01// Keywords
free association, thesaurus, zooming, entirety/detail, gestalt, organic,evolving, architecture, topkapi palace, rhizomatic growth, infinite, time travel, non-linear, the image work, search engines, internet surf, data visualisation, 3D navigation, 2D navigation, charts.

Panorama of environment

02// Objectives
The aim of this project is to visually demonstrate my personal understanding of the evolution of the computer generated "image work" [01], by means of a free associative process that utilises the search and query mechanisms of the internet. I have tried to create a structure that displays this process by showing the data gathered in detail as well as in its entirety: Zooming in and out of objects and virtual navigation following free associations that can be evoked through online thesauruses, internet search engines and the ensuing surf mechanisms that can be utilised in the act of image creation, very much like collage/assemblage. Thus, seeking chance encounters and found objects, I turn not to second hand shops, bookstores and discarded magazines and catalogs, but to the internet. My process is inverted to that of collage in that it is the associations themselves that reveal the found object and not the object that reveals the associations [02].

I am interested to find out what a structure composed of these interrelated ideas and images will look like, whether it wil convey the overall content and aim of the project when looked at from a distance, and the specifics when looked at in proximity. I am differentiaiting between data visualisation that carries as its primary goal the quick and accurate location of a specific data within an overall structure without the user getting lost [03] and my aims: I want the wiever to get lost, to wander from association to association, to spend time, to zoom in and out and view entire chains of associations as well as read the details.

Zooming in and out

Zooming has, of course, been used long before, in film. What makes zooming interesting now, is that here we hand over the magnifying glass to the user as a tool of interactivity. As we zoom into and out of virtual spaces we encounter new levels of detail as well as view entire structures. Thus the artist is faced with the challenge that distance and proximity work equally well on one plane of perception, not just in certain instances, as would be the case in film, but consistently, almost simultaneously. One of the tasks that I have set myself here, is to investigate means and methods of making this duality of detail/entirety work.

The fourth dimension: The timelessness involved in the process itself in the computer environment continually fascinates me: Additions can be made to work at any time or the original work be modified into new generations/mutations. Thus working with a computer has an element of infinity, if not time travel, embedded in it. This open endedness is known in the online art/design communities on the internet as "tweaking". The young digital artists that hang out in online communities such as deviantart will often write comments about their own creative processes:

"+ekud vs ^alphakx

this is a cool piece that i have completely forgotten about for months. i opened it today, and was like 'fuck, thats good, why don't I tweak this?'

fixing that now.

j+b" [04]

Users of image processing software will be very familiar with the unlimited undo's and redo's of the history palette and that enable the artist to traverse backwards and forwards in time, the ability to save many many generations of the same work on the way, with absolutely no need of destroying the original. This project does not wish to demonstrate this endlessly regenerative nature of the computer generated image work itself. However, in itself it demonstrates this process of constant regeneration: The 3D charts that can be viewed on this page, when scrolled to the left, show the status of the environment on January 25th and February 28th of this year. Not only have things been added to the environment but building blocks have changed and shifted. If such a 3D chart were to be made on a monthly basis it would show how much the work changes as it progresses. On the other hand, I have been diligently saving previous versions of the work, and thus I am at liberty to go back and "tweak" any one of them as the start of a totally new project or a different version of the existent one.

03// Methodology

  • Content: Writing one keyword, into the online thesaurus kicked off the project. I chose the word “bridge” primarily because I use a bridge daily to connect between work and home and thus a bridge is something that I already have a lot of connections and associations with. I also decided to use this particular word because I sensed it had the ability to generate multiple associations from the start, which it did indeed fulfill, immediately enabling me to choose multiple paths I could pursue.

