Edited by Anna Bentkowska and Hazel Gardiner

ISSN 1742-3376

Frieder Nake presents at CHArt'08

The twenty-fourth annual conference of CHArt was titled, Seeing… Vision and Perception in a Digital Culture. It was held at Birkbeck, University of London on 6 and 7 November 2008. The conference (see programme) is reviewed here by Ann Ogidi, a student on the MA programme in Digital Culture and Technology at King's College, London .

I'd heard about the conference from an email discussion group. The title, Seeing…Vision and Perception in a Digital Culture was sufficiently encompassing yet opaque and sounded like the ideal hunting ground for a Digital Visualisation essay. Then the abstracts arrived: inspired, absurd, ingenious. Of the ones I could decipher, I was looking forward to discovering why the colour of night was not inherently or invariably inky blue; that vision is not a fluid pan across a scene; 12 is not a ridiculous amount of dimensions for an image; and possibly in the near future, art will create itself. As a writer I was interested in how visual artists see the world. Do they direct the eye the way a writer manipulates attention through a narrative? If not…through what? And what does perception mean in this frantic, exponentially expanding digital world. Like those autostereogram paintings of the 1990s, I was hoping that with a bit of cross-eyed concentration, understanding might emerge.

Eyes of the machine

Professor Paul Brown opened proceedings with Art, Science and Technology in the Late Twentieth Century: A Revisionist view of a revolution that never was, a discussion of the ambiguous relationships and historic splits between art, artificial intelligence and engineering that led among other things to Ichnatowicz's Senster (1970-74). Simone Gristwood (Amalgamating Vision: Photography, Artificial Intelligence and Visual Art) posed the question whether art begins at the moment of conception or doing, a philosophical dilemma typified by Aaron, the autonomous drawing agent designed by Harold Cohen.

Birgitta Hosea (Digital Synaesthesia: Hearing Colour/Seeing Sound/Visualising Gesture) explored the phenomenology of perception with examples from her own work; kaleidoscopic sound and vision performances seemingly free of human agency, discovering its own digital materiality. Apophenia: seeing connections where none exists; Gestalt: seeing patterns; and pareidolia: seeing or hearing meaningful things from random data, are visual compulsions that help us construct the world. This inspired the installation by Alan Dunning and Paul Woodrow: Seeing Things – The Ghost in the Machine. Random, computerised visual data (pink noise) is projected onto a screen and sequenced again as an audio stream. Then face and voice recognition software analyses the data for human symmetry, i.e. words or facial features. What was powerfully evident is that patterns once seen, like a face on Mars, is hard to unsee and that seeing also depends on context. The same installation is experienced differently in a prison, an asylum or in a different language. James Faure Walker (Machines, Drawing and Vision) considers the pencil now too blunt an instrument to record the rapid flow of life in an urban centre like Leicester Square. In the digital context where photos, lines and forms can be manipulated, documentary evidence is unreliable.

Technology can fill the gaps of an imperfectly discovered world. Luciana Bordoni's presentation: GIS and WebGIS Technologies for Enhanced Seeing in Archaeology. The Case of the Roman Aqueducts, created 3D objects from fragments and mapped the invisible highways of the Roman-era distribution of aqueducts. While Paul Edward Scattergood (Subject to Change without Notice: How Advances in Modern Holography and Digital Imaging Have Altered our Understanding of Vision and Perception) presented a vision enhancing technology emerging from engineering and industrial laboratories into the art space. Stuart G. English (Creative Perception; Sensory, Conceptual and Relational Ways of Seeing) reflected on the multiple ways of seeing and how this can be used to model the ideal office environment, free of creatively limiting mental constructs.

Aaron, Self-portrait	Alan Dunning and Paul Woodrow, Ghost in the Machine	Rune Peitersen, Saccadic Sightings

Picture imperfect

Night is unsettling: it draws colour away and warps time, but does this mean that an object has two bodies? Do we need a language of the night? Eduardo Abrantes (Night-Coloured-Eye: Night Vision in Video or the Mediated Perception of Invisibility) showed the visual spectrum of night: the green tint of video and CCTV, infrared of heat sensor cameras, dark blue of filmic skies and its orange equivalent in Eastern cinema. For me, it was a moment of revelation. What I took as time-based deterioration of a theatre recording, as I tried to see through the image to the action beyond was not simply a technical flaw, but a surface that communicated its own reality. David Crawford (Realism vs Reality TV in the War on Terror: Artworks and Models of Interpretation) challenged photorealism and representation with examples from Jeff Wall, Gillian Wearing, Banksy and Pierre Huyghe. A new type of war, the war on civilians, such as the 7/7 attacks in London, is experienced via mobile phone images taken inside bomb-blasted trains to the We're Not Afraid project. David Humphrey's Seeing What You Believe, Believing What You See: Revisiting 'Photorealism' further explored the authority of the digital image, now so persuasive as to be indistinguishable from a photograph, but what the image omits is significant - the gaps between key framing and tweening.

