Virtual worlds are governed not only by the source code used to develop the world, but also by civil code documents that establish a governance structure that constrains the interactions of users of the virtual world and regulates relationships among stakeholders of the virtual world. While previous research has examined specific aspects of these documents, this paper analyzes these governing documents as a totality. By examining the totality of and the interplay among the governing documents of a number of established social worlds, this paper seeks to discover insights that can prove valuable both for scholarly understanding of social world governance and for the various stakeholders of social worlds. Following this analysis, the paper offers a set of policy recommendations and considerations to facilitate the development of governing documents that more democratically and equally serve the needs and rights of all stakeholders in virtual worlds. The paper concludes that virtual worlds and their governing documents are boundary objects with agency, in that they are the result of interactions among stakeholder groups and in turn reshape the relationships among those stakeholder groups.
Welcome to the Obfuscatocracy
The framework of virtual world governance
Types of governing documents in virtual worlds
Stakeholder groups and their rights
Analysis of virtual world governance
Selection and analysis
Virtual worlds and governing documents as boundary objects with agency
Conclusion: Beyond the Obfuscatocracy
Obfuscatocracy, n. — a system that is governed by way of obfuscated code.
An ogre, two dark mages and a horde of undead elves walk into a tavern. As they sit down, they begin to bellow out their drink orders. You decide to go over and strike up a conversation with them. As you exchange stories of your many battles, the tavern owner interrupts you. He asks the group for their payment of three rubies per ale. All of a sudden there’s a spell cast. *Poof!* Next thing you know, that horde of undead elves and their friends have just dined and dashed, sticking you with a hundredruby tab. You immediately call out to those scoundrels, but by now they are long gone. The tavern owner demands that you pay the tab. Seething with frustration, you start to wonder ... Where is the law in this land? Where is the order? The justice?
Well, don’t you worry, level 9 barbarian, rest assured that there is order in that virtual world. Illustrated in that story is the understanding that whether you are in the physical world or the virtual world, people look for the core fundamentals of government, law, and order. In the virtual world, government comes from two sources, the ingame rules created by the game designers (source code), and the contractual agreements written by the game designers legal team (civil code). In both cases, the code is often shrouded from users either by 1s and 0s or by complex technical and legal terminology. While obfuscated code is typically meant to refer to the difficulty of understanding the source code (Obfuscated code, 2008), the documents used to govern the use of the virtual world can be equally obtuse. As such, due to the obfuscated nature of these governing documents, this system of government can be referred to as an obfuscatocracy.
The obfuscatocracy created by the various governing documents of virtual worlds raises several questions for stakeholders of virtual worlds. Given their importance to the development and use of virtual worlds, these governing documents merit serious attention. This paper will use a stakeholder analysis to better understand how these governing documents affect stakeholders in virtual worlds. First, we will introduce complex issues surrounding governance in virtual worlds. After an introductory exploration of these issues, we will then survey six different virtual worlds to analyze how governing documents used in these six virtual worlds operate and are interconnected, and how the three major stakeholder groups — users, designers, and managers are affected by them.
Before we begin, certain key terms need to be defined. The term virtual world will be used to specifically refer to massively multiplayer online games (MMOGs) or better defined as any largescale persistent computerbased simulated environments where users inhabit and interact via avatars. This paper focuses on two major types of virtual worlds: massively multiplayer online role playing games (MMORPGs) such as World of Warcraft, and massively multiplayer online social games (MMOSGs) such as Second Life. The main difference between MMORPGs and MMOSGs is that MMOSGs focus on socialization, while MMORPGs focus on objectivebased game play.
As noted above, there exist two mechanisms for governmental control in virtual worlds — source code and civil code. This conceptual duality of code in an information technology context emanates from the work of Reidenberg (1998) and Lessig (1999). Both types of code are necessary and important in supporting a virtual world system, therefore each will be addressed in this section. However, the core focus of this paper is on the civil code aspect and its impact on virtual world governance.
The first mechanism for governance is through the use of source code, or the actual software itself. In the physical world, we are bound by certain laws of physics and nature. These types of laws cannot be changed or challenged; they simply exist. But in the virtual world, such laws of physics and nature are malleable and debatable. How fast can a person run? Does gravity exist? Is water wet? Questions like these are all programmable choices that designers can make. They are written into the source code of the application, and, in essence, comprise the very fabric of that virtual world. If the designers wish that no user ever gets sick and dies, they don’t have to call a quorum or declare an edict. They merely write that law into the source code, and it is automatically enforced within the virtual world. No one can violate this rule of law, no more than we can violate the laws of physics in the physical world (Lessig, 1999). Designers and programmers are so powerful in virtual worlds that some have drawn parallels toward viewing them as deities, if only for the fact that they could always simply pull the plug on the entire virtual world (Bartle, 2006; Malaby, 2006).
While this form of theocratic governance via a Newtonian God is quite powerful, it is by no means omnipotent. It does have certain limitations, because not everything in a virtual world is completely programmable. Some issues of the game cannot be generalized into simple coded statements, or rather that the answers to these issues are simply too time consuming to incorporate into the code.
In addition to this, while the programmers may control the physics of the world, they cannot always control the people inhabiting these worlds. Through avatars, users have some degree of choice and free will, which is not completely determined by the source code. This allowance is not an oversight on the part of the designers but rather a necessary comprise of the medium. Virtual worlds are an interactive and participatory media. If no choice existed, then the value of virtual worlds would be severely diminished if not entirely lost.
