Designs of interactive, navigable 3-D worlds abound. Computer games offer more and more elaborate imaginary worlds, and architectural 3-D renderings of individual building projects and whole cities are provided. While there is a growing body of computer graphics research on tools and techniques for the construction of such 3-D worlds, we can hardly find any literature on appropriate visualization strategies for their design.
A classical reference is Benedikt (1992) who argued against realistic representations that turn the world into a facsimile, a literal interpretation of the real. Benedikt talks about "a sliding relationship between the symbolic and the literal" as one of the basic principles for designing intuitively knowable worlds. Based on this early discussion, Norman, O'Brien and Rodden (1998) refer to the visual artsnamely cubism, futurism and suprematismas a source of inspiration for designing 3-D environments. They point to the aptitude of abstract art to communicate spatial depth and dynamics. While it is of great relevance to examine a diversity of painting styles, imitating abstract art is not necessarily a good approach to 3-D design. Mitchell (1994) has argued eloquently that abstract art raises the same issues of representation, iconography, and storytelling than more realistic (in the sense of "dense and rather saturated with significant features") figurative art.
The challenge is to develop visual languages which create meaningfullegible and reflexiveimage worlds, taking into account the affordances but also the limitations of a medium, which is primarily visual (and only peripherally auditory), non-immersive, and blurs familiar points of reference such as inside/outside, chronology, and spatial connectivity (Sobchack 1988).
The case around which we construct our arguments is the design of an interactive 3-D world in support of architectural work. The 3-D Wunderkammer is a visual environment in which multimedia materials can be placed, stored, encountered, found, displayed, and integrated with the flow of the work. It is created interactively, with users placing their own collections of inspirational objectsscanned images, sound, video, rendered 3-D objects. While the concept of Wunderkammer and the architects' experiences of use have been described elsewhere (Wagner Ed 2001, Büscher et al. 1999, Kompast et al. 1999), we focus here on the graphic design of the visual interface. It involved a multidisciplinary team of professionals representing different visual culturesarchitects, graphic designers, computer graphic specialists, 3-D designers, and a social scientist. It resulted in a variety of designed 3-D worlds, which can be explored via the Internet, and in a set of design principles for such worlds.
In this article we will present the concept of 3-D Wunderkammer, analyzing the challenges involved in developing visualization strategies for its design. We will describe the different stages in the design of the Wunderkammer worlds, and finally, discuss the set of design principles which resulted from this multidisciplinary design experience.
The concept of 3-D Wunderkammer
The 3-D Wunderkammer took shape as a result of fieldwork in an architectural office, in close cooperation with the architects, and inspired by the cabinets of curiosities of the 17th and 18th century, such as the Wunderkammer of Rudolf II in Prague. These cabinets of curiosity housed collections of heterogeneous objects, featuring "hundreds of icons, alluring apparatus, a multitude of mirrors, maps, charts, drawings, instruments" (Stafford 1996, p. 28). Visitors used them as an inspirational resource for their work in the arts, the sciences, philosophy, and politics. One of the most interesting features of a Wunderkammer was that it presented these objects in a way that did not impose on the user an ordered set of relationships and ways of interpreting.
The need for a modern cabinet of curiosity was confirmed by observations of the ways in which architects collect, archive, and search for inspirational material, and how they use it for developing, expressing, communicating, and presenting their work. From these observations we learned:
- One of the main features of architects' work practice is the intensely visual way working material (most of which is visual and graphical itself) is handled and arrangedon tables, in the meeting room, on the studio walls. This visuality is an integral aspect of architects' work environment and of their ways of capturing and communicating ideas.
- Memories of "place" influence the architects' strategies of searching for and retrieving material. In particular, association material is strongly connected with places, such as a book, file or catalogue, a movie, museum, building, or particular landscape.
- Inspirations often arise from the transient and ephemeral way in which these association objects are encountered, their peripheral presence in the back of one's mind, and with activities that are connected to movement (such as browsing, flipping through, etc.).
- Architects' aesthetic categorizations and the inspirational character of material are strongly dependent on the context in which they arise. Professional designers reflexively construct the imagined qualities of a space, and the objects that help them capture and express these qualities are deeply intertwined with descriptions in speech, embodied action, and people.
The designer team translated these observations into an approach, which builds on the metaphors of travel and of the world as exhibition, as stimulating ways of encountering materials. The idea was to design worlds of intuitively knowable places, which provide visual associations for inspirational materials.