The eye does not see in continuous motion. It constructs a picture through saccadic sampling, fast ballistic movements of light shooting erratically across a scene in fractions of seconds. Ian Gilchrist and Catherine Baker (Scanpath) presented on the science and art of vision. While Gilchrist addresses muscle movements and biology, Baker uses the saccadic process to create art - like a thousand crystals suspended in space and asymmetrical, geometric lines drawn instantaneously and continuously. Our movements in space are all potential drawings. Rune Peitersen, in a presentation of Saccadic Sightings, reflecting on the process of working with a MobileEye and on the difficulty of visualising sensory experience, attached eye cam and scene cam to his head in an attempt to record his own saccadic movements. He found that it was impossible to train the eye, it moves too rapidly and the stream of visual data revealed distraction as well as inner logic. Does the eye convey messages?

The evolution of the eye

Ada Henskens (Perception and Representation: the Visual Cortex and Landscape Art) also experiments with ‘eye pictures' or rather brain pictures as she visualises the motion of light particles passing through the visual cortex and bombarding the brain. She set out an interesting theory that vision is evolutionary. Drawing on examples from classical landscapes, abstract art and other cultures she showed that where eye scanning may once have been linked to survival, the basic visual forms are being re-engineered in our current culture, dominated by flux, fragmentation and speeded-up vision.

Neural Darwinism was also proposed by G. Brett Phares (Attentional Surplus: Ambient Media Art and the Myth of Looking) as a reaction to our visually over-saturated, speeded-up environment. A consequence of this defensive brain process is an increasing blindness to the real world, or as Paul Virilio hypothesised, the right not to see. In a further paradox, the news we see on the Internet is converging across continents as shown by the Marumishi Newsmap. Mass narcissism obscures our individual and collective abilities to attend to the world. Phares introduced ambient media art, pieces that drift in and out of our awareness such as a drawing on a bridge or the work of Olafur Eliasson's Weather Project (2003), as well as awareness tools that wait unobtrusively for human interaction. The connection with subliminal art was noted, but the difference is that the former is perceived as benign, foregrounding our attention - not the image - and unlocking opportunities for individual growth and social discourse.

Speeding up images in the digital landscape, Dirk de Bruyn's Play it again, SAM, brings into conflict two parallel memory systems: SAM, the situated, accessible memory of computer games and subjective thinking versus VAM, verbally accessible memory characterised by linear processing and verbal recall. This underlies the so-called digital native and digital immigrant divide. In the high-bandwidth digital age, SAM is more dominant. However, a third position is possible, of simultaneous and parallel, nomadic, disassociated and dynamic memory.

Godard's actors running through the Louvre (Bande A Part, 1964) and Martin's Creed's current exhibition: Work No 850 (museum sprints) at Tate Britain, is a metaphor for the tendency of the digital database to exceed the possibility of experience, argued Jennifer Steetskamp in Configuration of the Unseen: Installation Art and Information Overload. Just as in a museum, it is not possible to see everything in a database and the narrative logic of beginning, middle and end has been suspended. Referring to examples from modern cinema like The Matrix (Wachowski brothers, 1999), Memento (Nolan, 2000) and installations such as T-Visionarium, Steetskamp deconstructs the database structures associated with the interactive archive. The image and the logic of selection can seem at once random, overloaded and infinite in its attempts to reconceptualise and reassemble itself in the twenty-first century.

But what is deep storage storytelling? How to present knowledge as a visible shape? In the digital domain, with its new paradigms of instantiation, interactivity, matrix and media flow, Wolfgang Strauss, Performing the Archive: Building Knowledge Space, forces a designer into novel strategies combining text, image and sound navigation such as the Digital Sparks Matrix from his own institute, Fraunhofer IAIS.

Is it art?

Throughout the conference as speakers re-defined notions of perception and seeing – the what and the how. The activity of art was also challenged by the shape-shifting ability of digital technology: Aaron signing its/his pictures, performances created by software programmes, automatic eye drawings. Janez Strehovec (Not-Just-Seeing, Not-Just-Reading), on the perception and cognition of digital literature, proposed Victor Schlovsky's theory of art as technique: 'a way of experiencing the artfulness of the object'. If you can manipulate, possess, touch, or grasp a word physically, it is art. Digital poetry such as Brian Kim Stefan's e-poem, The Dreamlife of Letters and Camille Utterbach's Text Rain forces a hybrid form of perception, or a techno-shaped perception. These are dazzling pieces; however they are still poems written by poets. The software defamiliarises the content and brings complex associations of physics but possibly not meaning; you don't read the poems, you see them.

