About | Project Members | Research Assistants | Contact | Posting FAQ | Credits

CaveWriting and the CAVE Simulator

Research Report by Nicole Starosielski
(created 2/6/07; version 1.0)
[Status: Draft]

Related Categories: Immersive Text Environments | Alternative Interfaces

Original Object for Study description


Cave Writing is an interdisciplinary artistic practice developed at Brown University for the CAVE simulator, a virtual reality environment typically used for scientific visualization. Cave Writing began in 2002 when hypertext fiction writer Robert Coover initiated a series of workshops in Brown University’s CAVE that brought together faculty, students, artists and scientists in the development of creative projects integrating text, visual imagery, narrative and sound. Several notable projects from the workshop include Screen, developed by Noah Wardrip-Fruin, et al., John Cayley’s Torus and William Gillespie’s Word Museum. The release of CaveWriting 2006, a spatial hypertext authoring system designed by workshop developers, allows authors to directly manipulate text, imagery and 3D models in a graphical front-end environment. CaveWriting now expands beyond the physical limits of the CAVE simulator, making it relatively easy for anyone with a compatible personal computer to experiment and explore writing and reading in three dimensional environments.

CAVE is a recursive acronym which stands for CAVE Automatic Virtual Environment. The original CAVE simulator and library were developed in 1992 by a team of researchers led by Thomas A. DeFanti, Daniel J. Sandin and Carolina Cruz-Neira at the Electronic Visualization Lab (EVL) at the University of Chicago Illinois. The EVL is an interdisciplinary graduate research laboratory which works at the intersection of art and science and specializes in advanced visualization and networking technologies. A CAVE consists of four, five, or six walls which display rear-projected stereo images, creating an immersive space navigable by means of a handheld device. Unlike the traditional virtual reality format of the head mounted display (HMDs), CAVEs allow several people to be co-present in virtual space: it is a one-to-many device rather than simply a one-to-one device. The CAVE has since provided the basis for at least two other virtual reality devices developed at EVL, the ImmersaDesk and the IWall. CAVEs and CAVE-based environments have become a well known form of virtual reality; the system is also a commercial product marketed by the immersive display/virtual reality company FakeSpace SystemsTM.

A number of institutions have constructed CAVEs, which are primarily used for scientific research and visualization. Brown University’s Center for Computation and Development installed a CAVE in 1998, and in 2002 the Cave Writing workshop was initiated by Robert Coover, novelist, hypertext fiction writer and co-founder of the Electronic Literature Organization. These workshops mark a significant and innovative use of an immersive virtual reality environment for interdisciplinary textual research and artistic experimentation. The Cave Writing workshop at Brown facilitates the collaboration of undergraduate and graduate students, writers and programmers, artists and scientists from a number of different fields. The initial objectives of the workshop included introducing text, positional sound, and narrative movement into what was predominantly a visual, silent, and spatial environment. The focus was to explore the word in an immersive, virtual context.

The Cave Writing workshop brought in a number of renowned literary theorists and new media artists to work with the CAVE simulator. Noah Wardrip-Fruin, Andrew McClain, Shawn Greenlee, Robert Coover, and Josh Carroll collaboratively developed the well known project, Screen. The description on Wardrip-Fruin’s hypertext.org site reads: “It begins as a reading and listening experience. Memory texts appear on the Cave’s walls, surrounding the reader. Then words begin to come loose. The reader finds she can knock them back with her hand, and the experience becomes a kind of play – as well-known game mechanics are given new form through bodily interaction with text.” {1}. The two-dimensional text becomes a series of three-dimensional objects, movable and playable, read across and through a number of layers in immersive space.