The current working prototype of the 3-D Wunderkammer offers a set of well-developed functionalities for navigating, searching for, and viewing (collections) of multimedia materials. It runs in a Web browser in an ordinary, affordable technical environment. The program is a set of Java Applets, and uses a standard plug-in to display VRML. We have installed the Wunderkammer in a real working environment for an extended period of experimentation and made a Web-based version available as a virtual world in a hands-on way (http://as15.iguw.tuwien.ac.at/desarte/index. html).
The 3-D world with its diversity of places, atmospheres, and landmark objects provides a rich and significant context for placing visual material, with the third dimension dramatically increasing the space for storing, viewing, presenting, and interacting with objects (Figure 1). Clicking, browsing, and scrolling between places with material is substituted by continuously moving walking, flyingthrough a particular geography. The continuous movement has a zooming effectimages grow "into the screen" and disappear again. The "magical" features of digital worlds, such as floating, flying, teleportation, and moving through solid objects can be used for reinforcing the experiential character of the worlds.
People, who primarily use the Wunderkammer as a visual archive, tend to prefer the simpler "thematic" places for their collections of inspirational objects. Here memory is tightly connected to a place and images are retrieved by revisiting a particular place. Using the Wunderkammer as a source of inspiration puts emphasis on the unfamiliar, on surprise and discovery. A third focus that emerged is the Wunderkammer as an exhibition and presentation space (Figure 2). The Wunderkammer invites and enables users to experiment with context, scenography, and arrangements (Wagner Ed 2001).
The first Wunderkammer worlds
The leading idea was to design image worlds that are legible, navigable, and aesthetically attractive. Inspirational resources were the metaphors and images that grew from the architects speaking about actual practice and needs, in particular the metaphors of travel and of the world as exhibition. Wunderkammer places should not resemble any real place but be "mise-en-scène" for inviting the placing of inspirational objects and for stimulating contradictory associations.
Developing the visual interface of 3-D Wunderkammer was done in several steps, with different versions reflecting changing priorities and insights. The first version was a simple symbolic "village" with a few significant and easily recognizable places. The graphic design was a system developer's literal translation of the architect's first sketch into a 3-D environment, using a series of simple and compact objects with clear contours (Figure 3).
A more elaborate assemblage of a type of city and landscape replaced this version. This 3-D world was directly inspired by the architects' approach to urban planning. They started out from a modular structure based on a neutral grid on a slightly curved surface, and defined the qualities of different fields or places in this grid (Figure 4). The idea was to provide structure and construct an open system, rather than a pre-defined world, which can be filled in with places of different qualities.
The selection and definition of places for this Collage City/Landscape was done as part of our fieldwork within the architectural office, with a view to the inspirational material used within current architectural projects:
- Skyscraper City - buildings around a plaza, which are packed with glitter, rich decoration, and icons of fashionable life, a cultural center, and a park leading into a derelict area
- Urban Void - a mixture of derelict industrial sites and slabs in Neoversimo style)
- High-tech industry - with a touch of futurism
- Medieval City Center - with a variety of historical elements and objects
- One-Family Housing - as a symbol of idyllic loneliness
- Landscape formations- mountain area, forest, agricultural area, ocean, and cost, leading into the desert.
The designs for this more elaborate world were provided by the architects themselves. The approach was architectural 3-D modeling with an emphasis on structural elements (Figure 5). The first version was still predominantly symbolic. It contained a series of significant objects, some of them pathsamphitheater, museum, highway, bridge, etc., which were placed in-between these modules. The next step was to focus on navigability and performance. The architects switched to 3-D Studio, a modeling tool with better 3-D compatibility. The symbolic-iconographic language was replaced by a structural approach. Each place was architecturally constructed. The screenshot of Industry captures some of the qualities of this world. Its language is abstract, focusing on building structure, but still naturalistic. Performance problems with these complex constructs induced the architects to reduce the details of their designs by removing significant objects. They started experimenting with textures
This world, while offering interesting views, in particular a bird's-eye view, showed some of the strengths but also the limitations of an architecturally constructed 3-D world. Architects' main field of experience has to do with structure, topography, the qualities of space, with light as a space-generating element, and with the effects of combinations of materials in the real world. Their visualizations in 2-D and 3-D allow the "immutable transfer" of information. They are detailed, optically consist scale-constructions of worlds to be built. In this sense they are "real" and at the same time highly abstract, lacking the kind of rich detail necessary for conveying highly differentiated content.
Searching for visualization strategies
Given the limitations of an architecturally constructed 3-D world, the team decided to engage the help of an experienced 3-D designer and a graphic designer. The idea was to combine the architectural approach with the language of film, comics, painting, stage design, etc., and to explore the suitability of these visual languages for a digital medium.