Beyond individual control, users can have much greater control and power in the aggregate. Communally, users can even go so far as to redefine the purpose and scope of a virtual world through the invisible hand of collective intelligence. For example, Second Life has proven particularly vulnerable to commodification  through the efforts of users to use it as a platform to make money rather than a space for pure social interaction or aesthetic achievement.
The other mechanism for governance is achieved through the use of civil code. In virtual worlds, civil code is determined by an elaborate mosaic of legal documents and policies. These individual governing documents contain all of the written codified laws for a virtual world. The corpus of all of the governing documents for each particular virtual world creates the contractual framework for governance of that virtual world. These documents are absolutely essential when looking at legal mechanism as they effectively purport to replace the entire panoply of common law obligations that regularly bind communities. 
There are a variety of different types of governing documents used in virtual worlds: software license agreements, terms of service or use agreements, privacy policies, and community standards and practices. All governing documents generally can be grouped into one of these four major types. It should be noted, however, that while all of the various documents can generally be assigned into these four types, not every virtual world must have a representative document from each type.
The attention in the literature has been to focus mainly on End User License Agreements and occasionally on Terms of Service Agreements. However these documents, while of key importance, are just individual components of much larger frameworks. It is only by broadening the scope and including other equally relevant documents that we can get a truer representation and a better understanding of virtual world governance.
Software license agreements
Software license agreements determine the conditions under which a user may use the virtual world. Since commercial computer software is licensed, not owned by the consumer, this document functions as a contractual agreement between managers and users. It determines what the user is permitted to do with that software. While this document is technically an agreement, the software user has little ability to negotiate the terms of this document, as agreement with these terms is a requirement for installation and use of the software. The software user is typically made aware of this contractual agreement during the installation of the software. These documents are often referred to as end user license agreements (EULAs).
Terms of service agreements
Terms of service (ToS) agreements dictate how the user may use the service component of a virtual world. The service component of virtual worlds is the subscription to the servers that host the online persistent virtual world that is used when playing the game. Since the game can only be played online, users are required to agree to the ToS if they wish to participate. As the software and the service component are inextricably tied together, often EULA and ToS are incorporated together as a single document.
Community standards are the most nebulous and eclectic type. It consists of documents that are, in essence, codified community standards and practices. The main purpose of these documents is to explain to the user their social roles and responsibilities as participants in these diverse communities. Since virtual worlds are shared by many users, these codes of conduct help to enforce the unpredictable and nonprogrammable human elements of the virtual world. It should be noted that often the policies included here could not be enacted through legally binding EULAs. These community standards address highly contextualized problems. As opposed to the strict and finite legalese found in EULAs, ToSs, and privacy policies, community standards are often written in an elusive and haphazard manner.
A simple example of a highly contextual issue in virtual worlds might be the speech of a user. Depending on the theme of the game, users should have the ability to speak freely and openly to other users. Users, however, are generally frowned upon for using speech in a malicious way. A user should not be allowed to verbally molest another user. The problem is, how does one create an automatic prohibition against such behavior in terms of civil code or source code? How do you define precisely what malicious speech is? Such decisions cannot be made automatically by the system but can be addressed broadly in community standards.
Three major stakeholder groups are involved in the design, management, and use of virtual worlds: users, designers, and managers. Stakeholder analysis is an essential aspect of evaluating the intents and impacts of an information policy in the electronic environment (McClure and Jaeger, in press) and can also be used to explore how values shape the design and use of information technologies (Fleischmann and Wallace, 2005). Each stakeholder group has different overarching goals and motives. These conflicting and sometimes overlapping interests play themselves out in the governing documents. Interestingly enough, even though governing documents are written by designers and managers, they do not and cannot completely ignore the interests of the users. The reason for this is that all the stakeholders share the same common goal: they all desire to have a successful and thriving virtual community. All other interests are secondary to this overall common goal. In addition, it is important to note that individuals may simultaneously belong to multiple stakeholder groups: for example, designers of virtual worlds may also simultaneously manage or use virtual worlds; as such, they may simultaneously be designers as well as managers and/or users.
Figure 1 below represents the various and sometimes conflicting interests of the three main stakeholder groups discussed above as the governance structures. The issues that are important to each group are all accounted for in the governing documents, but the overall balance in any particular virtual world may end up favoring the interests of one stakeholder group over the others.
Figure 1: Diagram of stakeholder interests.
Users are those who actively participate in the virtual world. In both the physical world and the virtual world, users have certain expectations about their rights and freedoms. What these rights and freedoms are in the virtual world is a subject of much debate. Scholars have argued that users have a right to play  or a freedom to play.  The major highlights of this concept can be summarized and abstracted from Koster’s (2006) Declaration of the Rights of Avatars. Users have the right to speak and express themselves freely, the ability to seek redress of grievances and to secure their property, and protection from various forms of harm to both their digital and their physical selves. All of the aforementioned rights must, however, be taken with respect to the context of the game. For example, if the game restricts a users communication because of a spell that has been cast through the normal course of the game, this is not a violation of their freedom to speak and express themselves. If a user is restricted from negatively speaking out about a recently added feature of the game, then that would be a violation of her or his right to speak.
