Virtual Reality, Get Real by Denise Caruso Media Letter, August 1990, Vol.1, No.3 I'm sure that when Myron Krueger wrote his 1974 doctoral thesis on "artificial reality," he had no idea that he was creating a media monster for the '90s. "Virtual reality," a buzzword of the highest water -- delivered complete with its own buzz-acronym, "VR" -- has already been sucked almost dry of meaning since there is so little evidence to substantiate its many claims upon our future. The viable concepts within VR, and there are many, are masked today by technology groupies who have become extraordinary media magnets by hyping VR as either "electronic LSD" or the preferred mode of communications for the next century. The net effect is that some of the very real, excellent work that's being done in VR is becoming obscured. When you boil VR down to its elements, it's much less complex that it appears, and is developing along some parallel paths. The one with the most presence in the physical universe is photorealistic, real-time 3D graphics animation. You can call it cyberspace if you want, or virtual reality, or digitally mastered hallucinations if you must, but what we're really talking about here a simply "simulations." Not particularly romantic, but it certainly closes a few "doors of perception" along the hallowed halls of hyperbole. Parallel to work being done with 3D simulations are projects to create VR operating systems. The Cyberspace Project at Autodesk Inc. of Sausalito, headed by Randal Walser, is working on an operating system for what he terms "affordable machines," i.e., PCs, Macs and Amigas. The Human Interface Technology Laboratory (HITL), located on the University of Washington campus in Seattle, is also working on what it calls VEOS, for Virtual Environment Operating System. The VEOS project is headed by principal scientist William Bricken, most recently of Autodesk, and one of his goals is to create a software architecture that allows you to imbue objects within a virtual world with intelligence, allowing them to act autonomously and govern their own actions. HITL is doing some of the most thoughtful and thought-provoking work in VR, including projects for the severely handicapped. "The person who builds the OS for virtual reality," says Bricken, "will have the option of defining what virtual reality really is." Another path, which is much more difficult technically and progressing more slowly, is the development of human-computer interface devices like NASA Ames Research Center's famous "eyephones." The DataGlove and the DataSuit, created by VPL Research in Redwood City, CA, are connected to $250,000 workstations with drivers created by SimGraphics in South Pasadena, CA. Other devices, such as 3D stereoptic binoculars, are also being developed and tested. Such interface devices allow people to fool their eyes and other sense organs such as the skin into believing they're operating "inside" such simulations. Last, and certainly most difficult to describe, is how and why we should care about the ability to create synthetic experience at all. Brenda Laurel, an author and consultant who is widely considered an expert on the subject of human- computer interaction, thinks that the steaming piles of hype have buried what VR is actually really trying to accomplish. "Super-duper graphics are only the scene design," says Laurel. "Some people are working on the question of what do you do when you get there, but that's not as easy to flame about to the press." Laurel and Joe Bates of the Oz Lab at Carnegie-Mellon University are working on a project that maps interactive fiction to artificial intelligence technology. "We're working on how to create interesting characters and orchestrate action so that it has a dramatic shape," she says. "We're talking about manipulating action, not docking molecules." Though most of her work is directed toward what Laurel calls "location-based entertainment," she sees great possibilities for other applications as well. "Training is the biggest application, in the near term, for VR," she says. "Consider a work scenario, where a customer and the business person have to work together on something mediated by a computer. This could apply to anything from a vacation choice to repairing a machine." Laurel is part of a vast network of serious technologists probing uses for VR. But if we want to see progress continue in what's definitely a critical technology area for the 21st century, we must clear away some of the miasma surrounding VR and concentrate on applications and the people creating them. THE COMPETITIVE BENEFITS Though much of the work in virtual operating systems, headgear and heuristics is progressing nicely in the labs, it's possible today to track simulation technology by looking at the more tangible progress of 3D hardware and software. As the late Robert Noyce, founder of Intel, once said, "Ideas are a dime a dozen. The real ideas are the ones you can push through to the marketplace and sell." To that end, in 1988, the Mountain View, CA-based 3D graphics workstation company Silicon Graphics Inc. commissioned Peat Marwick to do a one-year study of its customers. The result, a report titled "The Competitive Benefits from 3D Computing," contained some interesting factoids. Peat Marwick studied Hitachi Power Group in Hitachi City, Japan, which produces nuclear power plants; Douglas Aircraft in Long Beach, CA; Karsten Manufacturing in Phoenix, which produces custom gold clubs; NASA's Ames Research Center in Mountain View; and Chrysler Motors in Detroit. It found that 3D technology "positively affected" quality, time to market, external relations, innovation and flexibility. For example, Hitachi reduced a major project's test schedule by 50 percent by using 3D workstations to increase both the number of simulations and the detail of the final model. It also used 3D models to "walk" customers through power plant models to make sure that specifications are useful to mechanics who will be operating the machinery. Chrysler forecasted a 20 percent reduction in time-to-market for new cars, partially based on the ability to link models created on SGI's Iris workstations to numerical control milling machines which carve out clay replicas automatically. More unusual uses of 3D by participants included Karsten, which used the ability to view and interact with 3D data in a patent dispute (the opposing party settled out of court); Hitachi, which created a bill of material using 3D data; and Chrysler, which interchanges actual production data with 3D models for plant loading simulations. If such benefits are possible today, it's not hard to extrapolate the future benefits of 3D as the technology continues to improve. For the most part, the software is already here, though the best of it remains on expensive specialty hardware platforms like Silicon Graphics workstations. Even given the almost scary rate that the cost of processor power is dropping, it's going to take a while. But the technology is moving to the low end quickly. One example is Indianapolis-based graphics hardware company Truevision, which just announced its Horizon 860 board, using Intel's i860 RISC processor. Bundled with the PC version of Pixar's photorealistic Renderman software, it is capable of reducing a 90-hour rendering job to four hours. And it's rumored that Silicon Graphics is readying a PC version of its Iris 3D workstation. Software companies are tooling up for a booming 3D market as well. Pixar's Renderman, the computer picture-making standard for photorealism, has migrated its software from high-end graphics workstations to the PC and Macintosh platforms. Two companies -- San Francisco-based Paracomp, the granddaddy of Macintosh 3D computing, with a new, speedier version of Swivel 3D for the Mac, and Autodesk, which just announced a 3D package that works with its Animator program on the PC -- are putting their resources behind simulations. San Francisco-based MacroMind is just now delivering developer versions of MacroMind 3D, a rendering and animation tool for 3D. Truth is, though, that almost every hardware and software vendor in the industry is working on virtual reality in one form or another. The Japanese are up to their necks in VR development, especially for entertainment. Apple hopes its work with "agents" -- i.e., representations of other people or ourselves -- will help future knowledge workers "interface" with information retrieval systems, and prove that a spatial metaphor is more valuable than a text-based one. Silicon Graphics engineers are working on virtual reality for programmers, trying to make it possible to walk around inside a dimensional representation of a program and watch it operate. Scott Fisher, who formerly headed the virtual reality effort at NASA Ames Research Center, has founded a new company to research and develop non-military applications areas for virtual reality in entertainment and the arts. So despite the hype around virtual reality, it would be a big mistake to ignore the substance behind the rhetoric and miss out on the applications that VR could open to the community at large. Just keep your bullstuff detector running at high speed, and carry on regardless.