[-empyre-] Animating Google: or Dragging a Fridge Across the Kitchen Floor

[Richard Wright]

Yello everybody,
here’s my opening gambit for the empyre…

For several years I worked as a “new media artist”, which I  
originally  thought would be a logical progression after some having  
spend some time making computer animation, digital film and video.  
Things went reasonably well and I ended up working in one way or  
another as a full time artist. But a couple of years ago I finally  
gave it up after coming to the gradual realisation that “new media”  
is an unsympathetic environment for the moving image and if I  
continued working in this field I would drift further and further  
away from forms of animation and moving image which I regard as my  
“core practice”.

The reasons for my disillusionment are connected with the way that  
“new media” has come to be dominated by ideas of the network as its  
overarching model. And as opposed to time based media, the network  
has developed (so far) as a spatialising mode of expression. By this  
I mean that in this concept of the network, everything is treated as  
happening simultaneously across the entire system. Messages travel  
from one desktop to another but we are only interested in them when  
they arrive or the pattern of their interactions, not the  
“travelling” itself. You can see the effects of this everyday, when  
you use Google for instance. Google is supposed to be a search  
engine, but the one thing you never see is any “searching”. You just  
type in a keyword, press the button and results instantly appear (or  
so we tell ourselves). The actual processing involved (which is what  
computers are really about, their transformations, exchanges,  
collisions) are compressed until they recede from human scale, out of  
sight. The only form of duration that is admitted is latency – the  
“dead time” of waiting for a request to complete, staring at the  
little disk or hourglass spinning in some kind of suspended trance,  
unable to continue with our work or understand the path of a progress  
bar which habitually gets stick at 95%.

This perhaps gives some idea of why I would argue that “new media”,  
digital media, networked media, has developed as a distinctly  
spatialising media. Along with this, in the new media art world,  
conceptualism has been adopted as the aesthetic and critical  
framework of choice. This would make sense, because a concept is an  
essentially static object (unlike an “idea”). The temporality  
produced by an artwork such as when simply watching a film, the sense  
of an articulated structure of time which unfolds around you, seems  
to have no place here. It is not even really understood (Note 1).  
Digital artists make artworks about time but not in time. In the  
online world things are constantly updating but not moving. Film  
brings you to the world but the internet brings the world to you (who  
said that?).

There is a struggle to find (to construct) a temporality for the  
machine, algorithm, the information society. Even in Lev’s famous  
book the only temporal structure in new media he admits is the loop.  
Something that I think you might agree has a very specific kind of  
structure (perhaps more akin to music?). Now I’m not claiming I am  
the first person to notice this. Recently I saw David Rodowick, the  
film theorist, quoting artist Babette Mangolte saying “why is it  
difficult for the digital image to communicate duration?” I remember  
Sean Cubitt has commented about this (note 2). There is also some  
connection with the arguments of Kittler, for whom the ascent of  
writing has imposed a spatialisation on other media, reaching its  
pinnacle with computer code which digitises and erases all other  
forms. But perhaps writing, like music, is a different case…

So for the last year or two I have been experimenting with ways of  
using software technologies, especially those most often associated  
with networks, to construct my idea of what a time based new media  
might look like. What would it look like, for instance, if Google was  
redesigned as an animation? My first attempt used a large database of  
images from random Google searches that I compiled for a previous  
project (my Mimeticon “baroque search engine” - a work that will  
probably be my last ever new media piece: http:// 
www.futurenatural.net/projects.html#mimeticon). The result was  
“10,000 Copyrighted Images”, a flicker film made by rendering 10,000  
random images, one image for every frame. I called it “crime at the  
limits of human perception”. It has a certain hypnotic, rhythmic  
quality to it. And I like the fact that if you pause the playback the  
image you see is completely unpredictable. But for me it’s only a  
token first step. (I’ll add a link to this soon)

At the moment many search engine companies are developing image  
recognition engines to provide the ability to search by visual  
similarity instead of entering keywords (in fact The Mimeticon  
project above used a simple open source image recognition engine). I  
want to see if I can use this technology, which is being aimed at  
organising databases and searching networks, to elucidate a quality  
of time. I am currently processing a film called “Psycho Similar”  
which reedits Gus Van Sant’s remake of Alfred Hitchcock’s film of  
Robert Bloch’s book based on serial killer Ed Gein who couldn’t  
separate himself from his mother. The word “reedit” doesn’t really  
capture what is going on as my software will resort or reorder every  
frame on the film according to their visual similarity (all 148,000  
of them).

Of course I could plot out these 148,000 columns of similarities  
spatially in an enormous rectangle, about one and a half kilometres  
wide, and it might be quite an attractive example of data  
visualisation (or ‘data mining’ or ‘cultural analytics’). But the  
project is to reconstruct a new sense of continuity or development,  
using visual similarity to rethread the film, perhaps against the  
algorithm’s own static logic. So far the results are curious (l’ll  
try and upload an excerpt some time). Because of Hitchcock’s use of  
long takes (and Van Sant’s copying) many of the shots are retained  
intact although sometimes reversed and with a constant nervous  
twitching as frames are substituted. But otherwise I believe the film  
looks as though it is being held back, straining to move on against  
the force of the similarities that try to keep each subsequent image  
as close to the preceding one as possible. Like pushing a heavy  
fridge across the kitchen floor that keeps getting caught on a  
protruding tile just when you think you’ve built up enough momentum  
(perhaps also a good metaphor for animators trying to work with new  
media…)


Richard

Notes (desperate attempt to cut down the length of this post):

1. What has happened to film as a time based medium in the age of the  
network is not the same as what has happened to music and (to a  
lesser extent) sound art. Music typically employs a different form of  
periodicity than film. Music is a much more repetitive artform -  
those beats, choruses, tunes, undulations. Sometimes I think there  
should be a general moratorium on music in film until we figure out a  
better way of how to use it.

2. I have come across a few attempts by other artists to address the  
network through the moving image. One strategy is to concentrate on  
the traffic of material across the network or the linkages between  
nodes as though we are taking a journey through it, transposing the  
“journey” structure of a film. Michelle Teran made a work called  
“Looking for Mr Goodbar” that just won this years transmediale.10  
award. She used Youtube to find all the videos made by people from a  
certain town in Spain. Then she organised a coach journey for the  
“audience” that travelled through the town, visiting each of the  
amateur video makers in turn. People would watch the next video on  
the coach monitor as they travelled like a mobile cinema across this  
actualised network.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mail.cofa.unsw.edu.au/pipermail/empyre/attachments/20100221/55044d4e/attachment.html