John Cayley and collaborator Dimitri Lemmerman crafted Torus to address the phenomenology of text in three-dimensional space. The central figure of their project is a torus constituted by sixteen vanes of text arranged around the central cavity. The reader begins by observing the distant shape and then moves in, around, and through it. The letters have no thickness and always turn to face the Cave explorer’s point of view. Rather than simulate a pre-existing phenomenology, Cayley argues that his project experiments with the inherent potential phenomenology of text in three-dimensional space. Through these experiments, Cayley forms a number of conclusions. First, that we approach literal forms with expectations of relative size: if all of the graphic figures are on the same plane, we expect them to be approximately the same size. If they are different sizes, we assume there must be some paratextual explanation (e.g. it is part of a title). In three-dimensional space, size easily becomes a signifier for distance. Cayley forms the hypothesis that “literal forms are highly effective for delineating space in immersive virtual environments” {2}. He also suggests that in these environments, the surface of the letter becomes a space of inscription where other letters may reside.

Both Cayley’s and Wardrip-Fruin et. al’s work shows the way in which Cave Writing is not simply about the creation of new textual arrangements, but a mode of generating further observations about our textual experience in both two and three dimensional spaces. In fact, the aim in many of the Cave Writing projects has been on abstraction, metaphor, and symbolic imagery rather than the creation of an immersive, realistic environment where a representational story plays out around a viewer. Robert Coover states of this approach: “I encourage people to try to keep things relatively simple, using texture mapping instead of intricate modeling where possible, achieving their effects by means other than the attempted imitation of the real world, using their own imaginations to challenge and stimulate the imaginations of others.” {3} The approach of the Brown workshop projects have situated Cave Writing as a critically engaged practice, a process which often involves reflection on the implications, phenomenologies and of digital words and texts.

One of the problems in the Brown Cave Writing collaboration was that “despite the success of the final product, the production process often frustrated the writers, who spent much of their time waiting for the programmers to finish the next iteration of the code” {4}. After the completion of Screen, the workshop programmers began developing what would become CaveWriting 2006, a graphical front-end environment in which authors could directly manipulate a model 3D world on a desktop computer. The Java front-end is known as the Cave Text Editor, and gives authors a range of potential functions (Fig. 1). They can define objects (for example images, video, or text), group these objects, and attach parameters, movements and hyperlinks. The author can also define a timeline for object actions and can specify the occurrence of certain events at either pre-specified times or in relation to user input. When the author saves the file, it is immediately navigable on either a desktop computer or in the CAVE. The features of the system can be changed or extended by programmers in C++; the programmers are not replaced by the Cave Text Editor; rather a new space is opened for collaborative development.

Figure 1. Interface of the Cave Text Editor

In making the content of the original CAVE self-sustainable on an individual computer, the Cave Text Editor expands the potential space of Cave Writing. Anyone can download and use the program on the CaveWriting 2006 site. This development addresses John Cayley’s initial question of Cave Writing: “How can such a recondite and exclusive process of artistic making address audiences or generate a socially engaged practice?” {5} While the original CAVE represents a transition from a one-to-one towards a one-to-many virtual reality, the innovation of CaveWriting 2006 is part of a shift from one-to-many towards many-to-many. While the Cave Writing projects are not yet viewed on a large scale outside of the Text Editor or the physical structure of the CAVE, the developers plan to port CaveWriting to a number of additional platforms, making it accessible in a variety of places, including schools and museums.

Research Context:
The development of CaveWriting exists at the intersection of two longer research trajectories: the scientific investigation of virtual reality and the literary exploration of hypertext. On one hand, the invention and adoption of the CAVE simulator takes place in the context of the development of immersive technologies, virtual reality environments and information visualization. CaveWriting creatively incorporates text, narrative movement and sound into these spaces. On the other hand, CaveWriting represents an expansion of pre-existing hypertext practices onto a new platform and provides a new environment for hypertext authors.

Technical Analysis:
The CAVE simulator can have a variety of dimensions and can be projected on four, five or six walls. Brown University’s CAVE has three walls and a floor. High resolution images are rear projected on the three walls and projected from above onto the floor. CrystalEyes shutter-LCD glasses are used for 3D stereo-viewing. Intersense and Polhemus sensors are used to track and register the user’s movements, altering the environment to correspond with their motion.

