3D Animation Workshop: Lesson 49: VRML--Back From the Brink | 2
Lesson 49 - VRML--Back From the Brink - Part 3
At Sony, Chris is working on the brave new world of digital broadcast television, and he speculates that VRML may find its true home here. On the Internet, the universal presence of VRML requires a hardware base that, in Chris' view, is two or three years away. But the new digital set-top boxes for television now being brought into production for the purpose of decoding digital broadcast signals will contain powerful processors and graphics hardware.
Digital television is a certainty, a requirement imposed by the federal government. So there's no question of its future. But digital television will permit the integration of traditional video with the kind of text display, graphics and interactivity that we already associate with computers. One of the standards now taking shape is called BHTML (Broadcast HTML). Included in this proposed standard are MPEG-4 for streaming video, Java, HTML (with some modifications) and our beloved VRML. Chris argues that the future of VRML may well be as a piece in a larger new standard. Although Platinum is, in Chris' words, "in Internet space, not TV space," the company nonetheless wants the VRML tools market, wherever it ends up.
Chris has put some serious thought into the reasons for VRML's disappointing progress thus far. As already mentioned, the VRML specification was formulated with a view toward immersive worlds on the Internet for which there is little present demand. An improved VRML would be designed to meet the specific needs of creating rich embedded content in ways that integrate well with existing Web technologies. Another problem is in the sheer size and complexity of the VRML specification. It has been very difficult to develop reliable browsers that are fully compliant with this ambitious spec.
Chris has formulated his suggestions in a proposal for a new standard influence the future development of VRML. In our talk, Chris characterized EMMA(Extensible Media Modeling Architecture) as a "thought experiment" that has provoked discussion in the VRML community. The ideas here are bold and innovative. Chris is concerned both with the download size of the browser and the size of its "footprint" in memory while running. One way to reduce both is by eliminating the built-in navigation tools implemented on the Cosmo Player dashboard. Chris notes correctly that most authors don't use these navigation tools, and that, where navigation is desired at all, it is better implemented by custom scripting.
An especially innovative concept is a method of rendering views within the scene that are not directly displayed, but rather saved as bitmaps for subsequent use. At a minimum, these renders can be subjected to image processing to produce high-quality anti-aliased results, or to create special effects. Even more impressively, these can be stacked up as a sequence and run as a flipbook texture, like a little motion picture. In effect, an EMMA-enabled browser would pictures within its own scene, and then display them on a surface.
Just when things were looking darkest, the VRML world is opening up again with new hopes and grand opportunities. And we'll continue to cover them here for you.
|To Return to Parts 1 and 2, Use Arrow Buttons||
Created: Sept. 15, 1998
Revised: Sept. 15, 1998