From: GAVAND01@ulkyvx.bitnet Subject: Cyberfundament (LONG) Date: Fri, 4 Oct 1991 17:18 EDT In article <1991Sep27.205307.13858@milton.u.washington.edu> brucec@phoebus.labs.tek.com (Bruce Cohen) writes: >First a couple general comments. Discussions of the philosophy behind the >concepts we are developing in the "Virtual Reality" paradigm are very >important; I'm glad to see postings which raise these issues. Thanks for addressing the central issues I seek to address. I appreciate that. To me, "philosophy" is the means of critically evaluating conceptual work, not spinning fine spider webs of thought to retreat into insular protection. I am most interested in technical applications, but I do not find the psychological readiness to accept premises that seem to me necessary for doing so. So, as the French existentialists say, "Engage!" >However, there is a wide range of professions, experience, education, and >vocabularies represented in the readers and posters of this newsgroup. It's >vital that all of us who post take great pains to be as clear as possible, and >eschew the jargon of our own technical specialties as much as possible, >supplying a gloss where precision requires that we use the jargon. . . . >If we can work out a common vocabulary many . . . misunderstandings should go >away. I agree we must take great pains to avoid technical jargon and (believe it or not) I try to do so by supplying, at times, gloss instead of precision. Even glossing doesn't entirely solve the problem. Communication is a difficult process. Even closely related fields have vocabulary problems. It requires tremendous patience, mutual regard, acceptance that what another says is in some way true, and painstaking effort to understand how it might be true. Quick fixes and semantic quibbling are blind alleys and ripplying eddies. I also suggest that whenever ACRONYMS are used, the full name be given. They are the worst, and appear often in postings. >In article <1991Sep27.031332.11748@milton.u.washington.edu> >GAVAND01@ulkyvx.bitnet writes: >>... >>The context here deals with fundamentals necessary for rigorous *theory* of >>virtuality, something I don't hear too many seriously discussing. >It's not clear to me what "rigorous" means in this context. In my >mathematical education, this meant that a theory was a set of statements based >on the axioms of a formal system, whose truth or falsity within that system >could be determined unambiguously. I agree that some formal theories of the >nature of virtuality vis-a-vis epistemology, semiotics, and the various >psyhcological and sociological disciplines is desirable. This electronic medium (and available attention spans of today's listeners/readers) requires brevity and brevity requires that I rely on another to seek clarification of points, just as you have done. One can never second guess everything. By theory I mean exactly what you learned in your mathematical education and as you defined above. But I am questioning the premises (axioms) themselves that prevent mathematical theory. As a teenager I was sort of a mathematical whiz so do have a developed mathematical intuition, but I found "metaframe" issues demanding my attention so do not work directly in mathematics. My context is "necessary fundamentals." As Kurt Godel's proof of his "incompleteness theorem" showed, truth or falsity has unambiguous meaning only within a given system. Recursive regression into infinity occurs if one attempts a complete, non-contradictory, general determination of truth and falsity. And history shows that it is very difficult to get a productive dialogue going once the very premises become the topic. There is a way out of that recursivity; as the poet, Wallace Stevens, wrote (quoted from memory), "The sound of a trumpet resounding/ is not another trumpet sounding." The way out is to abandon reliance on mathematical and linguistic intelligences as exclusive models for cognition and abandoning the premise that "truth value" is either simply a "1" or a "0" -- that is, the principle of the exlusive middle. I cannot begin to attempt any type of demonstration here. And I agree with you that it is desirable for some formal system (theory?) of virtuality vis-a-vis epistemology, semiotics, psychological and sociological disciplines. They can provide more satisfactory images to organize and enhance our impressions. I find it unsatisfactory, however, to stop with *descriptive* or *prescriptive* systems that they provide. They suffer the problems any axiomatic system does, but built on the logic of words alone, carry assumptions that are invisible and impossible to get at, as symbolic logic can demonstrate. >... >I agree that virtuality is not primarily sensory because I believe it is >primarily interactive; that is it exists in the same sort of space in which >human consciousness exists: at the interface between sensory and motor, >between perception and action. >>... >>Imagine total virtual space as a body of water into which we may sink certain >>vessels, and thus, be able to define individual volumes of water without, >>however, destroying the idea of a continuous mass of water enveloping all >>(this image is attributable