From: vanevery@blarg.net (Brandon J. Van Every) Subject: Re: DESIGN: Is there really such a thing as text-based VR? Date: 24 Feb 1996 00:52:48 GMT Message-ID: <4glnh0$a6d@guysmiley.blarg.net> Organization: Blarg! Online Services 206/441-9109 From: vanevery@blarg.net (Brandon J. Van Every) Robin Hollands (R.Hollands@sheffield.ac.uk) wrote: [Robin asserts a common definition of VR:] : 1) Representing a 3-dimensional environment : 2) Must be updated in real-time : 3) Must be interactive (even if the interaction is limited to purely : moving and looking around) : I do feel this is rather a generous definition, since this includes : telepresence, augmented reality, and 'text-based' VR. Perhaps a better : crieria would be to extend 1) to read 'Representing a synthetic, : continuous 3-dimensional environment'. This change would have two : benefits. Firstly, the inclusion of the word 'synthetic' would remove : telepresence systems, which aren't virtual reality because they're : real and not virtual (but would still allow augmented reality since : its VR superimposed on the real world). Secondly the word 'continuous' : would remove those stupid Quicktime-type systems which just pan : through large bit maps instead of rendering 3D geometry, and would : also remove my pet-hate - text-based VR. : No let's be serious here, no-one in their right frame of mind could : call MUDs VR, unless they also include role-playing or simply : day-dreaming! The user's experience of a MUD environment is purely : through the medium of the MUD creator. You go to a location and look, : and you get the creator's description of the local area, no ifs ands : or buts. I think you were doing better when you were trying to separate the notion of continuous geometry representations, from discrete samples of textual descriptions. I'm not sure this is a valid separation, as geometry representations are only discrete approximations to continuous representations anyways. Yet the _degree_ of continuity is certainly much higher, at least when you look at typical MUD implementations. At any rate, I think continuity vs. discontinuity is a stronger line of investigation to pursue. You aren't going to make much headway if you try to classify things in terms of "the author gives you an absolute representation," vs. "the viewer subjectively intereprets the representation." The viewer _always_ subjectively interprets the representation. It's just that in the 3d realm, it's done with a spatial language. We are often unconscious about the spatial langauges we employ, because non-specialists typically don't spend much time talking about them. However, it seems fairly apparent to an artist, an anthropologist, or a 3d interface designer. Let's say you run into the Monolith from 2001. That's a "big, imposing" piece of geometry. It has a subjective interpretation, and is linked to many other language constructs. Even if you, as the author, just want people to look at your rectangular parallelipiped as an uninteresting example of mathematical form, there are going to be people out there who point at it and say "That's the Monolith!" And then they're going to start beating on it and kicking it, just like the chimpanzees in the movie. The chimps were subjectively interpreting and interacting with an abstract form. Is it a threat? Is it a tool? Is it a God? Is it a parallelipiped? What a priori basis do we have for knowing any of this? To get back to text, I think if you made a sufficiently rigorous textual description of an environment, then it might be more likely to hold up as "real." We usually think about MUDs as fantasy adventures or works of interactive fiction, and the rate at which they sample the environment is rather low. What if instead you used a MUD to describe the entire inventory and personnel of an office building? And you described everything in the office in excruciating detail, so that any piece of info that somone could want was accessible? Then the MUD would be modelling a process that exists in real life. The MUD representation would be a virtual representation. This gets into the realm of whether you prioritize visual input, auditory input, haptic input, or higher level language constructs. It is not so simple to say "we don't do language." Most of the useful applications of VR simulation are about imposing language constructs upon various phenomena, in order to aid understanding. Cheers, -- Brandon J. Van Every | I am looking for work in the Seattle area. | Check out Free3d, my 100% efficient, 100% portable 3d Computer Graphics | 3d lib, at . C++ UNIX X11 Motif | E-mail: vanevery@blarg.net