3D Animation Workshop: Lesson 81: From VRML to Web 3D
Lesson 81 - From VRML to Web 3D - Part 2
For all intents and purposes, the Virtual Reality Modeling Language, or VRML (typically pronounced "vermil"), has been the guiding focus of 3D development for the Internet. I don't know how many of you remember the popular vogue of "virtual reality," some five or more years ago. A best-selling book on the subject titillated readers with virtual sexual encounters using electronic body gloves and head-mounted viewing apparatus, complete with sound and even smell. But these teenage sci-fi fantasies were only teasers for meaningful uses of realtime interactive 3D graphicsÂfor in those days the phrase "virtual reality" or VR meant nothing other than realtime 3D. 3D games had not yet appeared, and the entire focus of realtime interactive 3D was on scientific and commercial visualization uses, such as training brain surgeons or molecular engineering.
Over the past few years, the VR industry, as it is still called, remains identified with scientific and high-end commercial visualization, typically making use of the most powerful workstations to crunch numbers in realtime. It is a much more mature industry than it was a few years ago, and has wasted much energy fighting off the hype and immature vision of its early evangelists. But it's critical to understand how 3D graphics for the Internet was first born against a background of VR fantasy.
Just as VR was clouded by nonsense in its early days, so was the early Web burdened with the vague intellectual romance of "cyberspace." It's not too much to say that the foundation of VRML was forged in the stellar gases of minds who, without much precision, pleasured themselves in the dream of a 3D cyberscapeÂsome kind of information universe embodied in the intuitive language of a navigable 3D space. A friend of mine says that VRML is "too geeky," and there's a lot of truth in that. Those who first envisioned and established this standard are owed our sincere thanks (as we shall see) for getting the groundwork into place. Yet these same visionaries (some of whom were really just salesmen) undermined the success of their own efforts with the strong "techie" VR coloration of their product. For most of the emerging world of 3D artists and animators, the language and notions of virtual reality were foreign and obscure.
Yet, whatever its problems, a standard was born. VRML 1 was essentially a simple adaptation of Open Inventor, a 3D scene file standard developed by Silicon Graphics, Inc.Ânowadays officially called only SGI in an effort to distance the company from its roots in graphics workstations. An Open Inventor file describes a "scene graph," which is the proper term for the description of all the models in a scene and their spatial relationships. The Open Inventor format was fleshed out with features appropriate for network use, especially over the Internet. Being able to grab an existing standard right off the shelf definitely helped get things going, but it had an enormous consequence. SGI became the dominant commercial force in VRML, to the meaningful exclusion of other players.
Using VRML required tools for developers to create the content and viewing applications for end-users to experience it. It seems that everyone misjudged how large an undertaking this would be. However, SGI was a wealthy and arrogant enterprise in those days, anxious to enter the investment-sexy space of "virtual reality for the Web." These were the days in which (believe it or not) people were still unclear whether they could charge for client-side web applications like browsers, or whether they must be given away for free as part of some larger business plan. SGI was the only company in the field that could even dream of financing the development of a VRML player (often called a VRML browser) for free distribution, and to distribute it widely enough to make it ubiquitous. For now we are touching on the main point that still resonates today. Unless a client-side 3D player has somehow found its way onto the vast majority of user platforms, no one will care to develop content for it.
|To Continue to Part 3, or Return to Part 1, Use Arrow Buttons||
Created: Dec. 7, 1999
Revised: Dec. 7, 1999