When participating in virtual worlds, users seek the freedom to play. If users are being maliciously obstructed from participating in the normal experience of the world, or if they experience emotional harm, then they will become dissatisfied with the world and they will exit it. If a user runs the risk of physical harm to her self or property in the physical world, then she may not participate out of fear. Risks and concerns like these must be addressed or alleviated to ensure that users will be willing and able to participate.
In terms of governance, each individual user has very little power to change the terms of the governing documents. Since there is no voting and no arbitration in this government, the primary option that the user has is to take direct action, which is a form of political activism. Direct action in the virtual world can take many forms. Nonviolent methods for a user to take direct action would include temporarily logging off, refusing to participate in online activities, staging virtual demonstrations or sitins, etc. More aggressive methods, which are often frowned upon by the general community and deemed illegal in many of the governing documents, are sabotage, vandalism, and griefing (intentionally disrupting the gaming experience of other users) (Foo and Koivisto, 2004).
In addition to taking individual direct action, users can also seek to create broader community responses. By working collectively, users have a much stronger ability to influence producers to make alterations. Functioning as a collective group in the form of a team or guild allows users the ability to exert more formidable pressure than they could ever accomplish individually.
Finally, some virtual worlds allow individual users to seek redress through informal channels. This approach allows a user the ability to make requests, alert problems, or issue challenges. There is, however, no guarantee that any request will be acted upon. Ultimately, more progressive models are needed for understanding and integrating their [users’] work in virtual worlds .
Game designers use their creativity to create virtual worlds. Designers look at a virtual world much like a painter looks at a canvas. They envision the world that users might wish to participate in. Designers, like users, have rights. As discussed by Balkin (2004), they have the right to design, which can be thought of as a permutation of the freedom of speech and expression. The virtual world largely shaped by the expression of the game designer. This expression is a continual process, as the creation of a virtual world is not a singular onetime event. The interest of the designer is to exert and maintain a modest form of creative control and artistic freedom over their virtual world creation.
To designers, the concept of the virtual world is not merely the software code or the graphics but also the experience of users. Virtual worlds are an interactive medium. The actions of the users and the community breathe life into them. Therefore, designers do not merely design the world for themselves but for the community of users. The happiness and enjoyment of the users is an important component of the designers’ thought processes. Since the designing never stops, designers listen and are often very receptive to feedback from users. The shared goal of a prosperous virtual world results in a mutual respect and interaction. As Balkin (2004) explains, the designers freedom to design and the users freedom to play are often synergistic. 
The rights of managers are also a significant aspect of virtual world governance. Since the managers’ interest in the virtual world community is typically financial in nature, they have economic rights to profit from the virtual worlds that they manage. Since the majority of profits from most virtual world communities are received through user subscriptions rather than the initial purchase, managers attempt to prolong the subscription service. Therefore, they seek to maximize profits by optimizing the virtual world, provide the best possible experience for the largest possible number of users, and thus increase their profits.
The role of the marketplace in virtual worlds is crucial as it gives users their only leverage in negotiating the typically onesided and unidirectional nature of governing documents. The interest of users and managers partially overlap in this sense. Managers attempt to increase users’ enjoyment by giving them what they want and by protecting their experience. In this regard, managers have financial incentives to make some sacrifices in an attempt to ensure that users receive the best experience possible.
To determine how different virtual worlds employ governing documents, an analysis of selected prominent virtual worlds was conducted. This analysis aims to provide a better understanding of how the governing documents used in these specific virtual worlds interact with each other and, in turn, impact stakeholders.
The specific virtual worlds to be studied were selected based on several factors, including: size of user population, general popularity, accessibility of governing documents and uniqueness (Woodcock, 2007). To present a balance of both roleplaying games and social games, effort was used to maintain a representation of both MMORPGs and MMOSGs. The selection of virtual worlds surveyed are as follows: World of Warcraft, Dark Age of Camelot, Lineage II, EverQuest II, Second Life, and There. Table 1 presents the links to the source documents consulted in this study.
After the virtual worlds were selected, all of the possible visible and relevant governing documents were acquired from their respective company Web sites. If another relevant document was hyperlinked or referenced in a governing document, it was also acquired. Once all the documents were acquired (as listed in Table 1), each document was then separated into the four main governing document categories, based on name and general overall similarity of content. Analysis of core documents is an essential part of any information policy discussion (McClure and Jaeger, in press). The results were then tabulated in Table 2 into four categories based on the document content: software license e.g., EULA, terms of service agreements, privacy policies, and community standards.
As Table 2 demonstrates, none of the virtual worlds examined had all four types of documents available to users but every world had a least either two or three documents under an assortment of titles. Each document was then reviewed line by line and compared for unique patterns between the text of the document and between the documents themselves. This paper builds on the insights of Jankowich (2006), who analyzed the governing documents for specific regulations about virtual environments, such as costs, prohibitions, account regulations, privacy, liability, and account regulations. While such information is incredibly valuable for comparing virtual worlds, there are other important issues in the governance that merit analysis.
In this analysis, virtual worlds were analyzed and compared for patterns across each world’s governing documents, searching for broad themes of governance and the impacts on the relationships between stakeholders rather than specific issues in the documents. After being reviewed in terms of the specific regulations, each document was then compared to the other governing documents for that virtual world to see how the documents functioned as a totality. The factors for analysis focused on comparing and contrasting documents based on:
- Crossreferencing and interconnectedness;
- Parallel terminology and grammatical structure;
- Overall form and function;
- Uniqueness; and,
- Selfcontainment (i.e., could the document function by itself regardless of other documents?).
