On the Edge: Statistics & Computing

INTHE FIELD of statistical computing and graphics a major challenge for researchers is to keep abreast of rapidly changing technological advances. It is important to understand new technological advances and what they may offer for statistical analysis. Being aware of technology, understanding the advantages and pitfalls of the latest innovations, allows us to influence the technological developments so that they cater to the needs of statistical analysis. Be aware that it was a statistics giant, John W. Tukey, a pioneer in exploratory data analysis and graphics, who introduced the computing terms "software"and "bit"(an abbreviationofbinarydigit)to the world. In thisarticle,1discuss the new technologyofvirtualreality(VR) and how it may be important for statistical data analysis. The technology of virtual reality dates back to 1965 when l. E. Sutherland proposed, and shortly thereafter built, the first head-mounted display, the "Sword of Darnocles,' with cathode ray tubes and a ceiling suspended system. A forcefeedback system was developed in 1971 by Frederick Brooks, and the "Data Glove," which measures finger angle, was developed in 1985 by Thomas Zimmerman. A comprehensive overview of events leading to the current array of virtual reality hardware can be found in the book Virtual Reality 11lrough the New Looking Glass, by K. Pimentel and K.Teixeira (McGraw-Hili, 1995). The term "virtual reality" has been used, and often overused, to describe an overwhelming variety of computer software and hardware applications three-dimensional (3D) movies,videogames, Web software, cybernet existence, 3D scientific visualizations, and multimedia. There is a glut of buzz words surrounding virtual reality,each suffering similarlyfrom vague and often different definitions for exam-