Kjetil K. Haugen
Molde University College
Let us imagine the following experiment. All modern computers, phones and tablets are equipped with a graphics card and a screen. Such a screen consists of thousands of pixels, each of which assumes one of the millions of colors in the color spectrum to create an image. If, as most are, the device is equipped also with an operating system (OS), a programmer may control the screen’s color output. Given this setup, assume we write the following simple computer program: For each pixel the screen, loop through all possible color combinations of all available colors for all pixels. In other words, Produce every possible combination of pixels.
This should be a relatively easy program to write. Even with my programming skills having long left me, I am sure that I would be able to achieve it. I have not done it, however. Why? Because I have not the time; such a program would have to present trillions upon trillions of combinations before it was finished. Thinking on the output and consequences, however, is more feasible. Even for a lazy old professor like myself.
What would happen as a result of this program’s execution is simple to foresee. Suppose a further program has enhanced the human mind to sort through all the possible pixel combinations in a blink of an eye, allowing us to eliminate the issue of time. As we performed a full enumeration, we would soon come to the realisation that the program has produced every possible, imaginable image conceivable to the human eye. Somewhere among the program’s immeasurable combinations you would catch images of your own birth, of Hitler using the bathroom, and even those of Jesus and Abraham, though you would of course have problems identifying them.
If we were to take the experiment one step further and add another out loop, collecting all possible sequences of still images and merging them to make movies, the reality runs as previously: with all possible moving images being generated. You could witness the Battle of Hastings in sparking 4K resolution.
Of course the problem with this is that the program would produce all possible images, not only images that have or have yet to occur. The program would generate every feasible occurrence, every slight nuance of reality, from the perspective of any simulated agent around the universe. Any images portraying true events would be a grain of truth buried beneath a world of falsehood. Or would they? This thought experiment raises the important question of what would make these images ‘truthful.’ Could the qualification be that they portray images of events that have actually happened, and if so how could we verify this? If we were to witness a scene of Marie Antoinette sipping at a glass of wine, could we claim any sort of truth about the image? It is unlikely that it would correspond with any visual data received by an agent alive at the time, but—even then—what if it did? Even if we were able to liken the present images to those experienced by real-life witnesses, the said agent’s memories would have long passed into nothing, into nonexistence. How could we regard these images with any notion of truth? It would be interesting to view images of familiar objects performing unfamiliar tasks, but what uses could such a gargantuan archive of visual date have?
These considerations are still somewhat obviated by the aforementioned time constraints. Even though modern computing devices are astonishingly fast and consistently getting faster, the time needed on a single computer would be entirely intractable. But if we were able to utilise all available computing power, that of all the billions of computers in existence, we must at very least be able to create a few images of interest. So, why not? I am sure some youngster out there could write the program and arrange some efficient distribution mechanism making us all a part of this great experiment. I leave it to you.