This article was written in January 2002 for a book of essays which a friend
    was co-producing. It was never published, for various uninteresting reasons.
    The first sentence is quite good, but it tails off rapidly after that.


It is a commonly-held belief that we see the world with our eyes,
not Geiger counters.  The world around us looks analogue.
We observe smooth changes in sunlight as clouds pass overhead,
we donít see pixellated, jaggy road-markings if we tilt our heads,
and we can even see individual grains of salt - they donít simply
disappear into thin air due to quantization phenomena. Current scientific
models of the universe muddy the waters a little.  We donít know if the
light entering our eyes is a continuous wave-like entity or composed 
of discrete quanta. The answer, of course, is that it is neither.  
Between the eyeball and brain, finitely many rod and cone receptors 
convert light into electric currents, which in turn are thresholded
within the brain to produce binary fire/donít fire messages at the
synapses. At the other end of the chain, we see that a painting, if
viewed closely, is itself quantized by the coarse fibres of the canvas,
and the image on your computer monitor is merely a grid of glowing
phosphor dots.  At this level, the discussion rapidly become uninteresting,
because, if we agree that atoms are simply very small tennis balls,
then we come to the conclusion that everything is digital, end of story. 
However, things are more interesting if we concentrate on the notions
of analogue and digital at the human scale.


Digital art, produced using the computer, and whose natural habitat
is the computer monitor, is more important to us emotionally if it
exists as a physical object that can be imported into our lives and
surroundings. Once we become the proud owner of a painting or print,
we welcome it into the family by putting an expensive frame around it,
arranging the lighting to illuminate it nicely, and we always make sure
itís hanging straight and dusted regularly! Compare this with your screen -
when did you last wipe off the greasy finger-marks in order to view the
images on an art website more clearly?

Art presented digitally is "open" in the sense that sweeping variations
can be made quickly and with ease using powerful software. Viewing art 
on the computer screen, we are left with the nagging feeling that the
artist might become dissatisfied with the current incarnation, and so
make some changes and present a rather different piece the following day.
I myself have been guilty of doing this, albeit on just the one occasion
so far, when I uploaded an image to my website, hated it within hours,
and promptly changed all the colours and put up the new version. I didnít
like doing that, and felt bad about re-editing something which was supposed
to be final.

Art presented in analogue form, such as on paper, canvas or in clay is "closed"
in the sense that modifications are limited to time-consuming, low-power
processes such as small retouches and local corrections. If you wanted
to make a gross modification such as giving the Mona Lisa a perm and blue
rinse, you could do this quite straightforwardly once in the digital domain.
The real problems lie in the analogue-to-digital and digital-to-analogue
conversions which must take place at either end of the digital processing.
It would be highly nontrivial to digitise the painting with sufficient 
integrity so as to capture every nuance of every brushstroke. Paintings
are actually three-dimensional due to the varying thickness of the paint,
so flatbed scanning or digital photography is not enough! Even supposing
we succeed here, the digital-to-analogue conversion would be just as hard - 
how do you actually get the three-dimensional brushstrokes onto the canvas
from your digitally processed image?

People like closed art. It represents an ending, a resolution of many
possible paths. We like the reassurance that the artist has deemed the
piece sufficiently complete and correct to release it for public scrutiny.
Thus, the digital-to-analogue transition is important to the digital artist
because it requires a threshold to be crossed. This threshold is multifaceted,
consisting of monetary cost, time cost, distribution cost and psychological
risk ("Is it really good enough now to commit to paper?"). Collectively
and individually, we place a high value on thresholds being crossed, on
irreversible decisions being made. We respect strong politicians who donít
make U-turns. Company CEOs are paid big bucks for making irreversible,
far-reaching decisions. Works of art open to fast, endless variations and
mutations, either by the artist, or even by the public in interactive pieces,
are of lower value. Human life is finite, so we have to make decisions.
Thereís no value in keeping all our options open forever, because we 
wonít achieve anything at all.  The final realisation of an artwork 
as a heavily-thresholded, irreversible, immutable object is a metaphor for that.

Destroy the Mona Lisa

Consider the following hypothetical actions.

Action: Destroy the Mona Lisa.

The Mona Lisa is a unique physical object. The paint, colours and brushstrokes
are not to be found anywhere else.  To destroy it would cause psychological loss,
monetary loss and a loss to society. 

Action: Delete a computer image file of the Mona Lisa.

The computer file is a digital representation of the real Mona Lisa, and a poor
one at that; the painting has either been photographed with a digital camera,
or an analogue photograph has been scanned. Deleting the file does not delete
the original, so we feel indifferent about this action, and suffer no losses.