to the sculptor, Adolf Hildebrand). The >>enveloping space may be called cyberspace; the sunken vessels constitute a >>particular virtual world. >I'm confused. If I read this right, you're implying that there is a single, >connected "cyberspace" which can be entered by anyone. I don't think this is >true. Here is an example where "supplying a gloss where precision requires that we use jargon" can also create problems communicating. Words are so inadequate. >As I use the term, cyberspace is an artifact of the interaction between a >human and a world-description (a "VR database" in technical terms) via the VR >interface. So there may be many cyberspaces, and they may be public or >private. I generally accept your use of the term cyberspace. Technically, however, cyberspace is not an "artifact." It is a state (dynamic view) or a condition (static view) of an interaction; to reify it will have devestating conceptual consequences. The Eastern philosophical concept of the "No thing" universe is totally appropriate and necessary here. Cyberspace is the Void; there are not many Voids. There are several "attractor states" in the human/VR database interaction. That is, a given experience in it depends on the choices made, and chaos theory informs us how many of the choices, or which ones, will "flesh out" into a world-description. In this sense, there are several potential public or private experiences, but only one actualized experience (which encompasses both the private and public aspects) at a time. The actual "fills" the participant experience. The objects encountered because of the choices made are what I termed "vessels" in a virtual reality. Given a single VR database of any complexity, the participant can encounter different events each time and several unique experiences. What I term "cyberspace" in contradistinction to a virtual reality (an experience turned through interpretation to a world-description) is the interface which includes many public and private potential experiences, only a few of which may be actualized. Cyberspace is the potential field of human/VR database interactions (mathematically, a manifold). Out of the one, many. >On the other hand, if you are saying that there is a conceptual space of >virtual world designs, in which each world is represented by a point, then I >agree, but I don't see why this is important.. >From my last statement, it should be evident that, yes, I am saying there is a conceptual space of VR designs, mkore specifically, I would say of VR experiences. Even in one VR design of any complexity, several experiences may occur. And as a "metaframe" issue: To say that each world is represented by a point is actually translating a mathematical or conceptual intuition into visual intuition to lodge that conceptual space into a visual/spatial space. Scientific visualization demonstrates how useful, even necessary, that is. However, your compressing each world-description to a point is forcing them into a 3-D visual box. Your visual intuition employs a spatial frame modeled after the nature of objects in the middle ground of human focal vision. I confess that once you do that, it is difficult to see vitally important issues in the conceptual space we are discussing. 3-D representations of 4-D lose something (temporal dimension) during compression just as 3-D loses volume in 2-D. Although techniques have been developed to represent volume in 2-D, volume loses its "palpableness" and choice of perspective. Your technique of compressing what is essentially a 4-D event into 3-D loses the temporal element entirely. There are better ways (which 19th and 20th century thinkers tend to dismiss) for translating 4-D into 3-D where the time dimension is not entirely eclipsed. 2-D Byzantine art represented 3-D events in a way that totally ecliipsedvolume. The introduction of a virtual vanishing point created a more "realistic" model. Eastern (particularly Indian and Tibetan) dialectical systems don't force conceptualization into such limited logical boxes so there is a great deal of precedent for what I am saying. >>... >>Unless you subscribe to some sort of "ether" theory, action must be seen to >>construct the functional forms of imagination and are primary over the >>geometries among its vessels. This being true, stationary models for space >>or categories of substance and quality are inadequate for exploiting the >>potentials of cyberspace. >I really don't understand what you're saying here. Please explain. My response immediately above this one also pertains to an explanation. My 4-D/3-D/2-D analogies were meant to convey the idea that cyberspace is not an artifact, in no manner substantive or capable of being given a stationary frame. Below I will briefly give a technical then "gloss" explanations. (technical): Cyberspace is one relational complex that precludes metrical geometries. It is not a fact prior to a world; it is the first determination of real potentiality (a plenum) arising out of the interface between the sen- sory and motor. It is merely the potentiality for division; human action effects this division into a geometry by spatializing and temporalizing a gen- eral scheme of