By Matthew Becker
Edited by Matt Gelfand
The application of copyright in the space of virtual worlds has been a subject of increasing consideration in the legal community over the past few years.[i] Literature on this subject has often centered on two focal points: the existing laws and approaches that are likely to produce successful litigation in this evolving arena; and the idea that the best approach to settling copyright disputes might be to try to find recourse through the entity that owns and operates the virtual world, rather than through litigation. Less common, however, is a substantive analysis of why the existing copyright regime generally fails to provide a suitable venue for addressing grievances, and how it could be reformed to better suit the virtual context. The purpose of this article is to foster such a discussion by exploring the disconnect between the copyright regime in the United States, which has evolved in a physical environment, and the distinct problems and requirements that arise in a virtual environment. In the process, this article will explore three options for ameliorating the situation – two that are legal in nature, and one that is extra-legal.
The term “virtual worlds” is quite broad, ranging from goal-based games like World of Warcraft, to the so-called “pure” environments like Second Life (worlds that are open ended, without explicit objectives). This article will limit its study to the latter category. “Pure” virtual worlds present a wider variety of potential copyright issues because they afford users a greater opportunity to act as content creators and, in some cases (as in Second Life), to retain copyright for their creations.[ii]
A brief explanation of content creation in virtual worlds is in order. In general, users construct every aspect of these worlds, from the landscapes and buildings, clothes and objects, to textures and animations (the actual movements or actions that a user can make his or her avatar perform). Just as these products can be created, they can also be sold to other users, often via in-world stores. However, within this market for content, there are users who employ sophisticated programs to allow them to “copy” this content and sell it as their own, often sparking outcry from content creators, and sometimes inspiring litigation.
The Existing Copyright Regime
The copyright regime was developed to incentivize and support two columns of production: idea creation/expression and reproduction/distribution. We want to incentivize not only the creation and expression of ideas, but also their spread and distribution. Society has an interest in both the creation and spread of new ideas, such that others may reciprocate in turn. For this reason, the ideas themselves are not copyrightable, but the expressive works that serve as vectors for these ideas generally are.
In the physical world, we often take for granted that individuals and firms will be less inclined to create new products if they cannot exclusively reap the benefits of these creations for some defined period of time. Similarly, if we want firms to engage in the reproduction and distribution process on a large scale, it is customary to assume that these firms should be guaranteed some monopoly on new products (those without a successful history in the market). There are fixed costs to the creation and distribution of content in a physical environment, and it is often assumed that much of this production would not occur if the creators and distributors were unable to secure certain privileges with regards to their creations.[iii]
Distinguishing Elements of Virtual Worlds: Scalability and Intrinsic Motivation
The context of virtual worlds is one where the elimination of costs of reproduction and physical barriers to distribution, in tandem with the effects of cooperative and communal production structures, may render inapplicable much of the traditional dogma pertaining to innovation and distribution. As illustrated by examples of commons-based peer production,[iv] such as Wikipedia, and the research of scholars such as Clay Shirky[v] and Eric von Hippel,[vi] individuals often innovate for reasons other than immediate economic gain—sharing the products of their innovation freely with friends and strangers. This effect is especially prevalent in online communities, and pure virtual worlds are no exception. Many of the important innovations that take place in pure virtual worlds are produced without direct profit-making incentives (Edward Deci terms this type of production intrinsic motivation, in contrast to the pecuniary nature of extrinsic motivation).
Clay Shirky has taken Deci’s notion of intrinsic versus extrinsic motivation and applied it to the online context, noting that intrinsic motivation appears to be a significant driving force behind many forms of creation in that environment.[vii] One interesting facet of these motivating factors is that they do not appear to be additive. That is, taking an intrinsically inspired project (such as the pleasure of solving a puzzle) and adding a monetary incentive through a pay structure does not double the effort. Rather, it causes the individual to approach the project differently, as a job, putting in only as much effort as he or she will be compensated for. Subsequently removing the monetary incentive does not cause the individual to revert to his or her intrinsically inspired state, but instead diminishes interest in the activity.
Scalability serves as another important distinguishing factor between virtual worlds and the physical world. In the physical world, of course, there are per-unit costs of reproduction and distribution. As a result, large-scale reproduction and distribution is generally limited to firms. In the virtual context, scalability is not so limited, as virtual products can be reproduced and distributed at almost no cost. Mass distribution can occur in a virtual world without the need for firms. This is unlike distribution in the physical world, where firms are necessary in order to efficiently produce and distribute products at scale, perhaps requiring incentives (such as copyright monopolies) to support their business. An individual in a pure virtual world could, in theory, either create or appropriate through reproduction all of the products that he or she needed or desired. The actual reproduction is essentially costless.