The CaveWriting authoring system consists of a cross-platform application and software library, a web accessible, graphical, front-end written in Java with the JAXFront design tool and, a Max/MSP patch which spatializes audio. The application layer reads from an XML-based story-file which designates a range of objects: images, video and text, as well as events that take place either at specified times or in response to activation of hyperlinks by a user. Once the story is loaded, it is rendered in real time using the software libraries FTGL, OpenQuicktime, and the open-source rendering engine G3D. The application also accepts input from the head-tracker and a three-button handheld wand to track the user’s movement. {6}

Evaluation of Opportunities/Limitations for Transliteracies Topic:

The use of the CAVE for creative purposes has significance to the Transliteracies project as it represents interdisciplinary research and collaboration between the sciences and the arts. The result of this endeavor has generated new types of interaction between the artists and scientists (in the development and continual modification of the Cave Text Editor), the birth of a variety of navigable hypertext environments, and the production of software for 3D computer writing which is accessible to non-programmers, as well as the general public. CaveWriting provides a valuable example to the artists and scientists collaborating in content production for the Allosphere, an immersive virtual reality environment located on the UCSB campus.

Since many of the works produced in Brown University’s Cave Writing workshops explore potential modes of reading and phenomenologies of the word in virtual and immersive space, these projects provide interesting examples and reference points for the study of online reading. The spread of the CaveWriting program may increase the number and range of projects created, and potentially the reflections on digital texts and reading practices.

The CaveWriting system itself is important to the Transliteracies project: it is a free, accessible digital writing environment. A number of proprietary systems are currently available, but the CaveWriting program is intuitive, simple to use, and geared toward those interested in the manipulation of text. As it is written specifically for reflection on textual forms, CaveWriting might an appropriate platform from which further engage the study of online reading.

{1} From Noah Wardrip Fruin’s hyperfiction.org site: http://hyperfiction.org/screen/. Last accessed December 2006.

{2} Cayley, John. “Lens: The Practice and Poetics of Writing in Immersive VR (A Case Study with Maquette).” Leonardo Electronic Almanac. Vol 14, Issue 05-06.

{3} Interview with Noah Wardrip-Fruin, Robert Coover, and Josh Carroll. The Iowa Review Web. http://www.uiowa.edu/~iareview/tirweb/feature/cave/ . Last accessed December 2006.

{4} Baker, Damon, Sascha Becker, Robert Coover, Ilya Kreymer and Nicholas Musurca. “CaveWriting 2006: A Hypertext Authoring System in Virtual Reality.” Material presented at the ACM SIGGRAPH 2006 conference.

{5} Cayley, “Lens: The Practice and Poetics of Writing in Immersive VR,” p. 1.

{6} Baker, “CaveWriting 2006.”

Resources for Further Study:

Baker, Damon, Sascha Becker, Robert Coover, Ilya Kreymer and Nicholas Musurca. “CaveWriting 2006: A Hypertext Authoring System in Virtual Reality.” Material presented at the ACM SIGGRAPH 2006 conference.

Becker, Sascha, Shawn Greenlee, Dimitri Lemmerman, Morgan McGuire, Nicholas Musurca, and Noah Wardrip-Fruin. “Cave Writing: Toward a Platform for Literary Immersive VR.” SIGGRAPH 2005 Sketch.

Cayley, John. “Lens: The Practice and Poetics of Writing in Immersive VR (A Case Study with Maquette).” Leonardo Electronic Almanac. Vol 14, Issue 05-06.

Cayley, John. “Writing on Complex Surfaces.” http://www.dichtung-digital.org/2005/2-Cayley.htm. Last accessed December 2006.

Coover, Robert. “Cave-Writing: New Adventures in Mot-Town.” KOSMOPOLIS. Festival Internacional de la Literatura, Barcelona, 14-19 September, 2004.

Cruz-Neira, Carolina, Daniel. J. Sandin, and Thomas. A. DeFanti. “Surround-Screen Projection-Based Virtual Reality: The Design and Implementation of the CAVE.” Proceedings of the 20th Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH ‘93). New York: ACM, 1993. pp. 135-42.

Hayles, N. Katherine. “Is It Literature — or Art?” http://projects.design.ucla.edu/exhibitions/SecondNatures/kh.html. Last accessed December 2006.

  nicoles, 02.06.07

Comments are closed.