• The world will be the metaverse. People often think of Stephenson’s metaverse as an “other” place, and the web as a window onto cyberspace, but as Paul Saffo and Mike Liebhold of Institute for the Future note, the best model for the metaverse of 2016 may be an information-drenched world, where the 3D web is just one particular instantiation. Mixed reality is likely to be the dominant user experience. You will use virtual worlds when they are an appropriate mode of interaction, but they are not your primary mode of communication – you have your chat, your email, your augmented reality, your 2D and 3D browser, etc. While people will continue to use online spaces and media centers for particularly high quality 3D content, the pervasiveness of information access and augmented reality will give world itself new layers of “metaverse-itivity.” The ubiquity of small, portable Sidekick-like and wearable devices will enable immediate access. Voice will be used for many basic queries, but text, even IM text, is private and unobtrusive, so it will not disappear.

• Privacy law will be an increasingly important area of political and legal debate in the coming decade, where personal freedoms must be balanced with law enforcement and national security need to keep electronic communities transparent. As usual most of our legal and policy innovations will be reactionary, in response to new invasive technologies, security calamities, or landmark court decisions. See Spying with Maps: Surveillance Technologies and the Future of Privacy, Mark Monmonier, 2004 [51], for more on the invasive technologies ahead, which can be exploited more readily by hackers and corporations than the government, if history is precedent. Building immunity and respecting civil liberties in the metaverse will be a great challenge, but one we can meet.

• In 10 years, mobile handheld devices will bring us a billion new global users of the web, and significantly more 3D microcontent. As usual, some will use the web for entertainment and some will use it for work. Virtual worlds will be terrific game spaces, good places for instruction, and moderately useful for collaboration, but less pervasive than many think.

• In 2016, a handful of early adopters of wearable (most) and implantable (few) sensors [52] will be physiologically connected to the web, but this will be small scale and most implant work may not occur in the U.S. One can forsee research devices (cochlear implants, “brain ports” for spinal cord injury patients, etc.) that can send wireless telemetry, even update their software and hardware (FPGA) remotely. A few implantable sensors may be in use by pioneers, such as an implant that reports out your day-by-day biochemistry, perhaps even to a refrigerator, which in turn can make a custom electrolyte and vitamin mix, a customized diet that will be proven to extend vitality and longevity (in mice). We can imagine an implant that reminds you to exercise when your stress hormones rise, even suggests how long, and that talks with your wearable GPS/heart monitor. Implants in addiction medicine (alchohol, drug, obesity, etc.) that monitor blood levels and provide nausea after the inappropriate behavior to condition behavior change are another fascinating frontier for behavioral science research (again, unlikely to be seen first in litigious developed countries). It follows that if FDA approval for use of many such pioneering implantable internet interface devices is very unlikely to occur by 2016, early adopters of any such devices will be overseas or a handful of law-breakers here. And with regard to implants vs. wearables, even beyond 2016 the large majority of implant proposals are unlikely to make sense for experimentation beyond the laboratory, as their benefit is generally quite marginal relative to cost. In the long run, given broad stigmas against modifying the individual by invasive implants, and the ease with which we can modify our physical and virtual environment instead, humans are far more likely to be augmented by wearable devices and by adding "situational intelligence" around us than by adding hardware inside our body. For example, biometric interfaces for identity verification, security, etc. (eg., automatic identification by face, in a crowd) will develop far faster and see much greater global diffusion than implant technology in the foreseeable future, regardless of science fiction scripts. As the saying goes, "human nature doesn't change, but our houses [surroundings] get exponentially more intelligent every year."

