• The Carpal PC and Forearm PC, one of the next frontiers in wearable computing. These popular future wearable systems might consist of the following key components: 1) a very lightweight OLED touchscreen display that covers the dorsal metacarpals (back of the hand) carpals (back of the wrist), and a few inches of the distal forearm, in the same configuration as the top half of a carpal tunnel hand brace (picture left). The other side of the hand will be bare wrist and forearm, for maximum comfort. The Forearm PC, an alternative design, again wraps around just more than half of user's forearm (a slimmer and lighter version of Eurotech's Wrist-Worn PC, picture right). Both versions can be put on or removed quickly using a slap-bracelet mechanism, allowing easy use on the arm, at a table or in the lap. Advanced versions of these devices would interface wirelessly with additional wearable items, such as: 2) additional processors and batteries integrated in a standard-sized belt, worn around the user's waist, 3) a full-sized, touch-typeable keyboard and processor worn at the waist as a 1.75"x5" sized belt buckle (four interleaved pieces of keyboard). A touch of the buckle would de-interleave it into a standard 3.5"x10" laptop keyboard, allowing 50-80 wpm typing at the waist. Also, 4) a wearable mini-mouse, worn on the belt on the opposite side of the body from the Carpal/Forearm PC display would complete the system for those users who wanted an interface identical to their desktop ergonomic, requiring no retraining. Simply looking down anytime at the back of their hand, while typing or mousing would provide useful visual augmented reality information. In high end versions, the Carpal and Forearm PC's displays could fold out to twice their size for more visually intensive tasks. As miniaturization improves these systems will contain a cell phone, GPS, digital music player, camera, etc. The system could be connected through the cell phone to mainframe-based high-accuracy voice-recognition systems, to maximize the quality and range of verbal commands to the wearable PC. The keyboard will be used wherever verbal commands are insufficient or are less optimal, as they will be for many years to come, for some tasks. High end versions might be waterproof, and inductively charge themselves wirelessly when you are near a keyboard or are seated in your car. They will also contain fast-charging nanobatteries. An obvious accessory would be a detachable bluetooth cell phone earpiece that fits into the Carpal PC at the forearm or at the waist when not being worn, and which recharges off the Carpal PC's battery when not in use. Wherever you are, when you look down at your forearm you'll see a local geospatial map, have access to locally relevant video, audio, and text streams, and be able to query them orally. Verbal commands and keyboard shortcuts will give you access to both automated AI systems, like today's Free 411, and to human beings in various service centers you are subscribed to. All kinds of specific data on the objects in your vicinity will be immediately available, including histories, recommendations, annotations, etc. As the participatory web grows, no matter where you find yourself you'll be encourged to add your own feedback to the global database, at point of experience.

• Wristwatch cellphones. The Dick Tracy "two-way wristwatch radio" was first conceived in 1946. The wristwatch cell phone is an obvious developmental attractor for many of us. Why hunt for your cell phone when you can wear it on your wrist? Would you even take it off to shower if it was small and waterproof? It could even charge inductively when your hands are near a keyboard, as many of ours are every day. You'd want the ablity to swap in charged batteries and full SD chips as well.

• Video Walls in 2016. Nielsen's law (see Constants - Sci-Tech, Nielsen's law) has charted a doubling of premium-access internet bandwidth every 21 months since 1983. This trend predicts that fiber-to-the-home initiatives, like today's Verizon FiOS (20/5 Mbps) will deliver over 300 megabits per second of wired bandwidth to typical premium users in 2016, allowing download of a 2 hour video in 2-3 minutes (impulse buy/on-demand). As today's mature (LCD) and newer (DLP, plasma, OLED) display technologies continue to drop in price and energy usage, we can forsee home and business users converting large portions of free walls in several rooms into video wall configurations. Video walls would allow parallel video multicasts (internet video, HD video, 3D worlds, etc.) to various areas of the wall. Such systems would also greatly improve multi-party videoconferencing, by delivering "spatial video" (see 8Bj) with cameras on consenting users as they wander through their rooms, delivering you their image to your nearest video wall, wherever you are in your own home/office. As with current internet browsers, users would likely have only a few sections of these walls displaying full motion and sound at any time, with the remaining displaying static images, slowly updating snapshots of other channels, menus, etc. These secondary images might be in still motion until the user "tunes" to them by pointing a remote or saying a verbal command, whereupon they would enlarge, animate, and move to center screen. As a video wall controller and wireless mobile accessory, a very light tablet remote, capable of displaying thumbnail pictures of your 100 favorite internet TV channels and virtual worlds on the top screen, and your next 100 only a tap away, etc., would for many be a very intuitive way to manage a large number of content specific 3D feeds. Collaborative filtering and multiple categorization systems could be used to manage this information, with thousands of ancillary channels and 3D experiences "competing" to migrate up to higher levels of the user interface, based on user feedback and public preference data. Once the internet's bandwidth can support them, video walls and tablet remotes seem likely convergence devices for many future users, as they would maximize the value of peripheral attention while minimizing distraction, and allow the intelligent filtering of vast quantities of video data. Their entertainment, educational, and collaboration value as tools in the participatory web would be immense, though issues of addiction, cocooning, narrowing of views, and other abuses will also be serious concerns, particularly with early versions of these systems.