The primary goal of this analysis was to understand the implications of these patterns for stakeholders through an analysis of the interplay between the documents and how they function as a contiguous unit.
Several common characteristics were found to exist independently of each individual governing document. These characteristics all relate to the overall conceptual framework. These factors have an impact on the overall integration and interpretation of these governing documents.
Inconsistency in framework
When comparing documents across different virtual worlds, it quickly becomes obvious that there is no universally accepted naming scheme or standardized document set for contractual frameworks. In essence, each virtual world creates its own individualized set of documents. This means that documents with similar content might be referred to by different names within different virtual worlds. Moreover, certain documents may exist for certain virtual worlds but not others, or may exist but may be encapsulated inside another document. This lack of uniformity makes direct comparison of different virtual world documents complicated and creates many potential problems for members of multiple virtual worlds.
While each document in the dataset has been categorized into the document types in Table 2, the boundaries between them can be relatively fuzzy. Governing documents do not function in a vacuum and are dependent on and related to each other. Beyond direct references in documents, issues are often amalgamated together across documents. For example, a document might outlaw a certain type of user behavior in one document, but then explain what that specific type of behavior is in another. This means that even though documents are physically independent, they are logically related and coexist in the discussion.
One of the major themes in governing documents is the amount of complex legal language that is employed, particularly in the EULAs and ToSs. In general, only a small percentage of the population would consider themselves fluent in the intricacies of the law. Therefore, the technical writing style of these documents creates a language barrier that constitutes a cognitive burden for participants. In addition, there is also a considerable amount of technical and virtual worldspecific terminology that needs to be learned before the user can adequately understand the full implications of all of the documents. For example, if a document points out that machinima or griefing are not allowed, these issues would mean little to someone that did not possess previous virtual world experience. Since users in particular have varying degrees of knowledge and experience with virtual worlds in general and with specific virtual worlds, it is important that these documents are written so that all stakeholders can understand them.
Accessibility is a crucial issue, when looking at contractual frameworks. Accessibility speaks to where the governing documents are located and how users can access them. In most cases, governing documents are located on companies’ Web sites. Within the company sites, the governing documents can be dispersed in several locations; sometimes they might be all located together under the sites support page, but sometimes they might be spread across various parts of the site. The issue of accessibility of license agreements in the software industry has been addressed by courts in the U.S. In recent cases, the courts have challenged documents that were egregiously designed to impede users ability to analyze or understand them (Pollstar v. Gigmania Ltd., 2000).
Another factor, similar to the issue of accessibility, is the dynamism of governing documents. Governing documents are not static entities. They are susceptible to frequent modifications and alterations by the producers of the virtual world. Governing documents are rarely presented with a change log, or a last modified date. This issue is further complicated by the lack of an official repository or archive of previous governing documents that users could use to track changes in the governance of the virtual world. As a result, users must frequently revisit the governing documents in order to be certain to know the current regulations in effect within the virtual world.
Influences and requirements
Though virtual worlds can be radically different from the physical world, they are not completely separate and disconnected from our physical existence. Human consciousness, ideologies, beliefs, and activities in the physical world transfer over with the users to the virtual world and vice versa. These virtual worlds can thus have very real consequences. The U.S. judicial and legislative branches are beginning to catch up with recent developments in virtual world technology. People are beginning to be held accountable for what happens in virtual worlds. The U.S. government is finally starting to realize the importance that virtual worlds can play, either for financial reasons (taxation of virtual world economies; see Lederman, 2007) or for safety and security reasons (monitoring for potential terrorist activity; see Singel, 2008).
In addition to the virtual world spilling over into the physical world, there are other Internetrelated laws that are required to be respected and enforced in virtual worlds. Examples include intellectual property rights in the form of the Digital Millennium Copyright Act (DMCA) and privacy with the Safe Harbor Principles and COPPA.
Governing documents of virtual worlds tend to be created by designers and/or managers. Development of governing documents grants those stakeholder groups with disproportionate power in the governance of virtual worlds. As Jankowich (2006) explains, by creating these documents, designers and managers of virtual worlds instill themselves with broad and incredible amounts of discretion in deciding issues . Since designers and managers control the code, decisions and rules are created in a unidirectional manner from the topdown. There is no ability for individual users to directly modify the rules in the documents themselves. This topdown nature of virtual world governance is typical of many forms of governance, and where possible, it is desirable to institute bottomup structures for standardssetting and policy creation in addition to or in place of topdown controls (Fleischmann, 2007a).
Users must vocalize their concerns through customer service or message board forums. Equally disturbing is that users who attempt to voice their concerns are often censored and silenced. Freedom of speech is not guaranteed in virtual worlds. Virtual worlds function as privately owned spaces, similar to corporate controls for private spaces, malls, and company towns in the physical world (Jenkins, 2004). The lack of freedom of speech reinforces the unidirectional nature of governing documents. If users cannot voice their existing concerns, their concerns cannot be incorporated into future changes. Governing documents unfortunately do not tend to be participatory in practice, and do not function well as constitutions or social contracts because social contracts are dependent on the participation of a variety of individuals.