    The synonyms generated for the word "bridge" were written into either search engines or the thesaurus was used in the generation of further associations that generated links. These links were followed not only to obtain definitions, visual representations but also as a means for generating further associations that in their turn generated even further links and associations. These associations were frequently tossed back into the online thesaurus: The online thesaurus works in a character associative mode and thus some very strange associations were tossed back at me: Typing the word “dragonfly” led me to “tragic flaw” , or while searching for the definition of the word "Mesa" I discovered that Mesa is an acronym for "Methyl Salicylate", which is derived from the birch tree, the latin of which turned out to be "Betula Lenta". Betula Lenta, typed into the thesaurus, led me via a character association to "Beulah Land", which in turn led me to "over the rainbow", which led me to "Judy Garland", who as one of the texts in a website dedicated to her had indeed had a "tragic flaw", which led me full circle to a point that I had arrived at pursuing a completely different chain of associations that had started out with moon > earth > magna mater > cybele > dragonfly > tragic flaw. Choices were made only based upon the material that was encountered during this search, although unexpected combinations were used in preference to the more mundane associations. As an example, a google search for “modus operandi” led to a website that also had links relating to moon phases which I immediately put to good use.

    A synonym search on the online thesaurus delivered unexpected results.

    I could of course, have determined upon a completely random kickoff point by using a text generator software and I may still do this in the future. This time I did not use the internet as a kickoff point for the search but determined what that kickoff point would be myself - in a physical environment, while I was on the Bosphorus bridge herself, returning home after a long day at work. I did, of course, realize that the word "bridge" would give me multiple associations and that is why I chose it. Whether a word provided by a text generator will provide the same possibilities remains to be seen in further incarnations of this project. But once the chain of associations were underway I found out that I moved from one word to the other in astounding rapidity: "Bridge" led to bone via "bridge of nose", which was an association that I would probably have made by myself anyway. However that "bone" would lead to "process" (see left image above), which in turn would lead to "modus operandi" was entirely unexpected and it was those moments that made the search a fascinating exercise.

    As far as the word search was concerned I did not make any associations outside of the internet. I did make personal associations when it came to images and did not hesitate to use these. Thus, I "chose" to go to the webmuseum to find an image that would represent "father" or decided to look for Rembrandt's "Anatomy Lesson" as the image for "Anatomy". However, these cases are isolated and the bulk of the images used here are images encountered by an image search. Conversely, some word associations also came from links that the images were harvested from: "Planet halflife", which led to "computer games" was found while searching for images associated with "black mesa".
  • Design: Starting out, I did not know what the visual manifestation of the project would actually be. I am comfortable working with html and thus my initial assumption was that I would collate all found visual material in a hyperlinked 2D format. However, unlike most html pages, I did not want the output here to be textual but images linked to images. I also wanted the chain of associations to remain visible and accessible in their entirety at all times.

    • Space: Given that any 2D space which let me use only the x and y axes, no matter how sophisticated in its hyperlinkage, was insuficient to display the material, especially in its entirety, I turned to navigable 3D environments. Initially I decided to build a maze, one very much like a physical, architectural space. Although the maze was exciting to walk through as well as pretty to look at, and was thus a rewarding experience for the user, in terms of the project's specifications I was no better off than in 2D: I still had only the 2 axes, x and y in html, and x and z in the maze:

      The maze looked pretty and was easy to navigate but ultimately didn't work.

      Since every 3D object 6 sides I could generate 6 associations out of each block and still be able to place most of them side by side. This could only be done by utilising the x, y and z axes equally and this led me to design a structure, built of rectangular prisms, that stretches itself into all directions. I proceeded to learn VRML and started assembling the environment according to these specifications in that language.

      There was a trade off in this: By relinquishing the architectural structure; complete with floor, walls and sky, I also relinquished the ease in navigation and user friendliness that this provided. Due to its associative content, the environment started growing organically which made the creation a cohesive whole far more challenging than a preplanned structure would have. I gave up all attempts to bring method to madness. Instead I decided to go with the flow, following free association not only in content but also structure and design.

      Evolutionary growth

    • Interactivity/Navigation: Currently the environment can be walked through by using the interface of the VRML viewer, using either the mouse or the keyboard arrows. Alternatively there are animated viewpoints that can be accessed in linear sequence. Both these methods are insufficient in achieving non-linear navigation, which is integral to this project and hyperlinks that join the blocks will be undertaken in future incarnations.

    • Gestalt: All images used come from the internet. Except in a few cases, I used the images as I found them, merely adding a typographic layer containig definitions. Thus, the images cover a wide range of styles and techniques, from digital wallpapers to archival photographs and from charts to scientific illustrations.

