Lucid Mapping and Codex Transformissions in the Z-Buffer
Copyright © 1997-98 Matthew G.
Kirschenbaum. All rights reserved.
A version of this work is also available on CD-ROM in The Little
Magazine Volume 22 ("Gravitational Intrigue: An Anthology of Emergent Hypermedia"). Copies of the Little Magazine may be ordered from the link above.
A companion essay entitled "Lucid Mapping: Information Landscaping and Three-Dimensional Writing Spaces" was published in Leonardo 32.4 (1999): 261-8.
About Lucid Mappinig
Lucid Mapping and Codex Transformissions in the Z-Buffer is an investigation of textual and narrative possibilities within three dimensional on-screen environments (specifically Virtual Reality Modeling Language, or VRML). Functionally, it is both a text to be read and a space to be surveyed. The various elements of the title -- "lucid mapping," "transformissions," "the Z-buffer" -- are all glossed within the VRML environment itself, so I will not discuss them in detail here.
The project evolved from a set of early schematic models. These experiments in "spatial heuristics," as I called them, were not conceived as displays of technical virtuosity; rather, I hoped to begin exploring some of the questions raised by three-dimensional information spaces: How well suited are they to organizing documents according to the non-linear principles pioneered by hypertext developers? How can the addition of a third dimension extend (hyper)textual visualization? What are the conditions required for "narrative" in three-dimensional on-screen environments? And so on . . . (Mark Bernstein of Eastgate Systems offers a skeptical treatment of some of these questions here.)
As I was finishing work on Lucid Mapping in the fall of 1997, I came across the following remarks by hypertext author and theorist Michael Joyce (from a recent paper):
Of my experiences of virtual reality thusfar, I remember only one with a visceral excitement and longing: the experience of moving in and out of planetary spaces of text within a 2D rendering of 3D typographic space which I experienced in the work of the late Muriel Cooper together with David Small, and Suguru Ishizaki at MIT's Visible Language Workshop. "Imagine swooping into a typographic landscape: hovering above a headline, zooming toward a paragraph in the distance, spinning around and seeing it from behind, then diving deep into a map." Wendy Richmond described it perfectly in WIRED, "A virtual reality that has type and cartography and numbers, rather than objects - it's like no landscape you've ever traveled before, yet you feel completely at home."
A paper describing Cooper, et al.'s work at MIT is available from the IBM Systems Journal.
As a typographical and topographical environment, Lucid Mapping owes something to these experiments in information design (see also the technical report from the Language Visualization and Multilayer Text Analysis group at Cornell). But both of these projects, as well as Lucid Mapping, are ultimately (for me) equally as compelling, if not more so, as aesthetic explorations -- aesthetic explorations conceived both within the historical contexts of visible language and amidst the luminous ray-traced contours of the virtual.
About the VRML
The VRML conforms to the current VRML 2.0 standard.
The workflow for the project was as follows: All of the text was first created in Photoshop 4.0; some modest use was made of the software's filters to lend variety to the appearance of the type. All image files were then saved as transparent GIFs. The actual VRML was constructed using Silicon Graphics' CosmoWorlds authoring package (version 1.0.2), running on the SGI Indigo2 Impact at the Institute for Advanced Technology in the Humanities. The transparent Photoshop images were texture mapped onto ultra-thin three-dimensional solids, thus creating the illusion of a two dimensional surface. The pieces were then arranged in their final positions. Viewpoints, light sources, and animations were added last, and the VRML was then packaged and compressed.
I should note that VRML is not a text-friendly environment. Indeed, most examples of text and type in VRML spaces that I have seen have been quite crude. Both the VRML 1.0 and 2.0 standards support a basic ASCII node which allows rendering of short strings of alphabetic characters; its purpose seems geared towards labels, signposts, captions, and the like. Therefore, in order to achieve the more sophisticated textual arrangements of Lucid Mapping, the typographic components were treated as above.
In addition to my own writing, Lucid Mapping is composed of both attributed and (occasionally) unattributed quotations from the following works:
Drucker, Johanna. The Century of Artists' Books. New York: Granary Books, 1995.
McCaffery, Steve, and bpNichol. Rational Geomancy: The Kids of the Book-Machine. Vancouver: Talonbooks, 1992.
McLeod, Randall. "From 'Tranceformations in the Text of Orlando Furioso.'" New Directions in Textual Studies. Ed. Dave Oliphant and Robin Bradford. Austin: U of Texas P; Harry Ransom Humanities Research Center, 1990. 61-85.
---. [AKA Random Clod, Random Cloud]. "Information on Information." Text 5 (1991): 240-281.
Rotman, Brian. Signifying Nothing: The Semiotics of Zero. Stanford: Stanford UP, 1987.
Special thanks are due to Michael Tuite and Will Rourk from the New Media Center for assistance with the video capture, and for preparing multiple CD-ROM copies of this material.
Viewing the VRML
I have provided for three different means of viewing the VRML: first, the actual Lucid Mapping VRML file itself. (If the preceding link does not produce a working VRML file, try this version instead.) Please note that I strongly recommend using an SGI workstation to view the VRML -- and to do so you will also need a VRML browser or plug-in, such as CosmoPlayer.
Please be sure to set your browser's Collision Detection to Off. I also recommend viewing the environment in Fly rather than Walk mode.
Though there is cause for optimism, at the present moment the graphics capabilities of most Macs and PCs are simply not robust enough to provide a satisfactory rendering of complex VRML environments -- in the case of Lucid Mapping, images and text may appear distorted or illegible, and movement may seem sluggish. Therefore, to compensate, I have provided video of the VRML, captured directly from an SGI workstation (QuickTime and MPEG). These should play on any Mac or PC that supports these formats. While the movie clips offer a reasonably accurate depiction of the VRML, the sacrifice, of course, is that the user is unable to control his or her own movements. The interactive element of the environment is therefore greatly suppressed.
Lastly, I have provided a number of static screenshots of the VRML (in JPEG format) captured on the SGI -- these are accessible by browsers or image viewers on any platform.