Of course, fixed initial costs of creation are an issue. Traditionally, much of the incentive for initial content creation is assumed to come from the pursuit of future profits. There are many examples of virtual entrepreneurs who have opened businesses in pure virtual worlds such as Second Life, creating and selling items and animations that would fall under the protection of copyright. However, there are also many examples of individuals and organizations that have created content without immediate extrinsic motivations.[viii] Unfortunately, no study yet exists quantifying and comparing these modes of production, although it appears that both forms of motivation play important roles for content creation in pure virtual worlds.
Copyright Law in the Virtual World Context: Problems and Considerations
So what is wrong with the application of existing copyright law to pure virtual worlds? The copyright regime developed over centuries in the context of the physical world, where large firms were generally required for the reproduction and distribution of content. The laws of that period are not well-suited for virtual environments, in which individuals yield low profits and often have neither advanced knowledge of the legal system nor access to counsel. Copyright infringement in pure virtual worlds would often result in small damage awards (given the low prices of most products in virtual worlds), and are usually not worth litigating after legal fees are considered. Although existing law does allow for the collection of statutory damages and attorneys fees, such collection is only available for copyrighted material that has been registered with the Copyright Office before the infringement began (or within a grace period of three months following publication).[ix]
The burden of formally registering a copyright is low for institutional actors such as firms, but is substantially higher for creators in virtual worlds. First and foremost, registration is unlikely unless an individual understands the benefits that come with registration. Secondly, there are associated costs in terms of time and money. The registration requires a filing fee of $35, which seems insubstantial for a product like a book or album, but may be quite steep for a virtual good that will sell for only a fraction of that cost, and for which there is no guarantee of a market.[x] In short, the costs of registration may outweigh the benefits for many pure virtual world creators. Additionally, further barriers to litigation arise as a result of the particularities of virtual worlds, such as the difficulties of identifying the individuals behind infringing avatars (online personas), obtaining proper jurisdiction, and providing notice of copyright.
For these reasons, the primary recourse for pure virtual world creators suffering from copyright infringement has been to contact the owner or operator of the virtual world. Perhaps the best-known operator is Linden Labs, which runs Second Life. One option is for a creator to contact Linden Labs with a complaint about copyright infringement by another user, and ask Linden Labs to investigate these claims and impose sanctions and remedies according to its terms of service (TOS) or end-user license agreements (EULAs). However, Linden Labs has often been accused of employing a “hands off” approach in policing the actions of Second Life users, even when such actions have infringed the putative rights of others.[xi]
It is also possible for a copyright holder to file a Digital Millennium Copyright Act (DMCA) takedown notice, which demands that Linden Labs remove the offending material. In many circumstances the infringement complaint noted above might qualify, de facto, as a DMCA notice, so long as it fulfills certain requirements.[xii] Although such takedown notices do not have the force of judicial injunctions, virtual world operators can be sued for infringement if they choose not to take action on a takedown notice.
Finally, pure virtual world operators have tried to stop copyright infringement through extra-legal means like software protections, such as those allowing creators to lock certain items so that they cannot be duplicated, or locating each merchant’s product inventory in a centralized repository to limit access by potential infringers.[xiii] Unfortunately, infringers have been consistently successful at discovering means around these barriers, and are likely to continue to do so in the future.
There are further complications that might arise as pure virtual worlds develop and expand in popularity. One such development is the use of what is referred to as a “hypergrid” — a means of interconnecting multiple pure virtual worlds so that avatars and items may be ported between them. Although the concept is still in its infancy, it suggests a number of potential issues with respect to the policing capabilities of a decentralized network, and foreshadows the complexities that may arise as this field continues to mature.
Copyright Alternatives: the “Strong” Approach
Given these difficulties, it appears that neither existing legal nor extra-legal options have been suitably effective in dealing with copyright infringement in pure virtual worlds. The question, then, is how the application of copyright to virtual worlds might be modified in a manner specific and restricted to that context, so as to suit its unique features and complexities. In an attempt to answer this question, this article will consider two options on opposite poles of the copyright spectrum, with a “strong” copyright regime at one end and an “open” regime on the other.
A “strong” copyright regime would be one that employed terms that are similar or identical to the existing U.S. legal regime, but with some modification to enhance or further enable enforcement. For instance, one solution might be to extend statutory damages and/or attorney’s fees to any registered copyright, regardless of whether it is registered after infringement has begun. The effect of this would be to incentivize virtual world copyright litigation. Such a system would effectively function as a stricter version of our current copyright regime.
Although this approach would probably be effective in reducing copyright infringement, it would do so by substantially increasing the prevalence of litigation, with all of the costs that this might bring. For instance, extending opportunities for the collection of statutory damages, starting at $750 per infringed work,[xiv] would probably stifle infringement. However, it might also produce a chilling effect on many legitimate activities, given many individuals’ limited knowledge and uncertainty regarding the boundaries of fair use.[xv] Although this might incentivize settlement early in the litigation process, the substantial penalties for the losing party might serve to undermine fair settlements, and could be used to extract substantial sums from small-time, non-commercial, or even innocent “infringers.”
