The interaction between human and software via intermediaries is a crucial issue when examining the artist’s relation to the computer. These physical and digital interfaces enable the artist to work on the visual appearance of data. Yet because this digital image is always at one remove from the artist’s physical existence, they can never work in this medium as they would on a physical material.
The computer artist, operating at one remove, instructs the computer via gestures and the keyboard. His plans are executed by the computer in a semi-autonomous manner, with the machine taking care of the laborious calculations and descriptions of the image. The artist surveys the results on the screen and then modifies them further, taking into account any unforeseen variations or changes that emerged from the computer process.
There are essentially two ways to create a computer image. Firstly, an artist can program the computer to generate an image, working almost as a composer with the computer as performer. Although this process has a visual outcome, it requires that the image be understood as a sequence of instructions. The principal limitation of programming Computer Art is the artist’s knowledge of the programming language and the operations they can perform. They must also consider how best to represent their art as a sequence of instructions. There is also the important factor of the chance outcomes and pseudo-randomisation which can be incorporated into the art. In most cases, the programmer is the ultimate arbiter of the final selection of images that forms the finished work.
Secondly, an artist can use the graphical user interface (GUI), working with hardware and software tools that allow them to draw directly in a digital space analogous to traditional media. The metaphorical connection varies with the software: Photoshop uses a darkroom, Illustrator connects with drawing boards, Maya effectively sculpts with solid forms. Through these analogies, artists can transfer some of their skills and understanding from physical tools into a digital setting. Yet the very structure of the graphics software imposes its own limitations. There are also restrictions on screen resolution, colour matching, sensitivity of hardware interfaces and the format of the image itself. These factors aside, the GUI has enabled more artists to use the computer because it presents them with a visual display and a direct form of interaction.
The issues of Human-Computer Interaction have become a very important topic for software interface designers, and are discussed exhaustively elsewhere.[1] I will concentrate on their implications for visual artists using GUIs as the conduit for their art. All computer artists’ approaches are closely bound up with the software they use; however, GUI users are perforce more dependent on software packages rather than custom-built solutions to their art.
For instance, John Whitney worked with programmers because he could not himself program, but the resulting software bore his artistic stamp; it was custom-made to suit his artistic approach. [ref to Cuba] With commercial packages, however, the major limitation of using the GUI is that the artist is bounded by the parameters of their chosen software: they cannot remake its image environment. By contrast, if the artist programs their own software, they can determine all the characteristics of the image, and its scope is limited only by their skills and the hardware they are using.
Artists who develop their own graphics packages may also evolve a more distinctive style than those who use ‘off-the-shelf’ software, though the software is only a support to the artist’s vision. The structure of software plays an important role in delineating the artist’s workspace. If certain parameters and tools are already in place when the artist begins a new work, then the software has already exercised a strong influence on that piece. Much depends on the range of tools available and the structuring of menu options: 3D graphics programs will obviously contain a markedly different selection of tools, with very different purposes, compared with 2D programs.
The interaction between artist and program is constrained at certain points by the hierarchical structure of the interface and its underlying metaphorical basis (see Van Dam). Because the computer is not necessarily a passive partner, its responses (and foibles) add another layer of interaction. Also, many of these constraints are offset by the malleability of the virtual canvas, the ability to erase mistakes and the practice of saving multiple versions for further experimentation.
Programmers develop a better conceptual picture of the computer’s capabilities and the framework of the software. However, artists with a thorough command of a commercial software package will develop similar abilities and may even be able to devote more time to creating than to coding. Since the variety and power of commercial graphics packages increases all the time, an artist writing his own program would undoubtedly be reinventing the wheel in many cases. The programmer’s advantage remains the ability to customise software to reflect their artistic concepts.
The roots of this disagreement are partly historical, since many artist-programmers come from the generation that had no other option but to program. Others are represented by Harold Cohen, who views programming as a discipline; they learnt it primarily for that reason. They disparage the facility of GUIs and (see Reffin-Smith) assume that the metaphorical resemblance between the screen and the canvas, reinforced by paint programs, is a false comparison. I tend to agree that the parameters of the GUI can serve to restrict artists, or at least channel them down a particular path imposed by the software. However, as I hope to show in this chapter, the more imaginative “users” have found innovative ways of working with the powerful real-time tools that are the GUI’s greatest strength.
A knowledge of programming has also been justified for its general educational potential, opening the artist up to new concepts and possibilities. Richard Wright considers that the difference between contemporary computer-using artists and earlier programmers resides in the latter’s desire to comprehend the computer more fully. This was a necessity in the days that computers were scarce and exotic; however, the GUI and the personal computer have obviated the need for such intimate knowledge. Wright now sees the debate in terms of the practical use of tools versus “the potential of expanding the range of creative stimulation and access to alternative modes of cultural production.”[2]
This need to circumscribe Computer Art and its techniques permeates most of the anti-GUI sentiment amongst artist-programmers. I am trying to show that Computer Art is not dependent on the technique an artist uses, but rather on the totality of concept, computation and result.
