RMT is here to stay, and will become a major driving force in creating new business models for MMOs. The amount of money that is being made out there is too large, and the cost of attempting to shut it down via the EULA is too high. As designers and businessmen we have recognize that fact.

The key is to reduce the negative effects of RMT, primarily the resentment it causes between those who “earn” their advancement in-game, and those who buy it using RMT. Eliminating time-sinks have been discussed in the past, but the sessions never really considered this, since the nature of most MMO is one where we want to provide the player with un-ending game play and advancement.

The real solution is to shift the character/player skill trade-off towards the player skill side. This does not mean that the game has to become an FPS. Eve and Second Life have shown that player skills other than eye-hand coordination can be used. If done correctly then all the player is purchasing is a higher starting position in the game. It is up to them to maintain and advance the character further, though the player's own inherent abilities.

References

A number of references where used when preparing for this round table. Some of the more informative are listed here.

Terra Nova - the blog to discuss the economic, social and legal implications of MMOs.

Synthetic Worlds - Edward's Castronova's book on the MMO's, a must read for any serious discussion on MMOs, economics, and sociology. I even only disagree with about four things in it.

	GDC 2006 Round Tables

09/13/06

Second and Third Day – Expanding the Magic Circle

Conclusion