Heo Jun-Young (PhD student), Aghlab Al-Attili (PhD student), Pedro Rebelo (lecturer), and Richard Coyne (PhD supervisor)

           School of Arts, Culture and Environment. The University of Edinburgh






Time is central to architectural design, but has not been fully investigated through computational media. The works of Heidegger [[1]], Bergson[[2]], Virilio[[3]] and Deleuze[[4]] suggest that the study of the elusive concept of time has more to contribute to an understanding of the human condition than space. One can argue that contemporary society is being governed increasingly by temporal structures, as the space of the town square is replaced by time-based broadcasting and digital communications. We present three digital media projects that start with the problematic of time, and examine the implications for architectural design of working with media where time is a primary consideration. In turn, the projects address “real-time” interaction, “evolution” over time, and time-space perturbation.


Keywords. evolution; phenomenology of space and time; time-based architecture

Smooth simultaneity

In his discussion of new (digital) media, Lev Manovich (1996) notes similarities in Walter Benjamin [[5]] and Paul Virilio’s[[6]] approach to “the intervention of technology into human nature.” (Manovich, 1996)[[7]] He points out that both Benjamin and Virilio equate our perception of the natural with spatial distance between the observer and the observed. Technology (film for Benjamin and telecommunications for Virilio) reduces this distance. The fact that anything can be transmitted anywhere at the “speed of light” makes the notion of distance redundant. This condition assumes the collapse of the spatial dimension altogether. Distance and the inevitability of time delay that once provided the opportunity for assimilation and reflection supposedly cease to exist. With instant communication comes instant reaction time, a feedback loop that ultimately can only be handled by computers. Whereas the collapse of distance is, for Benjamin, marked by the development of film and its ability to represent different spaces at the same time, Virilio transfers the collapse of space to telecommunications. The richness of our perceptual field is diminished, removing what Benjamin calls “aura”. The “collapse” of distance implies a condition which conflates the observer with the observed, the human with the machine, the subject with the object, transforming previously discrete dualities into a blur.


Perhaps this fusion and collapse in space and time accounts for one aspect of our experience with new media. On the other hand our media-saturated experiential field sometimes presents the character of a fracturing, a kaleidoscopic confusion of media and images (Tschumi, 1994). The troubling of distance through film and telecommunications can also be seen as aggravating the gap, instilling a separator in the observer-observed continuum, fragmenting our experience, or perhaps giving expression to the already fragmentary nature of human experience.

     Arguably, spatial and temporal discontinuities are best appropriated when space and time are processed in relation to each other. For example, film can use time to render spaces disjoint. Instead of the broad sweeping panorama (smooth), filmmakers cut from one scene to another. Instead of conveying a scene as one continuous time sequence they introduce an element of temporal disruption by switching from one spatial location to another. We note Alexander Sokurov’s “The Russian Ark”, filmed as one continuous 90 minute take, and spanning several centuries, where the “cut” is achieved at the threshold, the doorway, the passage, or the gaze that lingers (on a painting), and from which we awake to a new time frame. Of course cuts can be used to convey a sense of the smooth. Perhaps in a film and MTV-enculturated world cuts are everywhere. Whatever the “impression,” be it of smooth or distressed, the effect takes place at the cut, the conflation, and this is time-abetted. In any case, it is a temporal plus spatial control that gives new media its power to explore discontinuity.

Temporal 3D Environments


What kind of spatial exploration results from digital tools that celebrate and exploit the time aspect? How does the designer gain access to the unmaking of smooth space, the fabrication of distressed geometries? How does the designer play with fractured and disturbed unities? Time-based media provide an opportunity to play with the smooth and the rough, the continuous and the fragmented, and thereby explore discontinuities within our experiential field.

Real-time three-dimensional technologies have only recently become part of the designer’s tool box. Digital tools that are now commonplace suggest functions and processes that typically become extensions of traditional design practice. An architect might use 3D modelling software at various stages during the design process. The ability to mould and shape surface and volume is provided as an extension to traditional physical model making. The promise for photo-realistic visualisation comes as a well-received technological “improvement” on drawings and perspectives. The smooth ideal draws on metaphors of creation and process that pertain to temporal-continuity (Lynn, 2003) whereas the distressed exploits intervention, disruption, event-based discontinuities, the non-linear and the non-local (Novak, 1996).













The first project is a generative system for “real-time” 3-D modelling using a procedural programming language (C++) with access to OpenGL 3D graphics primitives. Generative models can be used to simulate 3D architectural spaces to successive levels of detail, thereby contributing to an understanding of design as a process of evolutionary refinement. The project raises questions of evolution and continuity through algorithmic events.


Casella di testo: Figure 1.1 A basic tree by L-grammar

This project develops a generative method to explore time-based 3D modelling for abstract spaces. The algorithm starts with a simple “key” parameter from which a “self-generative” complex form develops. The system is based on a “growing tree code” via an L-grammar [[8]]. Almost all elements in the structure are generated to the next stage by following the “growing tree code” process. The recursive algorithm selects its branches in this derivational structure according to “the law of possibility.” Some elements grow faster than others, and some atrophy. The derivation of the elements is governed by one constant key parameter.