Jussi Parikki, in Software Viruses: The Biennale.py Net Art Virus, and Søren Pold, in The (In)Visibility of Digital Images, looked behind the digital canvas to the algorithm. What Parikki proposed was astonishing and disturbing – that software viruses, visible through their effects: melted screens, crashing systems, deletion of files… is itself art, drawing on aesthetic-political agendas of network cultures. In this universe, survival is an artform and technology a battle-ground between gatekeepers and mavericks. Pold was also interested in the mutability of art: behind an image lies the potential of other images. It is difficult to say why the animated pie charts and graphs of Antoine Schmitt are funny, or why Christophe Bruno's, fascinum, a visual equivalent of the Marumishi news map is so entrancing, perhaps because we see a reflection of the absurd coincidences and adjustments of activity. In Christophe Bruno's Hallucination a software agent continuously scans the net for examples of copyright infringement, and fires off cease and desist letters, but the logos are hallucinatory, the software has gone mad.

Applied science

Graham McAllister, in Seeing in 3D: New Problems in Accessibility, offered this sobering challenge: imagine you are a blind person, how would you navigate Second Life? The Internet is supposed to be the great liberating influence yet 2D is not solved and the 3D net has complicated the situation. Manufacturers such as IBM are working on hardware solutions for sound and haptic devices with, so far, limited success. New ways of conceptualising the problem are required.

Art as an application was the topic of Dolores A. Steinman's Medical Imaging in the Digital Age: Fusing the Real and the Imagined. Where the body and its processes were once drawn by artists, now biomedical simulations - digital visualisations of the inner workings of the body - are created by computer scientists and engineers working to mathematical models. What matters is not the picture but the information conveyed - a wrong interpretation of the data can have serious consequences.

Light, psychology, art, space, physics, the image and the after image - Carinna Parraman's The Art and Science of Colour: Bridging the Gap between Art and Perception dissected the science of colour and the visual system and how these have been calculated across two centuries by Goethe and Edward Lande's Retinex theory. It seems that what we perceive, how the brain elaborates the physical signals, is increasingly amenable to computational force.

Frieder Nake, the famous and much quoted pioneer of digital art, in Behind the Canvas, an Algorithmic Space. Reflections on Digital Art delivered an exhilarating closing presentation. The digital image has two faces, the subface and the surface held together by filaments of code. If you break the resistance you create the possibility of art (see Manfred Mohr's examples of algorithmic movements). From two to twelve dimensions, with computers we can see through a painting and change the world.

References

Ann Ogidi, MA Digital Culture and Technology,
Centre for Computing in the Humanities, King's College London, UK

I simply don't understand how a man with such a scientific mind as yours, Leonardo, can be so unsypathetic to computers.A drawing by an unknown CHArt member, CHArt Newsletter 9, Autumn 1988

These notes are based on back issues of the CHArt Newsletter. The scope of these grey brochures, edited for a number of years by Jean Miles, is remarkable. As the interest in the history of arts computing grows, CHArt newsletters published in the 1980s are likely to become a researcher's treasure trove. They should be digitised! Apart from the regular content consisting of reviews of seminars and conferences, and announcements of forthcoming activities, the newsletter also aspired to reflect and advise on current and emerging developments in humanities computing. It was CHArt’s only publication and included occasional academic papers, presented at CHArt conferences or elsewhere, as well as book reviews. Here is the overview of the subjects covered in CHArt Newsletter 9 published in-house in Autumn 1988.

A variety of CHArt's activities and non-CHArt events concerned with the use of computers in art practice, art documentation and art studies, all held in the Autumn of 1988 are covered. Worth noting is a record, probably one of few available, of the exhibition titled Art and Computers, held at the Cleveland Gallery in Middlesbrough, UK, between August 20 and September 17, 1988. The show was advertised in the calendar published in the Computer Graphics Forum, 7(3) p. 241, as 'a major exhibition selected from a national submission and by invitation from artists working in Great Britain and Overseas'. It was a touring exhibition. After four weeks it moved from Middlesbrough to Aberdeen and Bradford, and, possibly to Utrecht as part of the International Symposium on Electronic Art. 'The work was interestingly varied.' – reports Jean Miles for CHArt – ‘Some artists had actually programmed computer graphic sequences with which spectators could communicate by pressing buttons on a 'mouse'. One used the Apple system Hypercard to invite the spectator's own decisions. Others had produced three-dimensional, 'real-world' works of art either generated or in some way controlled by computer. There is no mention of the artists' names and no further information about their art. It is the novelty of the technology, and the ability of the medium to engage the spectator, that were reported in this brief account.