There are several critical problems with the governing documents of virtual worlds. One pressing problem is the astounding number of users that do not read or concern themselves with these documents before playing (Grossklags and Good, 2007; Magid, 2007). Certainly, it is possible that this is related to the design of the documents themselves, given the issues noted above such as accessibility, dynamism, and inconsistency. The disconnect between users and these documents is a serious problem that warrants much further consideration and study. The priority of the community should be toward informing the user of the issues that concern them. A better informed user benefits the entire community. This disconnect could be attributed at least in part to one or more of the factors analyzed above.
The overall problem is that the current contractual framework model places an enormous cognitive load on users. This burden results in users not being able to remain informed of important issues that concern them or to receive updates when important changes are made to governing documents. This disconnect also threatens to undermine the work of the producers, who stand to lose from the disconnect from the governance structure experienced by users. This problem is exacerbated by the complexity of managing the scope and interlinking of governing documents in relation to one or another inside the contractual framework. Therefore, the bulk of recommendations address the problem of design and delivery of governing documents as a contractual framework.
The recommendations are broken down into three sets. The first set of recommendations is a list of methods and practices that the designers and managers of virtual worlds can implement to improve comprehensive delivery and cogitative reception of governing documents by users. The second set of recommendations is a list of actions that users can take to play a more active role in understanding and creating governing documents. The third recommendation set includes opportunities for possible solutions by nonstakeholders.
Recommendations for designers and managers
The designers and managers of virtual worlds can take several steps to improve the creation, organization, and delivery of governing documents. The first possible solution is for all designers and managers to collectively agree to create a universally accepted standardized set of governing document types and terminology. This would mean that all virtual worlds would share the same model, language, and naming scheme. They would not, however, be required to share the same specific internal rules, just a standardized system for separating them into governing documents. This is already done in effect to a very small degree with the privacy policies. This standard consistency would be a good first step toward tackling issues identified in this paper.
A related recommendation is based on the organizational structure of the governing documents. One method for improving the access to and delivery of documents to users would be to organize all of the documents in a single standardized place within each virtual world. By collocating all documents into a single, consistent place across worlds, the effort required for users to navigate and find relevant information would be greatly reduced.
A third recommendation is that virtual world producers create an extensive archival repository of all governing documents. All changes to the documents would be recorded and stored in this repository. The changes that occur between the modification dates should be highlighted and annotated to alert users to recent changes. This would allow for not only better understanding of changes but promote greater responsibility and transparency regarding document changes.
A final recommendation for managers is the increased inclusion of users in the design and development of governing documents. How else better to craft something intended for users than to incorporate the intended users through participatory design (Schuler and Namioka, 1993). An interesting example is Erik Bethke, CEO of GoPets, who is currently attempting this approach of inclusive design in governing documents for the GoPets virtual world community (Bethke, 2007), which has since been expanded into the BetterEULA project (Bethke, 2008). Much like the slow paradigm shift in the computer science field toward software and Web development that focuses on usercentered design and participatory design, the crafting of governing documents might equally benefit from this approach. By focusing on embedding users’ values in the initial and continual crafting of documents, designers and managers could make great strides toward a mutually agreeable form of governance.
Often users are invited to participate in the process of evaluating and testing changes in the code of a virtual world before these changes are made worldwide. This betatesting process allows users, designers, and managers to respond to the often dramatic changes in the overall game play and community structure. An example of a highly dramatic change in a virtual world is the act of nerfing, which occurs when a character type has its power modified and reduced because its previous powers created an imbalance in the game. Only so much can be initially anticipated in the design process, so designers often go back and nerf characters after watching the actual game play to better create the balance they initially wanted. Nerfing is a continuous issue, as it will make existing users less powerful than they already are. Changes in game dynamics like this are tested before being applied worldwide. Users have an opportunity to experience and provide feedback on these changes, while designers can see the impacts and make more corrections if necessary. This method of beta testing virtual world software could easily be applied to the crafting and maintaining of governing documents.
Recommendations for users
There are several possible methods that users can use to address this system of contractual framework. The typical — and often the most cliché response is that if they do not like the system or have problems understanding the system, they should simply refuse to participate. This overly used recommendation is a simplistic and inefficient response to the core problem. The invested time and effort of users, as well as the social relationships that have been developed, have intrinsic value to the users.
The first recommendation for users is they should seek to initiate a process where governing documents can be collocated, reviewed, and annotated collectively by all users. Community analysis would greatly help reduce the cognitive burden by spreading the effort of analysis equally across the community. Instead of each participant trying to determine how each statement of the document affects them, several experienced participants could abstract and highlight for other users what concerns need to be addressed. This recommendation is very appropriate as it plays off the strengths of the already existing social nature of participating virtual worlds. Therefore, using a community analysis system would build on the already existing strong tendency of collaboration and community. This community analysis could be achieved by using current social software such as wikis. This model has the additional benefit that, like most social software, the community can quickly adapt and respond when changes are made in the governing documents.
A further recommendation is that users look to form groups similar to trade unions. While it has been suggested that users would benefit from engaging in participatory design of the virtual worlds (Taylor, 2006), members could also lobby for participation in construction of the governance materials. These pseudounion groups could be organized based on current game divisions such as guilds or out of sheer necessity. While it is difficult for one user to force changes in governing documents, the revolt or boycott of a large number of users would hurt the success of the virtual world as well as the profits of the managers. The managers might then negotiate out of concern for their profits. These groups could provide users with the ability to do collective bargaining. Through a method of collective bargaining, contested issues in governing documents could be resolved to be more mutually beneficial.
