Over the past five years, I've had two major revelations about the nature of video. The first occurred in 2000. I realized that the standard way of playing video frames ("display 30 frames per second -- start at the first frame and end at the last" ) is really only an arbitrary convention. By default, a one-minute video clip is something intended to be experienced in one minute. But why not consider the one-minute clip as a collection of 900 still images? (30 frames per second times 60 seconds = 900 frames.)
Suddenly numerous possibilities appeared. Selected frames viewed as still images might reveal interesting relations when seen in proximity. The duration that each frame is displayed can vary in some pattern or other. And most importantly, the order that frames are displayed is liberated: frame selection can follow some type of pattern (eg, 1, 4, 2, 5, 3, 6...), or can be random (22, 726, 698, 204...), or can be
controlled by an external process, such as data coming from audio analysis, analysis of the video itself, etc.
The second revelation occurred in 2003. I realized that not only is the sequence of frames arbitrary; the frame itself is an entirely arbitrary convention. One minute of video can be considered to be only a collection of 900 stills. It can also be considered to be a collection of 300 million numbers: that is, the values of each pixel in the entire video. Each pixel is no longer defined by its role in its particular frame; its role and relationship to other pixels is defined only by rules that I create. Digital video, once stored in computer memory, is just a large mass of data, and this data can be handled or analyzed in countless ways.
Now the possibilities multiply infinitely. The question is no longer, "Which frame to display next?", but "How to create a frame out of this arbitrary mass of data?" I decided to consider the data as a 3-dimensional space. A frame is created by defining a surface in the 3-D space. (I came up with this strategy on my own, but later discovered that others have had the same idea. However, I do believe that some of the particular methods I'm currently using are truly unique.) I like to refer to this technique as "4D mapping".
The point of these techniques is that they allow research on temporality, on the nature of time. In standard video playback, time is identical to time as we live it. In the techniques described above, something is revealed about time: we gain a new perspective on events, wherein beginning, middle, and end are seen simultaneously. When ordinary time is removed from an event, extraordinary things are revealed. A single event can have any of an infinite number of manifestations depending on the time-angle it is viewed from.
Collectively, these time-perspectives can be called "the atemporal". Atemporality is a reality, not an illusion.
We cannot see the atemporal in life because we are bound by time. These techniques provide views of atemporality.
How to explain the significance of the desire to come to a new understanding of time? It could be an intellectual project, an epistimological inquiry into the nature of our knowledge of the world. Alternately, it could be an emotional response
to the universal experience of powerlessness before the inevitability of loss and mortality.
(Kurt Ralske, May 2005)