We approached this task by mobilizing searching for and collectinga great variety of visual examples from art books and journals, which were scanned and printed out. These examples were combined with (metaphorical) descriptions of the qualities of places in this world (their atmosphere and significant details) and prototypal 3-D designs to form mixed media representations of worlds "in formation." In a series of joint design sessions, an architect, a 3-D designer, a graphic designer, and computer graphics specialists talked through the design of three particular areasSkyscraper City, Ocean/Desert, and Industrial Landscapedeveloping ideas about the content of each of these worlds, describing atmosphere and details, and looking through the visual examples that had been collected. The architect produced sketches of each area, which were gradually filled with detail. The 3-D designer took the documentation of this conversation, together with the associated visual materials, as a script for his design work.
The first module exemplifying this approach was Ocean/Desert, an assembly of different formations of desert, mountains, and ocean into a highly compressed space (Figure 6).
While describing his notion of Ocean/Desert, the architect produced a sketch, which was arranged, together with visual examples copied from art books, into a rich visual representation of the area.
The designer team's main concern was to develop a visual language for expressing the identity of a place, its atmosphere, and its legibility at a glance. This required finding the right balance between the use of telling detail and expressing subtle, unexpected and surprising visual forms.
The graphic designer in our team not only explored a large variety of textures (such as images of façades) for conveying detail and for creating particular atmospheric qualities, and also supported our explorations of a diversity of artwork.
Intertwined with the legibility of a 3-D world are issues of orientation and navigability. Kevin Lynch's work (1994), with its focus on understanding the "practical grammars of perception" underlying people's orientation in urban spaces, has been widely referred to in user-interface design. Here the architect and graphic designer pooled their knowledge of space into designing the Wunderkammer geography. It contains some of those structures and objects that in our physical environment support orientation: significant points, signs, paths, crossings, regions, and boundaries.
Another problem and challenge in building interactive, navigable worlds has to do with performance. It is a well-known fact that rendering 3-D worlds in real-time so that users can explore and navigate through them with ease is a task that challenges the hardware and software configuration of a machine. Computer graphic specialists have the ability to navigate through a 3-D world, knowing how it is mathematically constructed. This helps them identify critical spots in a 3-D world, such as complex objects, disadvantageous object groupings, etc. Graphic specialists have developed the art of smoothening out, simplifying, and manipulating with optical illusion, without diminishing the visual qualities of the design. In doing so they have to combine their technical knowledge with aesthetic judgments, while absorbing, reflecting and translating what they learn from the architect and the graphic designer.
The 3-D designer's role in the project was to literally compose the worlds in great detail, and to decide when and what the user sees, from whole objects to detailed textures. As part of this he needed to find a balance between the different visual languagesarchitectural constructions, the language of film, comics, and stage design. If the 3-D designer wanted his scenes to be rendered in real-time, he had to take a few simple rules into account. Viewing and discussing the design with the computer graphic specialists helped him understand and apply these rules.
The result of this joint design work is three quite different Wunderkammer places, each of them designed in a different way and together with a series of design principles. The current visual interface is still a construction site, with placeholders for sub-worlds to be designed at a later stage, eventually by users themselves. We look at these principles, together with the rules for scene composition, which were developed by the computer graphics specialists in our team (Wagner ed 1999) as guidelines for people designing interactive, navigable 3-D worlds.
The basic layout of the visual interface is a neutral grid into which different places can be filled. Each field has been designed in a particular way; by evoking and underlining contrasts of style, detail, color, etc., differences & specifics have been emphasized (Figure 7).
Mixing visual strategies and styles
The design combines a diversity of expressive styles, such as the pictorial style of David Hockney, the photographic character of 19th century industrial buildings, the naturalism of old etchings, the expressionism of a painting by Marcel Grommaire, and the streaky lines of drawings. A 3-D collage of different images and styles visualizing a diversity of places and objects was created (Figure 8).
By leaving things unscaled or out of scale and placing them in close proximity to each other, open, unspecified, and sometimes confusing relationships between objects were created. This often counterintuitive scaling enables "uncombinable" things to be combined (Figure 9).
Mixing 2-D and 3-D
As in stage design, 2-D has been mixed with 3-D. In an architecturally constructed world stage sets acquire qualities in between unreal and "legible." Here, the painting "Hotel Sian" by David Hockney (Figure 10) has been made transparent. Travelers may walk through the gate to a place filled with objects of different qualities, some of which appear massive while others assume a floating, fragile, stage-like character. Figure 10b shows a collage of 3-D building (`Silo') with stage-set elements of bridge and ship.