Leaders of the virtual world community have already imitated this process by proposing declarations for avatar/user rights (Koster, 2006) and policy declarations (SWI, 2007). While such declarations are purely aspirational, they have generated dialog in virtual world communities, regarding the appropriate levels of powers and responsibility users should possess in the governing process.
Another recommendation for users is that if they are dissatisfied with existing virtual worlds and their governing documents, they may work individually or collectively to build new virtual worlds that better reflect and accommodate the perspectives of users, and in turn craft their own governing documents for these virtual worlds. Research has shown that users who are dissatisfied with existing software products have the potential to develop their own software which can better meet their needs, which can then in turn develop into a commercial opportunity for these usersturneddesigners. The role hybridization reflected by usersturneddesigners can reshape power relationships among stakeholders in more radical and meaningful ways than traditional forms of participatory design, in which users are brought into the design process as partners, but do not have complete control over the design process (Fleischmann, 2006a).
Opportunities for thirdparty developers
While the previous sets of recommendations are geared toward designers, managers, and users, there are also opportunities for interested third parties to develop tools to analyze governing documents for virtual worlds. Thirdparty software developers can create a product that can assist users in analyzing and abstracting key points of interest in governing documents. EULAlyzer, which was created by Javacool Software (2007), attempts to meet this need. This product searches a software license for specific keywords and phrases. Areas of concern are highlighted for users, warning that they deserve careful scrutiny.
While this application is successful in its purpose and intent, the EULAlzyer is only designed to broadly look at software license agreements, specifically EULAs. Its use is hampered when analyzing virtual world contractual frameworks because of the complex interconnectivity among governing documents that exist. Therefore, an opportunity exists for a software developer to create a program that is tailored to analyze all of governing documents of a virtual world. This proposed software could scan each document of the contractual framework highlighting keywords and phrases. Beyond that, conceptual grouping and plain English translations could be provided to assist users and researchers. This software could be tweaked to focus specifically on issues of relevance to virtual worlds, such as the prevalence not only of legalese but also of technical terminology and phrases specific to the Internet and particular virtual worlds.
One problem facing access and understanding is how these governing documents evolve over time. Are governing documents proactive or reactive to changes in the community or changes in the overall communities in virtual worlds? It is quite clear that virtual world governing documents do not exist in a vacuum, and are reflections of events occurring in technology, law, and society. Further research should be devoted to the crosstemporal analysis of governing documents to help understand their development in terms of stakeholders. Currently, the major problem preventing this historical analysis is the access to documents over time. Many companies do not provide access to previous manifestations of governing documents. Researchers can help by creating sufficient archives of governing documents for temporal analysis. An example of this can be seen in the efforts of Kat Fitz (2007), who has created an archive of Second Life’s TOSs. The task of creating governing document archives is the first step to a historical analysis. This would provide better understandings about the evolution of governing documents and how they impact and are shaped by internal and external artifacts.
Virtual worlds, as a product of multiple interacting stakeholder groups (or social worlds), can be viewed as boundary objects (Star, 1989; Star and Griesemer, 1989). The stakeholder groups of users, designers, and managers all overlap, through individuals who span the boundaries between these stakeholder groups, and it is through these overlaps and through the synergy among these stakeholder groups that virtual worlds come into being. However, once they come into being, virtual worlds do not passively sit at the intersection of these stakeholder groups, but rather play an active role in reshaping relationships within and among the stakeholder groups. One potent example is the social capital created through weak ties among users (as well as, potentially, designers and/or managers acting as users) within a social world. Thus, virtual worlds can be characterized as boundary objects with agency (Fleischmann, 2006b; 2007b; 2007c).
Similarly, the governing documents used in virtual worlds also act as boundary objects with agency. These documents clearly not only are a product of multiple stakeholder groups (although with less input from users) and govern and reshape the relationships within and among these stakeholder groups both in virtual worlds and in the physical world (as the popularity of a virtual world or lack thereof may influence promotion and/or retention of designers as employees or contractors by managers, for example). Ideally, governing documents should be a meeting place for common understanding and interaction among the various stakeholder groups.
As such, it is important for all stakeholder groups, including users, designers, and managers, to play a real and active role in the development and modification of governing documents for virtual worlds. In essence, it is not fair for users to be influenced by governing documents that they had no role in developing or modifying. Since users who perceive this lack of fairness may respond by not participating in these virtual worlds and instead choosing to participate in virtual worlds with more democratic structures of governance, or, in the absence of that, create their own virtual worlds, it is also in the best interests of designers and managers to seek to involve users in the development and modification of governing documents.
In many different ways, the shift toward a digital society entails fundamental reconfigurations of forms of governance broadly defined.  New information and communication technologies have historically forced the reevaluation of laws and standards (Bimber, 2003). The development of laws of governance of virtual worlds is no exception. In trying to deal with the new legal situations created by the Internet and related technology, the U.S. government, for example, has significantly increased the attention paid to information policy (Braman, 2004; McClure and Jaeger, in press; Mueller, et al., 2004).
While the governance structures of commercial enterprises like virtual worlds will adapt far differently than governments will to the new technological environment, the development of government information policy may still be instructive. Over the past few years, government laws and standards related to information technologies have begun to more clearly relate to social aspects of information technologies, such as filtering requirements to limit children’s access to certain materials and fees created to expand access to the Internet for underserved populations (Jaeger, 2007). This may indicate that governance of private enterprises, such as virtual worlds, may begin to evolve toward more useroriented governance systems. However, stakeholders probably should take an active part in dealing with the issues identified in this discussion rather than waiting for them to work themselves out.