I put it to you that we should treat digital art equally.  If the digital master
of an important digital art piece is destroyed, then that represents a loss. 
Conversely, if a representation of the digital art is destroyed, even if that
representation takes the form of a oil painting on canvas with an ornate frame
and subdued lighting in a respectable gallery, then no real loss has occurred.
The catch, of course, comes in the word "important": can a digital art piece 
ever be considered important in the same way that the Mona Lisa is? This is 
determined by society. Humans value humanity. This is partly because humans 
take a long time to produce intricate, detailed, fine art, and we have to put
a price on the artistís time. Another reason is psychological; we experience 
feelings of awe, pride and benevolence in owning a hand-crafted piece of art, 
and these feelings are perhaps amplified if the artist is poor, ill, or lives 
in a foreign land.

Digital art is at a disadvantage because it is often hard to tell how much 
effort the artist has expended in producing the piece. Impressive-looking
effects might be canned options in powerful software which take seconds to apply.
Using the computer, images can be easily rescaled, so we canít judge how long 
a piece took to produce by its size either. As for psychological factors, the 
very fact that a digital art piece was produced on a computer immediately conjures 
up unfortunate images of the well-educated computer cognoscenti, who certainly 
arenít poor because they can afford a computer in the first place! Furthermore,
we know that in producing the piece, the artist didnít get their hands dirty 
nor paint on their clothes. How can digital art express the humanity contained 
within it?

The Digital Process

The output of a digital art piece has to be of human scale, but the digital 
process doesnít. Digital manipulation is enormously powerful, far beyond what 
all the people on the planet working in unison, 18 hours a day, for their entire 
lives, could achieve without electronic machines. Creating a piece of art is a 
process with many inputs, many processes, and one output. Analogue inputs could
include photographs, paint, electricity, and feelings. Digital inputs could 
include computer code, algorithms, as well as digital representations of text, 
sound and images. The output is always analogue. This must be so for it to be 
perceived by our eyes. Analogue processes are actions such as mixing paints, 
moulding clay or welding metal.  Digital processes must involve a machine which 
operates on digits, which we might as well term a "computer". Digital processes 
are therefore mathematical, and range in complexity from simple linear filters 
to highly non-linear algorithms, such as calculating approximations to fractals, 
a highly fashionable art movement in the 1980s. By way of definition, for a piece
of art to be considered digital, all processes used in its realisation must 
themselves be digital.

The Glitch Practitioner

Glitches occur in computers due to  bugs in software or hardware. Memory gets 
corrupted by stray pointers.  Software crashes when the operating system runs 
out of scarce resources such as memory, CPU cycles or stack space. Such glitches
sometimes cause garbled patterns to appear on the screen.  The glitch aesthetic
seeks to select regions of interest from this often very rich raw material input, 
digitally manipulate it, and produce images which are pleasing to the artist.

In the simplest case, the video memory acts like a fixed peephole into the 
computerís inner workings, so that the images you see have already been pre-selected
by the computer. With modest software coding skills, however, it is possible to
make this video memory into a movable viewfinder which can zoom in on any region
of memory, and render the data it finds there visually. With this methodology,
the full structure of glitches can be revealed. The glitch artist (for want 
of a better phrase) thus assumes a role akin to that of a photographer,
exploring the environment, waiting for interesting events to happen, and 
capturing the image before it disappears.

To the traditional digital artist, the computer is heavily involved in the 
processing of the input materials. The high-power processing capabilities 
are emphasised in the work, and we marvel over the whizzy effects with 
transparency, morphing, 3D-rendering and the like. These are the artists 
who produce the hyper-realistic life-forms in Jurassic Park, for example.
The glitch artist is using the computer as an input to the art, primarily 
as a source of glitches, but also as a tool, a software camera to explore the 
environment. What are the processes used by the glitch practitioner? This is 
the artistís decision. In my own work, I only use only two types of process. 
The first is the process of selection. Itís as simple as it sounds, cropping 
an image to focus on a region of maximal aesthetic interest.  The 
second process is a sequence of global, injective transformations of colours, 
to my own taste.  Injectivity, a mathematical concept, means here that any two
different colours in the original image must remain different in the transformed
image. This ensures that no structural information in the image gets lost; the
image remains "essentially the same". The point is, the computer is being used
in the processing stage in a rather low-power way, namely, just changing colours
in the original image!  In fact, itís such a simple process that it could be 
performed on a physical printout without the need for a computer.

So, the glitch artist is a patient explorer and photographer who uses simple 
processing of raw images generated by the computer.  The finished work doesnít 
contain fingerprints or hand-painted brush strokes as certification of human 
endeavour, but it does have a digital integrity which can be enjoyed for what it is.