• A unified metaverse may never occur. Googleverse won't be a dominant world in the 2016 time frame. For the time being we can expect more boom and bust, and falling profit margins for virtual worlds. Increasing failure of the centralized content development and centralized distribution model. A "balkanization" of virtual world space, with a few common standards and a wide variety of creation platforms, the way websites are made today. The ubiquity and availability of creation tools will provide so much new content that specialty producers and value adders will gain new power and audience. In this "thousand flowers/long tail" environment (Chris Anderson, The Long Tail: Why the Future of Business is Selling Less of More, 2006 [50]), small producers will advance relative to the top companies. A new set of big companies that are successful aggregators will also emerge. Recall the diversity of DVD titles (40K of the 60K total inventory) that are rented on any particular day by the five million customers of Netflix, a current leading 3D content aggregator, in 2006. A significant minority of the population will be very interested in unique video and virtual world content personalized to their communities of interest. There will be some shared things – identities through syndication tools (RSS model) bridging a gap between environments. Larger commercial VR games will continue; existing VR worlds will be more about the infrastructure than the content provider. Companies providing content experiences will have turned to user-created content more than programmer-created stuff. Virtual worlds will continue to have differing currencies, so there will be a banking industry to help convert currencies. As liability law develops, at least one major VW operator will move operation overseas to dodge regulation. Bigger operators will tighten identity authentication and consequences for violating EULAs.

• One of the biggest places we will see the impact of virtual worlds is education. Educational software continues to improve. The JumpStart programs of Knowledge Adventure are an excellent and affordable start on using 3D games for education. Second Life has a number of small scale educational projects [53] underway in world. There are a number of small independents like Learning Sites (archeological visualizations for educational and research). Our current educational system is so bad, particularly primary and secondary, that youth will desire to move to this space any chance they get. There is great promise ahead, though the software must be tied to good pedagogy and assessment. Another huge advance will come when large numbers of kids know how to use 3D software, and to plug their creations into virtual worlds. Dassault's new Cosmic Blobs, a kid-friendly 3D animation software platform, is the first major tool for what we may come to call "youth-created content" for the metaverse.

• The need for an editorial role for all types of content will increase and diversify, moving from today's portal content editors, to social, community-based search and aggregation. The more participatory the web becomes, the more we need to be able to choose our favorite editors for content filtering. Some of these will even be paid by us, on a micropayments system.

• Virtual worlds enhance play. As developmental psychologist Jean Piaget notes (Play, Dreams, and Imitation in Childhood, 1962) [55], playing is a fundamental component of human development. Our evolutionary psychology is wired to experience joy from play, and we are strongly motivated by it, particularly in youth. To the extent that playing in synthetic worlds can be richer and more unique than in the physical world, and our play can be done behind a wall of privacy when needed, these destinations can satisfy psychological needs in ways the physical world cannot. By 2016, developmental psychologists have teamed with the larger educational software companies to give us new virtual and kinesthetic play worlds, each integrated with the other. Kinesthetic play toys like Lego's Mindstorms NXT robotic construction kit are integrated with virtual preflighting environments, as well as virtual worlds with looser laws of physics where robots can be built, and used to fight and explore.

• Scientific computing. Towards 2020 Science (86 pages, recommended), 2006, Microsoft Research.

• 'Virtual Reality' television. The Japanese Ministry of Communications has established a blue sky research group to develop plans to commercialize virtual reality television by 2020 [70]. The group is also investigating the potential of related technologies (haptics, etc.) to facilitate touch and other senses.

• International security and socioeconomic development. Mapping the Global Future, 2005 (123 pages, highly recommended). Report of the National Intelligence Council's 2020 Project.

• Open vs. proprietary 3D worlds cycle. There was much interest and buzz around the idea of the 3D-enabled Web in the mid- and late-1990's, at the time of the open standards VRML and VRML2. This was similar to the buzz we see emerging around virtual worlds like Second Life now, which are based on proprietary standards. Second Life is releasing APIs to encourage mashups and third party development, and has committed itself to eventually moving to open standards. Yet seems very likely that as virtual worlds continue to improve, new proprietary standards will be developed, by Second Life or a competitor, that will yield significant new VW functionality and thus be worth the effort of user migration to the new platform. At some point those standards will become open as well. In sum, a historical look at standards tells us that the move from proprietary to open is just one half of the development cycle, and when we focus only on that we see only half of the economic and technical picture.