• Networked shoes in personal area networks for virtual worlds navigation. Networked shoes may become a valuable interface for immersive virtual worlds. Feet pedals are already used in dictation systems and racing simulations. For virtual character steering and walking, minor shifts of your feet on the floor to change direction, slight tapping of the toes to move forward, tapping the heels to jump, etc. would allow a whole range of natural repertoires, and leaves your hands for hand activities. Kids would love peripherals like this, as they give them yet another excuse for kinesthetic activity. As personal area networking expands, shoes also represent a unique place to put electronics for wearable computing, so the first versions of multifunction shoes (sensing, recording, computing) might be emerge within the next 10 years.

• Early conversational avatars in 2016. Anthropologist Ray L. Birdwhistell, a pioneer of "kinesics" (nonverbal communication), estimated (Kinesics and Context, 1970) that about 65% of the information in a conversational message is conveyed by facial expressions and body language, and only 35% by spoken words. Once our computers are smart enough to speak to us in simple sentences via a primitive conversational interface (circa 2015-2025), to store primitive inferential, context-based personality and values models of their users (first generation personality capture), and to express those models using a facial action coding language, we predict avatar-mediated human-computer and human-human communication must emerge. For conversations that benefit from even crude kinsesics, avatars should be both more humanizing and more efficient than other communication modes. We forecast tomorrow's users will increasingly associate personal avatars (what we call a "digital twin" (DT) with their public personas, and as DT's grow in capacity, use them to mediate contact with the larger world in an attention-limited economy. Avatars-as-representatives, both of humans and of complex technological systems (cars, houses, computers, robots, tools) will become increasingly friendly, personalized, intelligent, and interactive relative to text, static images or linear narratives. Public data mining of user lifelogs will greatly improve context-sensitive DT values and personality modeling, and greatly enhance social networking, personalized education and online collaboration. Given trends in automated knowledge discovery and natural language processing, the DT's of early adopters and research users should be able to have very primitive yet useful natural conversation with inquiring humans by 2015, with performance improving significantly over the subsequent decade. This conversation will grow to include simple information about the user's background, interests, present location, availability status, and future plans, as well as the ability to schedule meetings with trusted parties, answer FAQs, manage e-commerce, and perform other simple transactions. Eventually, DT's seem likely to become the first-pass communication screeners ("trusted handshakes") to prequalify face-to-face interactions between individuals, an important element of future trust networks. See IMVU for a good example of today's avatar-mediated chat communities.

• Head mounted displays (HMDs) are likely to remain niche applications for virtual reality. Other than for specialty competitive gaming, the marginal benefit to HMDs and Spatially Immersive Displays (SIDs) beyond the standard Keyboard Mouse Monitor (KMM) interface do not seem compelling. There are severe drawbacks to using HMDs while navigating physical space around the home (eating food, interacting with friends in the same room, etc.). SIDs, by contrast, seem likely to gain modest adoption as prices drop in coming years, because they will be natural outgrowths of HD home theatre installations (eg., more screens going on more walls, either projection or wall mounted). A low cost SID that covers three walls of the living room seems an archetypal advance for immersive gaming and virtual worlds, one that might achieve minority market adoption by 2016.

• Immersive virtual world walking and running interfaces seem likely to remain only military and research devices over the next ten years. A notable exception is two degree of freedom treadmills (speed and elevation tilt) which might be combined with large display screens and virtual competitors and coaches (via videoconference) in a system that could emerge a small fraction of high-end home, corporate, and commercial gyms, and military and institutional training environments as part of the growing exergame market. But higher DOF and "unrestricted" walking systems like the VirtuSphere currently have such drawbacks as high complexity, cost, noise level, space requirement, user injury, and the promotion of unnatural walking behavior. Even such clever near-omni-directional treadmills as those of Virtual Space Devices (see video) are presently $1M systems, hoping to develop $30K systems in coming years. R&D programs like the European Commission's Cyberwalk Project are seeking to develop systems for "natural unrestricted 3D motion" in virtual space, but that long-dreamed-for objective continues to look out of reach beyond the lab, with all technologies presently on the horizon.