Figure 1.2 Anti-clockwise radius of gyration



Figure 1.3 Positioned primary elements and vertical structures by random code


In the examples shown here, a single element, which we call a “virtual brick,” is to be located in a certain position and forms the basis of an assembly. This assembly is then transformed into a complex spatial structure. This is a flexible, multi-cloning assembly which can be of any scale and extent. Colour, scale and the position of elements are all determined by the initial selection of the key parameter. Forms of great diversity can be generated. Figures 2-4 to 6 show the derivation of forms using different key parameter values.




a) test 1 and 2.



b) test 3 and 4

Figure 1.4 Linear form Structure: test 1, 2, 3, 4


Figure 1.5 Example of a full composition


As we see, a single block of computer code and variation of a single key parameter results in large differences in the composition. The compositional possibilities can be explored repeatedly over a short period of time. The project raises the issue of evolutionary derivation as a smooth or disjointed algorithmic process. Though the algorithmic process is smooth, successive runs of the program show discontinuities. Further discontinuity is introduced by the intervention of human agency, and multi-user agency in an immersive environment. These are the subjects of further research.




In our second project, we developed an event-based system using a set of programmable objects which are an extension to MAX/MSP [[9]] graphic programming environment-Jitter [[10]]. Jitter is a collection of 130 or so standard objects which deal with time-based graphic content. Subsets of these objects are dedicated to video processing, data manipulation, and 3D Open-GL programming and rendering. The Max/MSP/Jitter user community is mainly composed of artists, musicians and programmers who develop applications for audio and video processing, interactive installations and audio-visual performance.

Max/MSP presents the user with a graphic interface in which boxes with inputs and outputs are connected (following the metaphor of circuit design). Each set of boxes (patches) can contain any number of nested boxes (sub-patches) and communicate with them at various rates. Its design is primarily for time-based applications and therefore follows a strategy of rapid prototyping which to some extent by-passes compilation and de-bugging, hence exposing the error, the accident, the interference, and makes them available for exploitation. The (application) design process that is implied in Max/MSP/Jitter tends to be attractive to non-programmers and to some extent forces a visualisation of decision-making processes and erratic connectivity. The clusters of boxes and connecting lines effectively represent a kind of multi-dimensional flow diagram that underlies the computational process. One of the central aspects of Jitter is the fact that all data structures are represented as configurable matrixes of specified data types, dimensions and sizes. A great part of Jitter’s possibilities in the realm of processing and mapping relies precisely on this fundamental entity. The values of a pixel array on a screen of 400 by 400 would take the form of a 4 dimensional (RGBA) integer, 400 by 400 matrix. Figure 3 shows an by connecting/disconnecting, re-configuring in a time-dependent way, and experimenting while examining the visual or kinaesthetic result.


  Figure 2.1 Example of environment with which the designer interacts



The time-based medium of Macromedia Director’s Shockwave 3D [[11]] exposes a series of 3D primitives, user behaviours, 3D animation strategies and a real-time physics simulation engine (Havok). Shockwave 3D has been used by designers of computer games that, due to Shockwave’s standard specifications and  increased computational power of the average PC, can distribute and play their environments on-line.

We indicate something of the potential of this time-based environment in our third project that explores the character of the home as described in Bachelard’s (1964) Poetics of Space, the home as a cellar, a garret and a hut. The phenomenon of space is closely linked to intimacy and memory in Bachelard’s writing. Certain parts of the house, such as the attic, serve as “repositories” of memories. The house also provides a person’s protypical spatial experience, a reference point from which all other spatial experiences derive and with which they are compared. The house is also understood episodically, in relation to sequences of events (getting out of bed, bathing, dining, opening windows, etc). From this perspective, our being-in-the-world is structured narratively. The house serves as a space for Bachelard’s narrative, and a house is itself a narrative space.

To instantiate Bachelard’s spatial narrative as a 3D computer model available for game-like navigation and interaction introduces some startling incongruities. As users of this new space we sense a familiarity with it, though we are perhaps struck by the mismatch between the medium and our bodily awareness. Our physical presence is perhaps reduced and moved into hardware and software. Our sense of recognition is suspended and the spatial phenomenon reduced to concepts of digital interaction.

The Shockwave 3D environment works with concepts of the model, movement, interaction and frames. It is possible to “jump” from one frame to another, in the manner of movement from one frame, sequence, or episode in a film. It is also possible to overlay frames, as in the case of film overlays. We have extended the frame metaphor to a consideration of a room in the Architecture building [[12]], data projecting a window as presented in the Shockwave 3D “attic” space onto a window in the room (Figure 1.1). The actual window is covered by a screen and an open window is projected (Figure 1.2 & 1.3).