• Avatar representational cycle. There has been a cyclic history representational accuracy in avatars concerning how cartoony, lifelike, useful, and playful they are, alternating between highly realistic and highly caricatured. The Uncanny Valley principle, which encompasses a wide range of human responses to the representation of humans in robots and animations, is one reason we may be seeing representational cycles. Represented objects need to be able to avoid the valley of human distaste for "almost human" likenesses and responses, and either retreat in to caricatures or make the avatars so highly realistic that they don't trigger an unpleasant response. So far, caricature is the most accepted, though there are continual attempts at making lifelike avatars. Haptek's avatars, such as Baba Dim Sun, are one example of avatars that have been said to be "in the valley." Perhaps this has been one reason they have had much less market adoption than more cartoony alternatives. Such cycles also exist in behavioral representation. We've made NPC avatars that have too much or too little interactivity and chattiness relative to the sophistication of their AI. In general, caricature is preferred, although there are constant attempts to move beyond this into highly realistic simulation for specialized purposes.

• Game revenue model and content development cycles. We have seen a regular 7-10 year cycle in the MMO industry regarding how users pay to play and how game content is generated. We may be due for another such major shift in either revenue model or content development, or both. One such potential shift, involving game aggregation services, is outlined in Ideas and Proposals (15Ba). Raph Koster quote: "In 1997 there was an apocalyptic event that killed almost all the existing MMO providers. It was the shift to a subscription-based business model paired with game level production values. The thing was the dinosaurs that got killed off were the heirs of an earlier apocalypse in 1989, which was the shift in earlier business models to an hourly closed service model, who in turn were heirs to a shift in 1982 from academic VWs with no business models. In all those shifts, the existing companies pretty much all died. And we're due for another shift. The last explosion was 1996, 1997 (Dark Sun, Ultima, Lineage, Asheron's Call, Active Worlds, and others). I think there will be a production shift married to a business shift and the Blizzards of the world will face a new round of mammals. EA seems a particularly good Goliath candidate, to my mind."

• Game migration cycles and the "Virtual Gold Rush." Historically, when online virtual worlds achieved a mass audience, their popularity often chased out the early adopters, who went looking for new virtual frontiers. But the ability for users to make money in virtual spaces like Second Life is altering this cycle to some degree. Instead of leaving, the early adopters are learning to stay and offer goods and services to the newbies. In the same way that the folks who made the most money during the California Gold Rush were the folks (Levi-Strauss, etc.) who sold supplies to the prospectors, we are seeing individuals come early to colonize a virtual frontier, then stay to make money off the later arrivals. The most profitable of these virtual prospectors are selling clothes and a wide variety of other objects on websites like SLBoutique, or developing and selling plots of land, sometimes in zoned communities, as virtual land barons. As new virtual economy platforms emerge in coming years, some of these early entrants will pick up and move to the new worlds, or extend their operations, starting the "gold rush" cycle again.

• Fear of the new, followed by acceptance, is a predictable cycle. People have historically feared that new technology and media will dumb us down and cause us to lose touch with with history, our values, and our humanity. One finds this with technologies as old as writing, which was widely feared to cause us to lose our memory, to technologies as new as virtual worlds, which have been blamed for causing us to become alienated from and inexperienced in physical reality. But new technology is always adapted to serve age old human needs. To date we have always learned to use ever more sophisticated computing and communication technologies in the reexploration and attempted solution of our ongoing human problems. Summit quote: "I'll give you my polemic here. Entertainment is education. [Medieval MMOGs and] TV sitcoms are renaissance morality plays. There's no difference."