To recap, I proposed that the characteristics of “Computer Art” were as follows:
i) The artist manipulates information directly, without the limitations that intrinsic to a physical medium;
ii) The computer can interact with and react to the viewer after production, or even as a creative agent in its own right if programmed to do so.
Whether programming or drawing on the screen, the artist is effectively manipulating information. This is much more overt in the case of the programmer; the process is nearly invisible to a software user, because they use metaphorical “tools”. However, neither is bound by physical materials or processes, even in the case of software such as Painter which mimics the visual appearance of paints and other media. For instance, the physical boundaries of the screen are not necessarily the limits of the virtual canvas – these can be specified by the user.
The processes invoked to control the computer may have a definite end, or they might allow for open-ended, pseudo-random outcomes which cannot be foreseen by the artist. Even definite outcomes may suggest new ways of developing the image, or bring about something wholly unexpected. Thus the art emerges from the interweaving of the artist’s and the computer’s contributions. This is the vital heart of “Computer Art”: not only is the underlying process quite different from its physical artforms, but the outcomes would not have been possible without the interpolation of computer results.
Of course, the tenor of this interaction differs according to the artist’s approach. For the programmer, the results are seen over a longer period and effected by modifications to the program’s code. Their “interaction” derives from changing the program’s parameters to arrive at their intended image. By contrast, the GUI user has a direct view of the changes they are making, and alters the image in real time.
The software works at two levels:
1) It provides the environment, with its attendant limits and constraints. In this sense, it is the medium within which the art functions;
2) It forms the tools and processes at work within this environment. The software controls reaction and interaction to the artist and can lay the foundations for viewer interaction as well. But the software is most evident in the process of making the art, not its final realisation.
The use of computer-specific processes, such as erasing certain elements (a non-linear action) or inserting special effects, makes images which are the result of a computational operation. “Computer Art” is distinguished from previous artforms by the application of programs to the act of image making, not simply drawing or digitising existing work. The artist’s role in Computer Art is as initiator of processes. [Quote Franke on the “insertion of the computer into the act of making art”] Thus even pre-Computer artworks, if transformed and rearranged through the agency of a computer, can become Computer artworks, though of course wholly different from the original which continues to exist in its own right.
The format in which the art is realised is not crucial to its definition as “Computer Art”. Even if the final realisation is physical, the process of making the image has differentiated it from previous artistic media. The artist has deployed the computer, absorbed and modified the results and ended up with an image that would not have otherwise been created in another medium.
When using a physical medium, the artist creates the image’s physical form as he works – the initial choice of medium determines the output format. On the computer, by contrast, the output medium does not have to be a factor until the end of the work process. In this way, the computer divides the conceptual visual content from the final realisation. “Computer Art” is whatever you can achieve solely through the computer, images that transcend the physical limits of analogous processes such as drawing and sculpture.
It is true that something resembling the resultant Computer Art might appear in a format with which it has superficial links – perhaps a photographic montage resembling a Photoshop image. However, the underlying techniques are so different that it would be a case of convergent evolution, not a true correspondence.
A case in point is Tom Costello, a photographic artist who has chosen not to use a computer. The startling abstract photographs he makes might be taken for digital graphics, but are the result of darkroom experimentation. In this respect, he is an artist who works with photographic processes rather than a photographer.
Costello’s art involves the indirect construction of images through photographing the interaction of light and colour on objects. He utilises chemical combinations and filters in the darkroom to produce abstract forms distinguished by their deep colours and shifts of tone. Costello explained to me that such images were the expressions of his visual imagination; because he could not draw, he found ways of producing these shapes on film.
[Plate XXX Abstract photography by Tom Costello]
In the working out of the image, its interactions and changes would themselves suggest ways to develop it. Costello likened this to a dialogue, and considered that he “served” the picture by combining his own skills of composition with its emergent qualities. Most dramatically, the picture would come together at the point when he exposed the film: he said there was a tiny fraction of a second when it would be “right” to take the picture, when the “artist mind” that guided his art making had the impulse to do so.
Costello explained that for every image there were at least 25 “takes”, from which he would distil a version he was satisfied with. In the course of the process, up to a day’s work, sometimes another image might suggest itself and he would quickly set it down on paper before returning to the work in hand.
He notes all the parameters of the images and he can now duplicate them through photographic means. He consciously chose not to use a computer until he arrived at a point where he could not express the image in his mind through photography. Of course, the processes he employs give his images a very distinctive quality and produces very different outcomes from software such as Photoshop. What interests me – apart from the images’ intrinsic fascination – is the parallels one can draw with computer artists, especially the indirect nature of his technique.[3] Costello’s evident divergences from computer-using artists are also significant.
Costello consciously chose not to use a computer until he arrived at a point where he could not express the image in his mind through photography. Of course, the processes he employs give his images a very distinctive quality and produces very different outcomes from software such as Photoshop.
[1] Preece, Jenny and Keller, Laurie eds. Human-Computer Interaction (Cambridge, 1990)
[2] “The Pioneers of British Computer Animation and their Legacy” by Richard Wright. Article for Kino-Eye [website]
[3] From a conversation with Tom Costello 13-9-02