Different interactions are available, such as opening the virtual window or closing it, or opening a blind — simple prototypical micro-event that we may have performed many times before, and that frame our experience and structure our space for the moment.

Of course the familiar event is rendered strange in this encounter as we see familiar objects projected. As we encounter something foreign, we draw on the strength of the metaphoric relations between image and space in order to make sense of our environment. The interaction of the user of this space and the computer-mediated space draws on the power of metaphorical association. The project highlights issues of familiarity, interaction, augmentation, the virtual, narrative and metaphor. We expect that a phenomenological understanding of such interventions helps develop understandings of digitally mediated space.

The familiar, homely event of opening a window is rendered strange, and consequently gives us a new understanding of the spaces we inhabit. The next challenge is to test this interaction with subjects to see what narratives of augmentation (metaphors) emerge. The task will then be to examine multi-user interaction in the same space, to see how such experiences are negotiated collectively, and through digitally-mediated communications.


Figure3.1. Picture of house section showing projection.




Figure3.2 & 3.3. Pictures of architecture window projected as closed window first, and open in the other one.








The conclusion of our study is that time considerations can perturb the impetus towards smooth, seamless interaction and step-wise derivation, contrary to developments which seem to aim for techno-human environments that are integrated, seamless and smooth. It also questions the recurrent digitally-inspired theme of an architecture based on organic and smooth forms.

            The practical application of the outcomes of these explorations is a means of exploring and generating designs and patterns, in a way that is algorithmic, interactive and time-based, utilising computer animation and multi-user control. The challenges include further developing languages for designing in new ways, and new languages of interaction design. What are the best ways of interacting with such capabilities? The interface is unlikely to be smooth and seamless.

One of the most interesting ways for design practitioners to appropriate these capabilities is to experiment with tools that are designed primarily for other than architects and spatial designers. These include computer tools for animators, artists, choreographers, film makers, and composers. There is also benefit in collaborating with such practitioners, who bring different conceptions of space, time and computer capability to bear on the design process. This paper represents such a collaboration, in our case between architectural designers and practitioners of the time-based media of musical composition.



































Bachelard, Gaston.1964. The Poetics of Space, trans. Etienne Gilson. Boston: Beacon Press. First published in French in 1958.

Benjamin, W. 1992, Illuminations. Trans. H. Zohn. London, Fontana Press

Coyne, R.D. (2002). The cult of the not-yet. In Designing for a Digital World. Ed Neil Leach. London: Wiley-Academic, 45-48.

Deleuze, G. 1992, Cinema 1 : The Movement-Image. Trans. H. Tomlinson and B. Habberjam. London, Athlone.

Heidegger, Martin. 1962. Being and Time, trans. J. Macquarrie and E. Robinson. London:SCM Press. First published as Zein und Zeit in 1927.

Lynn, G.: 2003,

Manovich, L.: 1996, Film/Telecommunication Benjamin/Virilio. (May 2003)

Novak, M.: 1996, Transmitting Architecture: The Transphysical City (May 2003)

Schneider, C.W. &Walde, R.E. (1992), L-system computer simulations of branching divergence in some dorsiventral members of the tribe Polysiphonieae (Rhodom elaceae, Rhodophyta). Eur.J.Psychology.,29:pp165-170.

Tschumi, B.: 1994, Architecture and disjunction. Cambridge, Mass., London, MIT Press

Virilio, P.: 1991, Lost Dimension. New York, Semiotext(e)

COSMO3D REFERENCE MANUAL\Program Files\Silicon Graphics\Optimizer\doc\developer\cosmo3dIndex.html

Karl S CHU; Modal Space. On the


End note

[[1]] Heidegger, Martin. 1962. Being and Time, trans. J. Macquarrie and E. Robinson. London:SCM Press. First published as Zein und Zeit in 1927.


[[2]] Mullarkey, J. 1999.Ed. The New Bergson. Manchester and New York: Manchester University Press

[[3]] Virilio, P.: 1991, Lost Dimension. New York, Semiotext(e)

[[4]] Deleuze, G. 1992, Cinema 1: The Movement-Image. Trans. H. Tomlinson and B. Habberjam. London, Athlone

[[5]] Benjamin, W. 1992, Illuminations. Trans. H.Zohn. London, Fontana Press

[[6]] Virilio, P.: 1991, Lost Dimension. New York, Semiotext(e)

[[7]] Manovich, L.: 1996, Film/Telecommunication – Benjamin/Virilio. (May 2003)


[[8]] L-System,  Lindenmayer System, is a string-rewriting techniques developed by Astrid Lindenmayer in1968, which can be used the model the morphology of a variety of [rganisms(Schneider,1992),

[[9]] 1990-2003 Cycling 74/ IRCAM

[[10]] 2003 Cycling 74

[[11]] Macromedia Director MX, Macromdia,Inc. 1984-2002.

[[12]] Architecture Building, (Minto House), School of Arts, Culture and Environment, University of Edinburgh