relationships that melds objects into the real unity of one experience. By melding this unity, humans project a geometric locus to the "blooming, buzzing, chaos" of experiences. This melding is the foundation for conceptual abstraction, object permanency, etc.; it creates a world-description. That is we make representations. Humans live primarily in such conceptual spaces, not the physical, actual spaces of each moment. Usually we are quite inattentive to the present moment. Cyberspace is the capacity for giving spatial and temporal extension. And perceptual extension, often called "projection." is a continuum through many partial (fractal) dimensions. Humans are not perceptually delimited to 3-D spatial extensions. It is also to bear in mind that there are different kinds lf "space." Actual space (the presentationally immediate moment too filled with potentialities for us to take it all in; it is undifferentiated space within a moment), various metric spaces, and conceptual space (which is purely abstraction). Now let me gloss what I just said in the above paragraph. For nearly 2000 years Euclidian geometry was the only exact (measurable) visual-spatial description of extension. During the 1800s, non-Euclidian geometries developed. By changing a Euclidian postulate (e.g. Euclid's fifth, to postulate that not one, but either zero or infinite lines can go through a point off to the side of a straight line and be parallel to that first line) the visual-spatial description is radically discontinuous with the Euclidian. Mathematically, the non-Euclidian geometries are just as sound, but they were inconsistent with common sense, with earthbound experience. Such geometries did not seem to have physical meaning (modern examples of the same problem can be given). Later the non-Euclidian geometries lost their opaqueness to visual intuition. We realized they describe the geometries of spherical and pseudospherical, rather than planar, surfaces. One is not more correct than another. Mandelbrot's concept of perceptual "cutoffs" is pertinent to the issue because the appropriate geometry may depend on the scale of perception. From a bird's eye point of view of Earth, Riemannian geometry would seem appropriate; from our ant's eye point of view on the Earth's surface, Euclidian geometry usually works. Fractal geometry overcomes point of view problems by incorporating scale into the geometry itself. The point of these discussions is that it would be inappropriate to say that the conceptual space containing all geometrical systems is several spaces. That space IS cyberspace, which comes out of the interface between sensory and motor, perception and action. As a conceptual space, it is purely an abstraction made for the convnience of talking about it; as a potential state of sensory and motor interaction, it is NO THING until actualized. Each geometry that has been developed is a *systematic* measure of a virtual extension that has melded an entire world-description. Cyberspace cannot be conceived as an artifact or any type of substance, or even a quality. Cyberspace cannot be conceived as a model for space. From out of the plenum comes several models for space. Because of interaction, something comes out of nothing. Interaction is motion, and motion is potential temporal extension (if nothing ever changed, there would be no time). Another gloss: just as a motion, say the wave of a hand, cannot be substantively captured, neither can cyberspace; (1) graphing the hand's trajectory, (2) writing a program for representing the motion in a virtual world, or (3) freeze framing the hand photographically may abstract important aspects of its motions and inform us of a pattern (1), give the potential for (2), or suggest (3) actual motion. But it is important to remember none of the above are the actual motion; they merely model potentialities. Cyberspace is the interACTION, and specific spatial and temporal extensions, specific world-descriptions, are only representations. We're talking Koyre's, "The map is not the territory," but I chose motion because perception of motion is not a 3-D perception. Spatial perceptions may lodge motion into a frame, but spatial perception is not the same process as motion detection. The brain melds them into the locus of its own universe. All dimension is conceptual. >> >>Even though the participant is attentive to sensory impingements, if he or >>she has any choices at all, the construction of an experience (the >>participant's "reality") is not atomististically give; ... >Here, and in several places after this, you use the term "atomistic" or >"atomizing". Please explain this usage. In general the prose of your posting >starts to get a little dense here, and I got lost in what I guess to be some >technical terminology related to ontology. Sorry, it's been quite a while >since I studied this; I expect many others are in the same state. And I am sorry that I got too dense. I may be falling into the same hole with this posting, but I have spent considerable effort to communicate highly abstract issues in a manner as accessible as possible (this is the the third draft of this before posting my response). Please bear