A 3-D world allows variations of scale and density. The idea of Skyscraper City (Figure 12) can be condensed in a set of significant buildings around a plaza, which are packed with glitter, rich decoration, and icons of fashionable life. Variations of landscape or architectural styles can be assembled in close vicinity, just as a forest be packed with layers of woodlanddark, high-rising fir trees, boscage, tropical forest, "macchia."
Another strategy is to place contrasting environments next to each other without providing smooth transitions, emphasizing contradictions. In Industrial landscape (Figure 12) contrasting environments have been placed next to each other, emphasizing contradictions between artificial and natural, transparent and solid, derelict and proper.
Mixing complex and simple designs
Some of the fields in the current version of 3-D Wunderkammer are emptyplaceholders for worlds to be designed; others are symbolized by a single object such as an airplane, a church, a stadium (Figure 13). The idea again is to work with contrasts in a world, which may rapidly become too dense, loaded with detail, and in the end disorienting
From flying over to digging deep
Speed strengthens the discovery effect. It allows combining the "flowing, quick glance" with the possibility to dig deeper into some detail. The flaneur roams through the Wunderkammer, picking up what s/he likes, disregarding the uninteresting. While the sightseer follows some predefined path, the hasty traveler flies over the world, taking in transient and distant impressions (Figure 14).
Designing navigable, interactive 3-D worlds is partly an aesthetic and partly a technical question. While having to solve the technical problems of optimizing performance in practice, we focused on the challenge of creating designs that generate a different view on things, helping perceive the novel in the familiar, discover relationships between seemingly incongruous objects, and relate the unrelatable.
Obviously realism is not an answer to this challenge. The designer team approached it through bringing different design disciplines together, pooling their knowledges and expertise. The design principles we presented in this article are a result of this work. At the core of their development was an intense discussion of a great diversity of visual examples of artwork: paintings, film, stage design, comics. The designer team experimented with different strategies of combining the visual languages of these artworks into 3-D scenographies.
Although not directly translatable into the design of 3-D worlds, architectural debates about spatial qualities inspired several of the design principles we developed, such as the field concept, compression of space, changing scales, mixing visual strategies and styles. Contemporary architectural debates involved hybridity, flexibility, and the openness of space; enlarging the choices of materials and techniques and intensifying the effects of structures and materials (Wagner and Lainer 2001). There are traces of these debates in the design principles.
Also, architects' experiences with intensifying the effects and qualities of an object, with mobilizing built structures and multiplying modes of perception and use, have proven useful in 3-D design. They helped the 3-D designer move away from an architecturally constructed world, to experiment with stage sets, with contrasts of style, details, and color, sharp boundaries, and edges alternating with more fluent transitions between fields. Moreover, as representatives of a highly visual culture, architects know how to translate between different kinds of representations and between scales (between 2-D and 3-D, sketch and detail drawing, etc.).
The design principles, although described textually, have been made visible through particularly telling views of the 3-D worlds. We look at these screenshots as visual examples of directions to take in 3-D design rather than as accomplished prototypes.
The 3-D Wunderkammer was developed as part of Esprit LTR Project 31870 DESARTE. We wish to express our gratitude to our co-designers Johannes Siglär, Richart Schneider, Kresimir Matkovic, Maria Sciencik, Michael Strobl, Michael Gervautz, and Erich Monitzer.
8. Sobchack V. (1988) The Scene of the Screen. Beitrag zu einer Phänomenologie der `Gegenwärtigkeit' im Film und in den elektronischen Medien, In Gumbrecht H. U. and Pfeiffer K. L. (Eds) Materialität der Kommunikation, Frankfurt: Suhrkamp, pp. 416-429.
11. Wagner I. and Lainer R. (2001) Open planninginspirational objects, themes, placeholders, and persuasive artefacts. In: Proceedings Colloque Architecture des systèmes urbains, Université de Technologie de Compiègne, July 5, 2001.
12. Wagner I. Ed (1999) Desarte. The Computer-Supported Design of Artefacts and Sapces in Architecture and Landscape Architecture. First Year Report, Institute of Technology Asessement & Design, Vienna University of Technology.
13. Wagner I. Ed (2001) Desarte. The Computer-Supported Design of Artefacts and Sapces in Architecture and Landscape Architecture. Final Report, Institute of Technology Asessement & Design, Vienna University of Technology.
Ina Wagner & Martin Kompast
Vienna University of Technology
Academy of Fine Arts in Vienna
Design Column Editors
89 South St, 2nd Floor
Boston MA 02111
Rivendel Consulting & Design, Inc.
P.O. Box 334
8115 La Honda Rd. (for courier services)
La Honda, CA 94020 USA
©2002 ACM 1072-5220/02/0900 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.