• Sensory substitution technology in augmented reality seems unlikely to have a significant social effect for the forseeable future. Sensory substitution is a rudimentary interface that has been touted by some futurists as an additional channel to provide augmented reality sense data to the human. There are research devices available that use a wearable camera and a transducer on a subject's tongue, palm, or back of the torso, allowing the brain of unsighted individuals to interpret the touch patterns as a crude form of vision. Some have suggested that such tools are ways we will be able to increasingly augment our normal senses through unobtrusive secondary channels in coming years. Imagine, for example, permanently gaining the ability to see in 3D behind you, with a wearable camera and using areas on your back as retinas, and having a part of your brain learn to adapt itself from birth to interpreting such information. Or using that same subliminal system to be able to continuously and surreptitiously "read" incoming text and graphics from a 3D augmented reality display linked to the web, while your eyes are enaged in something else. The problem with this vision is that this interface strategy appears to be very limited and poor by compared to our existing forms of sensing, and is subject to all the same limitations of divided attention that we find today when humans try to do more than one thing at a time (e.g., drive a car and talk on the cell phone). Developmental biology and animal studies also argue it would work better in human youth than in adults, particularly if used from birth with no interruption (e.g., as some form of implant). But such use would raise serious ethical concerns, require extensive testing, and be very slow to emerge. Even when perfected it would remain subject to sensory substitution's decreased effectiveness relative to the naturally evolved senses. An alternative to trying to open new sensory channels to the brain would be a form of "sensory specialization," to permanently assign a part of your existing visual field to your AR display, as with glasses or implants in early adulthood. But again, while such tools mayl be extensively experimented with in future subcultures, they all seem likely to give only minor benefit, and would come with significant problems. In the few cases where early benefit might accrue, as with soccer players that have the ability to "see through their backs," including a topsight view of the game, relayed to them by remote camera, the social stigma involved with early and elitist use would be significant. Such players would for a very long time be relegated to playing in minor, "enhanced" leagues, the way professional bodybuilding now has minor competitions for those who admit to the use of performance-enhancing drugs. For the forseeable future, simply increasing the miniaturization, affordabilty, performance, and bandwidth of our wearable computing systems, as well as the quality of the artificial and human intelligence connected to them, seems to be a far more productive and socially acceptable prescription for progress in augmented reality.

• Avatar-based videophones and the limits of videophone utility. Within the next 2-5 years we'll see high-end avatars that can accurately mimic your facial expressions as you sit in front of the screen. A low-level version of this has already occurred, with Logitech's latest Orbit MP QuickCam webcam, which adds mixed reality features to webcam images and supports gesture-driven avatars. Something like this might make the videophone usable for folks who don’t want their entire personal environment accessible to the caller. Still, even these kinds of videophones will likely be used only very selectively. They make sense for family, dating, early stages of business contacts, and some forms of collaboration, but audio will continue to be more efficient for most conversations and allows more privacy and multitasking for the users. In a curious way, less is often more in communication. Video often shatters the intimacy of an audio conversation.