      The images cover a wide range from archival photographs to charts

      Creating a Gestalt that would be unified as a whole, despite the discrepancy of its parts, compounded by the organic architecture, formed a major challenge [05]. I tried to solve this problem by embedding this material into replications of a single shape that then formed a geometric pattern. I tried to provide further elements that would enable a Gestalt with the manner in which I used environmental colour and type:

      • Shapes: The building block is a simple 3/4 aspect ratio horizontal rectangular prism. The representative image is mapped, front and back, on to each block. Rectangular prisms can be stacked thus enabling the associations to be placed in an efficient manner.

        Stackable geometric primitives

      • Colour: Outside of the images themselves, the colour scheme is neutral: The rectangular prisms are a flat %80 transparent white and the background a white to black gradient.

        Neutral whites, greys and black

      • Typography: I used a sans serif typeface (DIN), which is highly legible since legibility was a major issue: The text would need to be read in a 3D environment and would certainly undergo perspective distortions.

        Legibility, even with distortion and perspective.

        I also used type as a navigational aid, setting up directional axes of type between nodes that were interrelated but could not be stacked. These axes are not always fully legible, since depending upon one's location they may be partially or even wholly concealed by other objects or be under unfavourable lighting conditions. I have however, tested them on numerous users and found out that they do eventually manifest themselves and prove to be a valid aid.

        Type as a navigational aid.

04// Future Work: Hardware restrictions aside, I do not envision a particular time when this project will be completed or indeed that I should be the only one working on it. As long as the interest in it holds up it can be endlessly tweaked - by me or by others. It can keep on stretching itself into its three axes and new chains of associations can be generated from the existent ones. I have only followed a few paths and walked down very few of the multiple branches that opened up to me. Many more are left open ended to be gone back to at later dates, not just by me but maybe by other users as well. So I will continue to add modules to this version of the project as well as take steps to establish a community of contributors.

I will also be working on new versions of the environment that will investigate hyperlinks, layering of modules within modules, using transparency and enriched zooming capability and also modules that contain motion graphics/video or embedded interactivity.

I intend to investigate organic structures in both biology and architecture in the hope of revealing parallels between these and the structures of virtual 3D spaces. I specifically plan to examine the Topkapý Palace, which is a stunning example of unstructured, evolutionary growth in architecture, having evolved over 4 centuries by additions that did not progress along a predefined plan but were added on as and when needed [06]. In doing this my objective will be to establish analogies and crossovers between the organic architecture of the palace and virtual 3D spaces such as mine since both seem to grow with the passage of time with no preset plans or guidelines and whether lessons from the former may not be utilised in the creative processes of the latter. Yet another area that I wish to carry out this research is the investigation of biological/rhizomatic growth, which unlike the palace, does progress along a predefined path but may give the appearance of haphazard, random growth nonetheless.

Finally, as I put this project together I encountered Gestalt theories, both in relation to Psychology/Cognitive Science and Art/Architecture Theory. These have interested me a great deal and that I feel that a deeper knowledge in this field may benefit this project as well as future research. Consequently I intend to embark upon further reading in Gestalt theory in the months to come.

05// References
01] Matthews, J. H. “The Imagery of Surrealism”. Contributors: Syracuse University Press. Syracuse, NY. 1977. P: 4
02] Seitz, W. C. “The Art of Assemblage”. Museum of Modern Art. New York. 1961. P. 5
[03] Chen, C. Information Visualisation, Beyond the Horizon (2nd Edition). Springer Verlag. London. 2004. P: 22
[04] http://www.deviantart.com/deviation/3416915/ Retrieved on 12/02/2006.
[05] Smith, P. F. "The Dynamics of Delight: Architecture and Aesthetics". Routledge. New York. 2003. P: 5, 167
[06] Necipoglu-Kafadar, G. "Plans and Models in 15th- and 16th-Century Ottoman Architectural Practice". Journal of the Society of Architectural Historians, Vol. 45, No. 3 (Sep., 1986) , pp. 224-243