with me, or tell me to go away! I do not do this simply to mentally ejaculate. By the term "atomizing" I refer to the mode of perception by which we "meld" a unitary world-description from the "blooming, buzzing, chaos" (William James, by the way) of experience. This is done through discriminations of differences of sense dta, including colors, sounds, bodily feelings, tasts, smells, together with the momentary perspectives introduced by the extensive relationships in actual space. By lodging experience in a locus, we discern discrete boundaries and project object permanency in a geoetric universe. Representations (atoms) are the furniture of our minds. We construct them, they are not simply given. Two people in the "same" environment can construct diametrically opposed world-descriptions, as the legal system knows too well. The Western tradition has largely been constructed based on assumptions concerning atom-like representations of material or discrete units, the building blocks of our thinking. All this is fine and good; I just add that it is only half the picture. Some phenomena, such as much of the biological, social, cognitive, and even such things as robotic vision, cannot be easily systematized using the axiom of some such fundamental "atom" simply given to experience (VR databases seem to be constructed on that premise). The perceptual interface is constructive action; discrete units aren't simply objectively "given." Actually, even the physicist's "atom" doesn't really fit into such simplicity; even they require complementarity. VR development needs a conceptual unit that is complements the discrete unit. That conceptual unit has been rigorously developed by Susanne K. Langer, which she calls the "act." But that is another story . . . . >>... >>Constructing a 3-D model of our experience, for example, emerges by acts of >>being attentive (out of all that which is presentationally immediate in the >>perceptual field) to slowly-moving or stationary, molar-scale tensions and >>contrasts perceived as external to self and in the middle ground of focal >>vision. >I agree. I think there's a very important point here that needs to be >emphasized: an understanding of VR needs to be based on an understanding of >how humans develop their models of the external world. There's a world (pun >intended) of philosophy and cognitive psychology which bears on this question, >and we ignore it at our peril. Clearly, from my last statements, I am in full accord with the importance you emphasize. The heart of my postings has to do with those perils and with productive directions that may be taken. I shout into the wind. And we need the next step beyond understanding: procedures for turning that understanding into technology. That has been my research focus for over twenty years, but I will not talk about procedures on this newsgroup for proprietary reasons and for ethical concerns. >>... >>We live in an era where the 3-D model is misapplied as if it were a template >>to all perceptual dynamics, and thus to all experience. Conceiving cyberspace >>issues primarily in terms of three dimensions springs from a "lethargy of >>custom." It indicates a lack of full appreciation for other constructive, >>assumptive patterns that are radically discontinuous with the 3-D model, but >>common in our perceptual experience. There are many common situations in >>which we see various fractional dimensions. Cyberspace is really an extensive >>continuum underlying all dimension and needs to be approached as such. >I'm really vague on what other models you think are applicable; that is, in >what ways the 3D model is insufficient as a paradigm. Examples would be >welcome. All the above statements, hopefully, shed some light on how the 3D model is insufficient as a paradigm. Most generally, the 3-D model is insufficient because it makes invisible perhaps the greatest perceptual and conceptual challenges we face in this revolutionary period of change. That involves what Mandelbrot calls perceptual "cutoffs," the process of perceptually moving toward the infinites of the large and the small. Our perception of dimension is not inherent in the perceived object, but depends on the scale at which we perceive the object. The 3-D model makes it very difficult, if not impossible, to break our thinking from the strict subject/object, mind/body duality. And the tradition of delimiting our conceptual models to 3-D scale ignores a great deal of our actual perceptions. Examples include rhythmic changes, such as of the seasons, of modern dance, or of music; a 10cm ball seen from from away; dynamic relations among 3-D objects; holistic attentional strategies; and transformations through time, such as phases of embryonic development. 