• The Conversational Interface (CI). Exponential growth in web-based Natural Language Processing (NLP) may create a Conversational Interface circa 2015. In 1998 the average query length to AltaVista was 1.3 words, at a volume of a few million a day. In 2005 the average query length to Google was 2.6 words, at a volume of half a billion queries a day. This doubling of query length occurred in seven years. If the future growth in query length follows the classic logistic curve of codebreaking (treating NLP as a process of collaborative filtering of useful seach queries by internet users in each language), and if we are in the growth phase of the logistic, we can project 5.2 word query averages (possibly including hidden context-sensitive words in this number) by 2012, and 10.4 words by 2019. Human-to-human spoken queries average 11 words. It is reasonable to expect that when our query lengths start approaching normal human sentences in length and sophistication (after 2012 and before 2019), talking to our avatars, using simple "pidgin" emergent grammars will become the preferred way to search the web, as well as a very efficient way to further prune the lexicon. Similarly, prosody analysis (emotional intonation, rhythm, lexical stress) would lead to increasingly natura-sounding conversation. This emergence has been called a conversational interface, or CI, and can be expected to enable a plethora of new developments. One might be an increasing desire to communicate with our devices through a helpful avatar ("Digital Butler") interface, rather than to converse with a disembodied machine. Avatars would have the ability to look quizzical when they don't understand us, to motivate us with their expressions, etc. Nonverbal interaction occuring in parallel with the conversation stream makes communication more efficient in the world of human-to-human communication, and at some point this must hold for computer-to-human communication as well. Furthermore, once our avatars possess crude personality models of the users they are interfacing with, we can use them as interfaces to all our complex technology, as well as proxies when we do not wish to speak in person. But these developments seem likely to come beyond the ten year horizon of this map.

• User-generated content. As another great tool of the participatory web, business models will move, at least to some degree, from prefabricated content to user-created content in a society of increasingly creatively empowered and networked individual users. We see this in the classic social software applications today, and in the rise of collaborative content like Wikipedia and "citizen journalism." A standout example of the latter is OhMyNews which is now the fifth largest news source in South Korea, has expanded into print from online, and has developed a syndication service (like AP or Reuters). It achieved this distinction in just six years (2000-2005) by aggregating the contributions of over 40,000 individual journalists, an example of the power of the coming 3D participatory web. Google Earth as a 3D wiki, with content creators aided by SketchUp and future tools, holds a similar promise, perhaps five years from now when we all can access both proprietary and open geospatial maps on our cellphones and navigation systems. Companies like Second Life know that as user-creation increases companies get new content without significant additional cost, and players invest themselves more fully in the world, which creates mutually reinforcing benefits.

• Global computer helpers (GCH), aka "global geek brother." Online computer tech support gets cheaper and more powerful every year. Today’s tech support centers in India and elsewhere, developed large organizations like Dell, Tata Consultancy, Infosys, Wipro, etc., are becoming increasingly helpful and affordable. Commercial service organizations like Best Buy's Geek Squad are waiting to discover the efficiencies of very inexpensive and technically proficient global online tech support using educated and specialized youth in emerging nations. Such support can be organized by intelligent call routing and human resources evaluation systems, to provide feedback, accountability, and bonus pay to the best helpers based on customer satisfaction. With today's powerful computers, the ability for the GCH to take over the local user’s computer (moving the mouse, etc.) using virtualization software (the remote access features in Windows Vista, standalone programs like GotoMyPC, etc.) offers a fantastic opportunity to provide affordable, accountable expert remote assistance for every computer problem, a global "geek brother" that you can give extensive access to your computer system, using a trust network. Such helpers would have access to recent histories of our computer use, would perform all specialized housekeeping (backups to remote sites, optmization, etc.), and could give us "just in time" learning, ten minute lessons every morning on just the computer software and procedures that would be most useful for us based on the mistakes, inefficiencies, and repetitive behaviors the remote teacher has observed us recently doing. This kind of high personal transparency and continuous learning approach to our technology, while it will be avoided by many adults and corporations with low trust and high liability levels, will significantly shorten technology adoption and learning curves for those users the developed nations who choose to pay for such support. A time may be reached where those who opt for GCH use (individual users, small offices) will gain significant competitive learning and adoption advantages over the larger corporations who resist such change. If this occurs, it would be a repeat of other technology adoption cycles we have seen before, as in the advent of the personal computer in the 1980's, which was adopted later in large vs. small companies and was responsible for significantly flattening of large company hierarchies, empowering of the individual relative to the institution.

• Many children will have their first web experience through a cartoony virtual world, and such worlds will be increasingly fascinating, and the exploration they do and problems they solve will be increasingly rewarding in the most successful of these worlds. Summit quote: "Apparently when the San Jose Tech Museum installed Second Life access in their children's area, it was the first exhibit they had to put a time limit on." It is possible that children growing up with virtual worlds today will expect to access the web through them as they get older. Finding them particularly rewarding for exploration, they'll want to do serious work through these spaces as well.