As the overall popularity and cultural penetration of virtual worlds increase, more attention will be focused on addressing its confusing governing structure. Regardless of stakeholders, the current virtual world governing structure is unable to handle the myriad of issues facing it, in a concrete way that is understandable to all parties involved. This obfuscated approach lacks stability, which in turn forces contentious issues between stakeholders to be resolved either by market pressures or by external government agents, most notably the legal system (Fairland, 2007). This volatile approach will likely lead to costly legal battles, which all stakeholders might avoid if governing documents were structured in a way that better incorporated the values of all stakeholders and were fashioned to be easily understood by all stakeholders.
As demonstrated above, the civil code documents that serve as contractual frameworks for governance of virtual worlds are indeed a complex and difficult issue to address, with a range of problem areas and a number of stakeholders with competing needs and interests. As a result of their inherent structure, governing documents often cross paths in a complicated webbing of rules and policies with no standardized naming structure, language and terminology, content, or location. This is compounded by the fact that each stakeholder possesses multiple and sometimes overlapping agendas and interests, which are not always easily visible or pronounced. While the task to interact with these documents is arduous, their integral importance as the sole source of civil code in the governance of virtual worlds demands that they be analyzed and understood. If these issues are not carefully considered, they will continue to become more difficult to resolve as virtual worlds proliferate, mature, and involve an ever-increasing number of users.
About the authors
Justin M. Grimes is a doctoral student in the College of Information Studies and a Graduate Research Assistant for the Center for Information Policy and Electronic Government at the University of Maryland.
Paul T. Jaeger is Assistant Professor in the College of Information Studies and Director of the Center for Information Policy and Electronic Government at the University of Maryland.
Principal contact for editorial correcspondence: pjaeger [at] umd [dot] edu
Kenneth R. Fleischmann is Assistant Professor in the College of Information Studies and Assistant Director of the Center for Information Policy and Electronic Government at the University of Maryland.
1. Malaby, 2006, n.p.
2. Fairland, 2007, p. 8.
3. Castronova, 2006, p. 68.
4. Balkin, 2006, p. 86.
5. Taylor, 2006, n.p.
6. Balkin, 2004, p. 2050.
7. Jankowich, 2006, p. 44.
8. Braman and Malaby, 2006, n.p.
Jack M. Balkin, 2006. Law and liberty in virtual worlds, In: Jack M. Balkin and Beth Simone Noveck (editors). The state of play: Law, games, and virtual worlds. New York: New York University Press, pp. 86120.
Jack M. Balkin, 2004. Virtual liberty: Freedom to design and freedom to play in virtual worlds, Virginia Law Review, volume 90, number 8 (December), pp. 2,0432,098.
Richard A. Bartle, 2006. Why governments aren’t gods and gods arent governments, First Monday, special issue number 7 (September), at http://firstmonday.org/issues/special11_9/bartle/, accessed 19 March 2008.
Erik Bethke, 2008. Better EULA, BetterEULA, at http://www.bettereula.com, accessed 24 March 2008.
Erik Bethke, 2007. $5000 to help me with the new GoPets EULA, Erik Bethke and The World of GoPets Blog, at http://erikbethke.livejournal.com/18138.html, accessed 19 March 2008.
Bruce Bimber, 2003. Information and American democracy: Technology in the evolution of political power. New York: Cambridge University Press.
Sandra Braman, 2004. Where has media policy gone? Defining the field in the 21st Century, Communication Law and Policy, volume 9, pp. 153182.http://dx.doi.org/10.1207/s15326926clp0902_1
Sandra Braman and Thomas M. Malaby, 2006. Preface, First Monday, special issue number 7 (September), at http://firstmonday.org/issues/special11_9/preface/, accessed 30 August 2008.
Edward Castronova, 2006. The right to play, In: Jack M. Balkin and Beth Simone Noveck (editors). The state of play: Law, games, and virtual worlds. New York: New York University Press, pp. 6885.
Entertainment Software Rating Board (ESRB), 2007. ESRB privacy online program, at http://www.esrb.org/privacy/, accessed 19 March 2008.
Kenneth R. Fleischmann, 2007a. Standardization from below: Science and technology standards and educational software, Educational Technology & Society, volume 10, issue 4 (December), pp. 110117.
Kenneth R. Fleischmann, 2007b. The evolution of agency: Spectra of bioagency and cyberagency, Information Society, volume 23, number 5 (October), pp. 361371.http://dx.doi.org/10.1080/01972240701572897
Kenneth R. Fleischmann, 2007c. Digital libraries with embedded values: Combining insights from LIS and science and technology studies, Library Quarterly, volume 77, number 4 (October), pp. 409427.http://dx.doi.org/10.1086/520997
Kenneth R. Fleischmann, 2006a. Doityourself information technology: Role hybridization and the designuse interface, Journal of the American Society for Information Science and Technology, volume 57, number 1 (1 January), pp. 8795.