3-D conceptual models either make such things invisible or difficult to systematize, because of indistinct, shifting, or overlapping boundaries, and other problems. The problem of developing VR to greater perceptual realism is a larger issue than delivering better rendered objects. It must also conceptually break out of the narrow perceptual domain of modeling. For a long time, we (and by we, I mean the professionally educated) got away with modeling the locus of our universes after stationary or slow-moving objects in the middle ground of focal vision, but such a narrow realism no longer predominates in our perceptual environment. Today we are bombarded with perceptual information of diverse scales. Watch one TV commercial break just for changes in visual scale and in the rate that objects shift in the perceptual experience to realized what I mean. Adults today complain of today's fractionating of experience and consider it psychologically disastrous. There is even evidence that it is causing changes in the physcial structures of a child's developing brain. Adults have even created a childhood epidemic: Attentional Deficit Disorder (ADD). Multimedia researchers often recognize that we live in a transformed, and transforming, perceptual environment. They are developing more adequate conceptual models. VR researchers don't seem to be as rigorous. For all of us, it is easier to go into the future looking through rear-view mirrors. Another insufficiency of 3-D models for VR comes from the fact that the environment substitutes for the human's use of focal vision. So many VR people seem concerned about giving participants the freedom to choose which 3-D objects they interact with; that that freedom overcomes the "fascism" of other media. But enslaving the participant's attentional strategy may have worse consequences than the media that restrict choice of interaction. A report in the August 1991 _JOURNAL OF NEUROSCIENCE_ by Corbetta et al. demonstrates that perceptual judgments that discriminate shape, color, and speed of a visual stimulus activates different cortical regions outside the visual system, depending upon the attentional strategy. Selective attention and attention divided by attending to multiple features simultaneoulsly show in PET measurements different patterns of activation involving different brain regions. Their study did not include perception of dimensional scale, so we can only extrapolate from that study, but it does flag the issue. And it is not clear how these experimental results involve conceptual strategies of analytical versus holistic, or systems, thinking. Personal experience and other neurophysiological and sensory psychology studies, not to be enumerated here, suggest that enslaving the attentional strategy in VR might actually affect brain functioning. It also makes invisible some of the most interesting technological applications that can be developed. >>... >>Thinking is framed with mostly unconscious spatial assumptions that have no >>more validity than the fact they they are part of an historical and cultural >>episode. >Well, they also get some validity from the fact that our spatial assumptions >are grounded in the evolution of our sensor-motor systems, and the processing >which goes on between perception and action, within an environment which >penalizes bad modelling. So there's some relation to an external physical >system beyond what we've invented in recent history. Yes, there certainly is a relation to an external physical system, but the internal/external dichotomy of our thinking caps is the problem, which I have been discussing through this posting. There is a wide belief, even among some neurophysiologists, that 3-D is "hardwired" into our visual system. I can see no reason for that assumption. 3-D can certainly be explained in terms of our sensory-motor systems (stereoscopic vision, persistence of vision, etc.), but it can just as easily be demonstrated that *preoccupation* with 3-D was gradual cultural development culminating in what seems to be a patently unscientific attitude about dimensional perception. Our mathematical and conceptual intuitions can deal with multiple dimensions, and so can our visual-spatial intuition. And I must seriously question the idea that the "environment penalizes bad modelling." Given the incredible variety of mythic, magical, religious, pseudoscientific, bad-science, and immature-science, and other phantasmegorical models that proliferate, I can only wonder where the penalties lie. >>... >>Today's computer-generation toddlers and the multimedia field will blow the >>preference for a 3-D box into the oblivion of irrelevance more quickly than VR >>developers can get them into the consumer market anyway! >Maybe. What do you think they'll want instead? In comments made above, I have given some of the basis for my belief. I think that the 21st century will be preoccuppied primarily with socially integrating 4-D to the point that the "man in the street" will come to find it so obvious that they consider 4-D the "natural order of things" as 3-D is today. Today's ethical crises will never be resolved without such an integraton. Which means a holistic integration of time and space as a constructive process of reality systems, where there is neither subject nor object, but a subject-object unit. I don't know how long it will take before 5-D becomes the preoccupation of science fiction and futurists and scientists the way that 4-D is now, but I doubt it will take as long as it has to go from 3-D to 4-D socialization. Gary Van Den Heuvel Univeristy of Louisville Gavand01@Ulkyvx.bitnet