Another argument against a “strong” copyright regime would be that it prioritizes the monetary (extrinsic) motivations of individuals, perhaps to the detriment of intrinsic motivations. One of the facets of online environments that demonstrate a high degree of intrinsic motivation (such as commons-based peer production environments like Wikipedia) is that the participants in those environments espouse a notably anti-commercial ethos. It is conceivable, therefore, that if the environment of pure virtual worlds becomes highly litigious, it would dampen free and open production by those who prefer to be intrinsically motivated. In effect, it might squeeze out free production and replace it with largely commercial production.
IP is RL: an “Open” Copyright Approach
On the “open” end of the spectrum, one might ask why it is necessary to provide such strong copyright protection for creations in pure virtual worlds in the first place, given the difficulties of litigation and the unique facets of production and distribution in the virtual space. Essentially, this argument would advocate for a copyright structure specific to virtual worlds that de-prioritizes individual ownership or control of rights, while emphasizing attribution to the content creator (as recognition provides an important source of non-pecuniary motivation), effectively creating a moral right of attribution while rejecting the economic rights traditionally associated with the copyright regime. As noted above, a substantial portion of pure virtual world content is created without monetary motivation, but is instead inspired by intrinsic and non-pecuniary motivations. Furthermore, the reproduction and distribution that might be required in order for other pure virtual world users to enjoy these creations is without cost.
Whether one accepts such an argument or not, the effects of an “open” regime would likely be just as radical as those of the “strong” regime. Most immediately, one would expect the dampening of much of the commercial production in pure virtual worlds, especially among those creators who have made this their full-time job.[xvi] In addition to reducing the output of these full-time creators, an “open” regime would also disincentivize commercially-driven hobbyists. Consider, for instance, the numerous cases of popular user-created Second Life games that have been picked up by outside companies interested in porting these games to cell phones or console gaming systems.[xvii] If such products were initially released in a virtual world without coverage by traditional copyright protections, there may be a question as to whether these creators would have traditional rights to their works outside of virtual worlds. This might disincentivize the creation of games or products by individuals who view these virtual worlds as testing grounds for their products, with an eye toward eventual commercial release.
Moreover, the implications that an “open” copyright system might have for deterring future investment by businesses in the virtual space form another concern. Many people who have studied pure virtual worlds have noted the potential for these spaces to become significant social and commercial hubs, analogizing the current state of the pure virtual worlds to the early Internet.[xviii] To whatever extent these predictions are accurate, if pure virtual worlds are to take a path similar to that of the Internet, they will likely require significant investment and participation by businesses and commercial entities, in addition to intrinsically motivated groups and individuals. An “open” copyright regime might retard pure virtual worlds such that they retain their smaller, largely intrinsically motivated, amateur character (much like the Internet exhibited for many years), unable to expand through large-scale commercial investment.
An Extra-legal Option
Although the alterations of the existing legal system noted above may not seem propitious, there remain extra-legal options that might reduce instances of infringement in virtual worlds. Most promising among these options is an arrangement that draws on a simple architecture to foster group-governance dynamics, much like the “code-backed norms” of Jonathan Zittrain’s writings. For example, operators of virtual worlds could set up a centralized system that would allow users to identify potential infringers, thereby flagging these merchants and encouraging other users to exercise discretion when considering a purchase. Such a system might be similar to the detailed seller rating system used by eBay, although tailored to address copyright issues (in addition to more typical product-related comments).[xix]
To provide a specific example, Linden Labs could build on its Second Life Marketplace website — a centralized store for Second Life— to incorporate comments or ratings on the merchants themselves. (At the moment users can rate individual products and leave product-specific comments, but cannot comment on the merchant itself.) Such a merchant-centric rating system could allow users to flag merchants that they believed to be selling illicitly copied products by placing a warning icon on the merchant’s products. Registering an infringement complaint about a merchant on this website might discourage other users from purchasing the merchant’s products, and could even be used to link back to the actual copyright holder’s account, encouraging users to purchase from the content creator. Like eBay, the site might also employ an algorithm to de-prioritize the search rankings of merchants that are the subject of multiple complaints. Such an approach would draw on the community dynamics of virtual worlds to help the community regulate its environment, reducing the need for recourse to an ill-fitting legal system, and employing the very characteristics that make this environment unique. Although such a system might not be effective in limiting all forms of infringement (such as infringement that takes place outside of a merchant environment), it would nevertheless provide an additional tool to aid content creators, and it would serve as an example of an extra-legal solution that does not restrict the incentives that have helped virtual worlds to grow thus far.