The unease in using the art-related computing terminology, still in the process of formulation, is characteristic of the late 1980s. In his review of the International Conference on Terminology for Museums, organised by the UK Museum Documentation Association and the Getty Art History Information Program, and held in Cambridge, UK on 18-24 September 1988, Michael Greenhalgh of the Australian National University in Canberra, notes ‘two very distinct' approaches. One, aiming to develop universal and hierarchical thesauri, the other restricted in scope to a discrete project. The criticism of the former brought forward the difficulty, if not impossibility of the task, as well as the fear of control such a large and over-arching thesaurus would impose over the use of language. ‘[…] these two approaches offer alternative philosophies […] they are mutually exclusive, and each seems to the other to place the cart before the horse.' Greenhalgh stresses the need for any subsequent conference on the subject of terminology for art documentation to look at user interfaces, ‘that is active computer programs that sit on top of databases, […] because such interfaces can be all the end-user need know about what happens underneath – whether it be comprehensive thesauri, simple word-lists or indeed natural language.' He concludes in his usual polemical style (which will be familiar to his colleagues) by recording a comment from the floor: ‘Linnaeus created a taxonomy, but he didn't have to live with it.'

Two areas where art-historical computing was at its most advanced in the late 1980s were concerned with the application of computer graphics to the study of pictorial arts. One was the use of three-dimensional modelling to explore the theory and practice of Renaissance geometrical perspective; the other direction in the application of computer graphics was the development of scanning hardware and software suitable for producing high-resolution images of Old Master paintings. These were also the applications most difficult to access and apply routinely in art-historical research and teaching; these continued to rely on photographic prints and 35 mm slides. Very few researchers had access to computers, and even fewer were able to experience digital images and three-dimensional modelling first hand. Early in 1988 the Hayward Gallery at the South Bank Centre in London held an exhibition of Leonardo da Vinci in which the computer was used to produce graphic sequences to illustrate the artist's ideas as expressed in his writings and drawings. The principal investigator was Philip Steadman, then of the Centre for Configurational Studies at the Open University, UK. Extracts of his catalogue essay were reproduced by CHArt. He explains the advantage of using the computer to model the artist's ‘visual science': his deep concerns with geometrical and structural order in nature. ‘Unlike photography which records surface patterns of light and colour directly, computer graphics techniques build realistic images of the world starting from geometrical and physical first principles. The first step in making such images is to construct numerical ‘three-dimensional' models of the objects which make up the scene to be depicted, by specifying the co-ordinates in space of points on these objects. […] The next step is to decide the position of the viewer and picture plane relative to the scene, and have the computer construct a perspective view by calculating the diminution of size with distance and determining which parts are visible and which hidden.[…] The last step is to decide the positions of lights in the scene – either sun or skylight, or artificial light sources. The computer then calculates the effects of this illumination on the appearances of objects in the scene. […] One technique which is now widely used for computing such optical effects is that of ‘ray tracing'. […] Once a three-dimensional mathematical model of a scene is built, it is possible to compute perspective views from different viewpoints by reference to the same model.' This description is a reminder of the intensity of the conceptual computational process that had to precede the displayed image. Twenty years on, those who model today enjoy the benefits of GUIs and WYSIWYG (Graphic User Interfaces and ‘What you see is what you get') which have transformed research projects of this kind into a predominantly visual work.

In 1988, the imaging standard for a painting in the National Gallery, London, digitised by the VASARI project, allowed for its image to be displayed on ‘a matrix of nearly half-a-million pixels (768 x 575, to be precise).' It was necessary to store approximately 1.3 megabytes of information to achieve ‘the astonishing result', i.e. the level of resolution and colour accuracy that no colour photography could match. The background to the VASARI project, or Visual Art System for Archiving and Retrieval of Images, which ran from 1989-1992, is given by Kirk Martinez, interviewed for CHArt by Jean Miles. The interview is titled 'Lecturer in what?', refering to the uniqueness of the post of Lecturer in Computing and the History of Art introduced jointly by Birkbeck and University College, University of London. Martinez also reports on a number of other projects, both in the UK and overseas, including progress with Will Vaughan's unique Morelli system for automated matching of forms in pictures. This pioneering content-based retrieval system, translated around this time by Martinez from BBC BASIC into the C programming language, stored – in the stark contrast to the 1.3 MB required by a VASARI image – only 36 bytes of information for each picture.

There is also a report by Michael Good on the progress with computerisation of indices to Pevsner's Buildings of England series which aimed to unlock its contents by facilitating searching through the vast amount of historical, artistic and topographical information. Envisaging the project to take at least five years and generate about 6,000 records per volume, the concern how to store a database of some 200 MB preoccupied the author in 1988. Today, the revised index is available for purchase, a tribute to its creator's dedication and persistance, and a testimony to the lasting success of his undertaking, so very rare with early computer projects in art documentation.

Anna Bentkowska-Kafel, CHArt