Terrorists in virtual worlds?

I came across this article in Computerworld the other day that starts by talking about how the FBI (yes, the freakin' FBI) raided the apartment of two University of Michigan students to investigate "potentially fraudulent sales or purchases of virtual currency that people use to advance in the popular online role-playing game World of Warcraft."

An article from annarbor.com provides some more details, saying that:

"Investigators seized laptop computers, hard drives, video game systems, credit cards, a cell phone, paperwork and other computer equipment, documents say. Investigators were seeking records of any online transactions with WOW, the Chinese-based gold-farming website www.gameusd.com, eBay, PayPal and the United Services Automobile Association, which offers services including online banking."

"No arrests have been made, FBI Special Agent Sandra Berchtold, a bureau spokeswoman in Detroit, said Wednesday. Berchtold said she could not comment further on the March 30 raid because many documents in the case remain sealed."

What in the? The FBI involved in something like gold selling? Really? Doesn't this seem kinda, what's the word... excessive to you? I mean shouldn't the FBI have more important things to investigate? But the computerworld article goes on to talk about a Canadian study that apparently found:

"a 'dark universe' where terrorists or 'targets of interest' have moved their operations. Virtual world terrorism facilitates real world terrorism: recruitment, training, communication, radicalization, propagation of toxic content, fund raising and money laundering, and influence operations."

This Canadian report also refers to something called the Reynard Project. The proposal for the Reynard Project was released in 2009 by the Intelligence Advanced Research Projects Activity (IARPA), an agency of the US federal government. The proposal says that it is looking for research to help them identify how behavioral indicators within virtual words are predictive of the real world characteristics of the users.

Some intriguing sections I found in this proposal:

"A new channel for information exchange and social interaction is emerging with the growing popularity of Virtual Worlds (VWs) and Massive Multiplayer Online Games (MMOGs). Industry experts predict the number of VW users will exceed a half billion by the end of 2009. As participation in these VWs broadens and deepens, the need for understanding the nature of interactions in these worlds arises."

It asks proposers to consider the following sections:

Avatars and Representation: What characteristics of an avatar may reveal something about the real-life person behind it? What motivates avatar choice, and can we predict when an individual might select certain characteristics over others? Does avatar selection reveal something about an individual’s real-life personal characteristics, attitudes, customs, groups, or culture?

Communication: Communication mechanisms in VWs may be both verbal and nonverbal. Verbal: Today the primary means of verbal communication in VWs is text-based chat and messaging. Unique language patterns have emerged in chat and texting, characterized by use of abbreviations such as LOL (laugh out loud) or BRB (be right back). Can we determine whether the person is a native-language speaker of the language (e.g., English) being used?

Can the path of “viral ideas” be traced through the chat history of the residents, as it apparently can in exclusively text-based media, which may then allow inferences about the degree of influence of the propagators of the ideas? Nonverbal: When and why is nonverbal communication used?

Things Avatars Do: Individuals inhabiting VWs engage in a variety of activities... to what extent might cues in these activities provide indicators about the RW person?

Group Formation and Dynamics: The massive multiplayer nature of VWs leads to extensive social interaction among its inhabitants. As in the RW, individuals form social bonds with others, usually after some period of interaction and trust establishment. What types of groups form, what are their characteristics, and what supports their continuance? Does group membership reveal anything about the RW individuals who belong to them? Do individuals recognize others as from the same culture or different cultures? If so, what factors or characteristics do they use to recognize each other?

Economics and Money: One of the most important modes of human and group interaction is bartering and exchange. Do certain economic decisions or activities vary depending on the culture or nationality of the user? Does handling of VW currency vary depending upon the RW culture of the user? What inferences can be drawn about the RW individual based on their treatment of VW goods and currencies?

Cultural Differences: Very little research has been done to date comparing the similarities and differences in motivation, usage patterns, and behaviors of VW users across multiple cultures. The Reynard program is interested in understanding possible cultural differences in VW usage, and encourages researchers to study non-U.S.-based players.

I find these topics and questions that the Reynard Project is asking to be really quite interesting. But for an agency of the federal government to be interested in this kind of research, there has to be a reason right? I mean they can't just be interested in game development or the proliferation of virtual worlds in our society or whatever. So are they really asking for research on these subjects so that they can really track down terrorists within these virtual worlds?

Finally, on a related subject a recent book was released by Emile van Veen, a WoW player. The fiction novel is called MMORPG: A World of Fun and Games... and Terrorism. In the book terrorists are using WoW for secret communication, to rehearse strikes, and to elude any possible electronic surveillance by intelligence agencies.

I'm intrigued by this book (interview with the author here), but also a bit freaked out by the possibility of federal intelligence agents tracking behaviors in game.