Considering the options that have been discussed above, it is with an eye towards the future development of virtual worlds that any changes to the existing copyright system must be considered. As new users, new businesses, and new technologies expand the horizon of virtual worlds, there will be a commensurate change in the requirements for this space, not least of which are effective legal and extra-legal remedies. Any such remedy must draw on the unique characteristics of this environment to foster further growth, not retard development through ill-fitting solutions.
[i] See, e.g., the Summer 2010 and Fall 2010 issues of the ABA’s SciTech Lawyer, which focus on legal issues in virtual worlds.
[ii] Second Life is also interesting because its currency, the Linden Dollar, is freely convertible into U.S. currency, and vice-versa. For the current exchange rate, along with other data on the volume of currency exchanged, see LindeX Market Data, Second Life, http://secondlife.com/statistics/economy-market.php (last accessed Nov. 26, 2010).
[iii] Although it is beyond the scope of this article, it should be noted that there are interesting and worthwhile arguments against some of the traditional assumptions that are said to underlie copyright, as well as proposals for alternative incentive structures. See, e.g., William W. Fisher III, Promises to Keep: Technology, Law, and the Future of Entertainment (2004).
[iv] Commons-based peer production is a term coined by Yochai Benkler. See Yochai Benkler and Helen Nissenbaum, Commons-Based Peer Production and Virtue, 14 J. of Pol. Phil. 349, 349 (2006). Available at: http://www.nyu.edu/projects/nissenbaum/papers/jopp_235.pdf.
[v] See Clay Shirky, Cognitive Surplus: Creativity and Generosity in a Connected Age (2010).
[vi] See Eric von Hippel, Democratizing Innovation (2005).
[vii] See Clay Shirky, Cognitive Surplus: Creativity and Generosity in a Connected Age 70-95 (2010).
[viii] Examples of both extrinsically and intrinsically motivated creators can be found by looking at the Second Life Marketplace, which features products available at a wide range of prices, including a significant proportion of free items. Second Life Marketplace, https://marketplace.secondlife.com (last visited Nov. 26, 2010); see also Robert D. Hof, My Virtual Life, BusinessWeek, http://www.businessweek.com/magazine/content/06_18/b3982001.htm (May 1, 2006) (noting that in 2006, there were 3,100 individuals in Second Life earning an average of US$20,000 annually); Life 2.0 (Palmstar Entertainment 2010) (documenting the lives and difficulties of virtual world entrepreneurs, including instances of copyright infringement).
[ix] 17 U.S.C §§ 412, 504.
[x] Although the impact of this fee might be mitigated by registering multiple works simultaneously, this requires either that all the works be unpublished, or if published, that they were all published on the same day. Whether it will be viable for a virtual world content creator to hold off on publication until a number of new products have been amassed and registered (sufficient to match the filing costs), or whether it is common for a creator to publish so many products on the same day, will likely differ on a case-by-case basis.
[xi] See Life 2.0 (Palmstar Entertainment 2010) (discussing Linden Lab’s reluctance to police disruptive behavior).
[xii] 17 U.S.C §512(c)(3)(B) (2010).
[xiii] In the Second Life Marketplace, restrictions may be set for copying, modifying, or transferring the products sold. Second Life Marketplace, https://marketplace.secondlife.com (last visited Nov. 26, 2010).
[xiv] 17 U.S.C § 504 (Statutory damages range from $750 to $30,000 per infringed work, and may be as high as $150,000 for willful infringements).
[xv] Similarly, it is worth noting that legal fees for any litigated claim would likely be far in excess of actual damages.
[xvi] Although donations would still be possible (either explicitly or through the purchase of items that could simply be copied for free), it seems unlikely that this would provide adequate income for most full-time content creators.
[xvii] Robert D. Hof, My Virtual Life, BusinessWeek (May 1, 2006) (describing two such examples of games that were purchased for distribution outside of Second Life).
[xix] It must be noted that Linden Labs has proposed a somewhat similar “content seller program,” described as “a program that sellers may participate in if they meet certain eligibility standards intended to show a level of trustworthiness and quality of content.” Such a program would require each merchant to opt-in and also meet certain minimum requirements (such as filing identity information with Linden Labs, and meeting yet-unspecified transaction thresholds), with the benefit being that they would receive some kind of certification or seal to differentiate them from other merchants (thus creating an “elite” class of merchants, a prospect that some have criticized). Although such a system might be effective in highlighting more popular merchants who have a proven track record, it must be differentiated from the model proposed by this article, as Linden’s system appears to lack any kind of content theft deterrence mechanisms. That is to say, there is no immediate reason why infringing merchants would be disincentivized from continuing to resell other merchants’ IP.