• Location-Based Cellular Radio. LBCR will be a multibillion dollar market for autos and mobile devices by 2016. Local wireless infrastructure will support streaming internet audio in the car and over the cell phone within two to three years (2008-9) for premium customers, and four to six years for the mass market. This may be one of the biggest coming new media industries currently below the pundit's radar. Tomorrow's GPS-equipped cell phones and handheld and in-car navigation systems will support the playing of short location-based images and video, and location-based audio delivered over cellular radio (and HD radio to a significantly more limited extent). Combine this with basic 3D virtual maps and you have a handheld information and entertainment device with compelling new abilities. Motorola's iRadio is a leading company in this still very early space, though there are a number of other entrants [60]. Cooper's law tells us that the spectrum efficiency of radio communication (both voice and data) has doubled every two and a half years, over 104 years, since radio waves were first used for communication. As a result, we are on track for this functionality being possible at the premium end very shortly. In late 2007, Qualcomm will release CDMA EV-DO Revision B data modem and cellphone chips to card makers, which will provide up to 14.7 megabits per second peak on downlink (something closer to 1/4 of this for real world average rates). When these emerge in 2008 they will enable such mobile features as television, and internet browsing while making VoIP calls, and the enough extra bandwidth to support location-based streaming radio in the car [26]. Imagine being able to receive cellular internet radio, as you drive, that offers you such channels as: 1) Educational and historical information for the landmarks you are passing, 2) Highly local, up to the minute news, politics, weather, and traffic, 3) Reviews and info on local restaurants, shopping specials, and entertainment events, as you are passing them. On the navigation and cell phone screen, location-based advertisements can pop up as clickable graphics, or audio headlines in a radio stream, and if you click on them you can hear ordinary or "breaking ads," i.e., the big special going on right now in the restaurant you are just passing, for the next 20 customers that wallk in the door. Over time, store owners will be able to update their ads in realtime from web interfaces, using their own automated and manual systems ("Only two of these left in stock, hurry!"). Such ads are likely to create commercial and civic institution-driven flash mobs increasingly in coming years. When you venture outside of your home zone, you can set your filters to let you know interesting tourist information, or when you pass a favorite type of restaurant, bookstore, coffeeshop, etc. For the radio shows, most of this programming can be auto-assembled digitally, without need of a radio DJ. Much of the local video, as well as the less changeable audio, might be updated daily by Wi-Max to the auto while it is parked, to be served up later as one drives through the local area. In a more advertising-unobtrusive mode, cell phone users and auto passengers will be able to click on visual ads appearing on their device's navigation screen, as you approach a location, which will trigger audio and video location-based advertising, entertainment, and educational content. Such ads will of course be able to hand off GPS coordinates of the event back to the device, which can then provide turn-by-turn directions to the driver. Such a platform will make weekend and evening "cruising" of ones favorite areas of the city a much more enjoyable experience for both tourists and locals. One of the obvious longer-run implications here is that within a decade, satellite radio's national-level programming (XM, Sirius, etc.) will be on the path to becoming a niche player role in the radiosphere, the way satellite television is beginning to become a niche provider in the television space. Another is that regulations will have to be drafted for the use of ads and video on navigation screens while the car is in motion, so that driver-initiated accidents don't increase. Handheld GPS devices, like Garmin, who have 50% of the U.S. handheld GPS market, may be first with this service. But telcos deploying smart cell phones are likely to be the major provider, as with Sprint Nextel's partnership with TeleNav GPS Navigation Systems to provide turn-by-turn GPS navigation for $10/month [65]. As a final piece of this prediction, once the telcos have built out the complex infrastructure for superior entertainment and news content based on voice subscriptions, and have trained up millions in how to use their cell phones and in-car navigation/cell radio systems as mobile entertainment platforms, the market will be ready to support open source versions of this geographic web. Or perhaps the open source versions will be competitive from the very beginning. These are interesting and uncertain issues for the future.

• Internet 3D content feeds to the home. The coming decade will bring an influx of both independent film and machinima (video animation rendered with virtual world engines) to the home. Leading platforms for amateur and independent video and machinima publishing, like YouTube and Google Video are poised to greatly democratize access to 3D visual stories, as soon as these can be autosubcribed by our home media systems. In 2005 Tivo announced a deal with Yahoo! TV, allowing users of the latter service to program their Tivo's through Yahoo's interface. But what is needed to open up this market would be allowing PVR users to automatically download internet video via RSS feeds, giving them keyword driven, ad-minimized content to view at home each evening, and an effectively infinite number of independent TV channels. Perhaps a future iTunes hack will allow this. Windows Media Center plug ins like Streamalicious allow you to subscribe to YouTube, Google Video, and other feeds with your computer, so it can't be long before we see this in our TVs as well. When videocasting equivalents to OhMyNews (Korea’s citizen-journalist newspaper) begin to pay the best amateur content providers for the right to display their work, an explosion of amateur video content will occur. As authoring tools for producing machinima and recording public events within 3D worlds improve, and as leading virtual worlds make these tools available to their residents, we will see more 3D animation downloaded to the home as well. The increasing popularity of all-animation movies makes it clear that discrimination between 3D and film is blurring. "Good stories not yet told" are becoming the primary currency, and niche audiences are increasingly accessible in the networked world. Netflix, for example, pays premiums to independent filmmakers for good social documentaries to add to their 65,000 titles, and Netflix customers rent as many as 40,000 of these titles each week, diving deeply into the long tail of content diversity.