Information theory: Is the light on or off

The following is a little thought experiment. It is not meant to answer any questions but to examine capabilities of cyber warfare from a metaphorical position (perhaps open). Using the principles of Schrodinger’s Cat (is the cat alive or dead) without harming any cats we ask a slightly different question. Is the electric light in a room on or off? Of course you could enter the room and look. Maybe you could peek under the door, or perhaps through a window. But, for our purposes let’s say the room with the light on is on the other side of the earth and you have to determine the condition of the light using other than direct observation methods.

If you examine the complexity of a light being on or off you might jump to the conclusion that the chances are 50-50 that the light is on or off. You might say that there is no way to know for sure if the light is on or off. The idea is to see if it is possible to reason if the light is on or not (sureness isn’t being measured). As far as guessing we can inject a few variables such as the amount of daylight in a given day, the occupants work habits, the weather, cost of keeping a light on, and more to help reason the likelihood that the light is on. Each datum or point if accurate can help us draw conclusions about whether the light is on or off.

What does this light on/light off question have to do with cyber warfare? It is an inherently non-cyber controlled device (though that is changing), but more importantly it helps us look at a piece of information regardless of the computers that might or might not be used. We are not looking for the sure answer that so often eludes us, but we are looking to see the flexibility inherent in the exercise of thinking. I have seen a variety of these thought exercises in the literature and thought it might be nice to walk through one here just for the fun of it. In the end the state of the light is simply a piece of information.

Since we can’t see into the room we might use other methods other than direct observation to ascertain the state of the light. The use of infrared imaging might give some indication of whether a light is on or off. Perhaps simply looking at a power meter might help ascertain if the light is on or off. In fact we could remotely watch the power consumption of the power meter and attempt to infer when certain activities occurred within the room. If thousands of watts of power is suddenly being drawn it might be a microwave, or if an entity enters the room and with seconds a hundred watts or thereabouts turns on it might be a light.

Looking at that usage remotely we might infer a series of patterns of electric usage and know that over time when certain spikes occur a light is being turned on. Though it might be hard to say for certain when a specific light is turned on. If we know the wattage of the light and all of the other lights a set of smaller lights might create a set of lights where the wattage of the specific light is only one. Thus, we would know exactly when the light was turned on by remote manipulation of the power meter (sensing is often unprotected on these meters). The wattage of no other device should be the same as the light in question, but neither should the addition of any multiple devices add up to the same as the target light.

All of this allows us to make an educated guess about the light. How many other ways can you derive the status of the light without breaking the rules? We have departed from the original structure of a “cat in a box” but our principles of unknown observation remain. I don’t want to mess up what might be a good point of discussion, but think about the problem and post in the comments if you have anything to add.

7 comments for “Information theory: Is the light on or off

Leave a Reply