Kenneth R. Fleischmann, 2006b. Boundary objects with agency: A method for studying the designuse interface, Information Society, volume 22, number 2 (April), pp. 7787.http://dx.doi.org/10.1080/01972240600567188
Kenneth R. Fleischmann and William A. Wallace, 2005. A covenant with transparency: Opening the black box of models, Communications of the ACM, volume 48, issue 5 (May), pp. 9397.http://dx.doi.org/10.1145/1060710.1060715
Joshua Fairland, 2007. Antisocial contracts: The contractual governance of online communities, Indiana Legal Studies Research Paper, number 89, pp. 168.
Kate Fitz, 2008. Terms of service (Second Life) — History of changes, Second Life LawSpot, at http://www.lawspotonline.com/lawspot/termsofservice.jsp, accessed 24 March 2008.
Chek Yang Foo and Elina M.I. Koivisto, 2004. Defining grief play in MMORPGs: Player and developer perceptions, Proceedings of the 2004 ACM SIGCHI International Conference on Advances in Computer Entertainment Technology. New York: ACM Press, pp. 245250.
Jens Grossklags and Nathaniel S. Good. 2007. Noticing notice: A large-scale experiment on the timing of software license agreements, Conference on Human Factors in Computing Systems. New York: ACM Press, pp. 607616.
Paul T. Jaeger, 2007. Information policy, information access, and democratic participation: The national and international implications of the Bush Administration’s information politics, Government Information Quarterly, volume 24, pp. 840859.http://dx.doi.org/10.1016/j.giq.2007.01.004
Andrew Jankowich, 2006. EULAw: The complex web of corporate rulemaking in virtual worlds, Tulane Journal of Technology and Intellectual Property, volume 8 (Spring), pp 160.
Javacool Software, 2007. EULAlyzer, at http://www.javacoolsoftware.com/eulalyzer.html, accessed 19 March 2008.
Peter S. Jenkins, 2004. The virtual world as a company town — Freedom of speech in massively multiple online role playing games, Journal of Internet Law, volume 8, number 1, pp 121.
Raph Koster, 2006. Declaring the rights of players, In: Jack M. Balkin and Beth Simone Noveck (editors). The state of play: Law, games, and virtual worlds. New York: New York University Press, pp. 5567.
Leandra Lederman, 2007, ‘Stranger than fiction’: Taxing virtual worlds, New York University Law Review, volume 82, number 6, at http://www.law.nyu.edu/journals/lawreview//issues/vol82/no6/NYU602.pdf, accessed 19 March 2008.
Lawrence Lessig, 1999. Code and other laws of cyberspace. New York: Basic Books.
Larry Magid, 2007. It pays to read license agreements, at http://www.pcpitstop.com/spycheck/eula.asp, accessed 19 March 2008.
Thomas M. Malaby, 2006. Coding control: Governance and contingency in the production of online worlds, First Monday, special issue number 7 (September), at http://firstmonday.org/issues/special11_9/malaby/, accessed 19 March 2008.
Charles R. McClure and Paul T. Jaeger, in press. Government information policy research: Importance, approaches, and realities, Library & Information Science Research.
Milton Mueller, Christiane Page, and Brenden Kuerbis, 2004. Civil society and the shaping of communicationinformation policy: Four decades of advocacy, Information Society, volume 20, pp. 169185.http://dx.doi.org/10.1080/01972240490456845
Obfuscated code, 2008. Wikipedia, at http://en.wikipedia.org/wiki/Obfuscated_code, accessed 18 March 2008.
Pollstar v. Gigmania Ltd., 2000. 170 F. Supp. 2d 974.
Joel R. Reidenberg, 1998. Lex informatica: The formulation of information policy rules through technology, Texas Law Review, volume 76, number 3, pp. 553584.
Ryan Singel, 2008. U.S. spies want to find terrorists in World of Warcraft, Wired, at http://blog.wired.com/27bstroke6/2008/02/nations-spies-w.html, accessed 19 March 2008.
Susan Leigh Star, 1989. The structure of illstructured solutions: Boundary objects and heterogeneous distributed problem solving. In: Michael N. Huhns and Les Gasser (editors). Distributed artificial intelligence, volume 2. Menlo Park, Calif.: Morgan Kaufmann, pp. 3755.
Susan Leigh Star and James R. Griesemer, 1989. Institutional ecology, ‘Translations,’ and boundary objects: Amateurs and professionals in Berkeleys Museum of Vertebrate Zoology, 19071939. Social Studies of Science, volume 19, pp. 387420.http://dx.doi.org/10.1177/030631289019003001
Synthetic Worlds Initiative (SWI), 2007. A declaration of virtual world policy, Synthetic Worlds Congress, at http://www.studiocypher.com/files/ludium2.html, accessed 19 March 2008.
T.L. Taylor, 2006. Beyond management: Considering participatory design and governance in player culture, First Monday, special issue number 7 (September), at http://firstmonday.org/issues/special11_9/taylor/, accessed 19 March 2008.
Bruce Sterling Woodcock, 2007. An analysis of MMOG subscription growth, at http://www.mmogchart.com, accessed 19 March 2008.
Paper received 25 March 2008; accepted 2 August 2008.
Copyright © 2008, First Monday.
Copyright © 2008, Justin M. Grimes, Paul T. Jaeger, and Kenneth R. Fleischmann.
Obfuscatocracy: A stakeholder analysis of governing documents for virtual worlds
by Justin M. Grimes, Paul T. Jaeger, and Kenneth R. Fleischmann
First Monday, Volume 13 Number 9 - 1 September 2008
A Great Cities Initiative of the University of Illinois at Chicago University Library.
© First Monday, 1995-2015.