Black Box Theory
Thoughtful application in a world of limitless unknowns
The black-box-white-box spectrum is a meta-system: a method of organization used to classify complex systems. It is often applied to the study of computer science, engineering, design, and psychology.
A black-box model is defined by clarity of inputs and outputs, with no regard or knowledge of internal workings. An example of a black-box model would be a piece of closed-source software, or a human being.
A white-box model is a system with clarity of inputs and outputs, as well as complete clarity of internal processes. An example of a white-box model would be a deterministic algorithm you have authored.
A gray-box model is defined by a mixture of both black and white box models. Note: complex systems are made up of many subsystems.
This described spectrum, like all models, has inherent inefficiencies and harbors reductionist tendencies. For example, some types of complex systems may have elements that are constantly changing states, nearer and further from black. A gray-box model can take the form of a snapshot of a group of one black and one white-box model.
For the purposes of this article, we will think of the models in a more basic form, as to apply them more generously to our daily lives while striving for greater individual benefit.
The child who tries to open a door has to manipulate the handle (the input) so as to produce the desired movement at the latch (the output); and he has to learn how to control the one by the other without being able to see the internal mechanism that links them. In our daily lives we are confronted at every turn with systems whose internal mechanisms are not fully open to inspection, and which must be treated by the methods appropriate to the Black Box.
Ashby, W. Ross; An introduction to cybernetics
Black Box Theory is a qualitative methodology by which to view the world. This mental heuristic can be applied in many ways, making use of the above definitions to first, define the level of mechanistic insight, and secondly create action. We can also use the concept of black box theory in creating our own designs.
Here are some potential questions to keep in mind while evaluating, creating, and designing complex systems: Can this become simpler? Is there a presence of ostentacity? What features are hiding? Can they be removed? Is the form following function?
A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away.
Antoine de Saint-Exupéry
DESIGN APPLICATION OF BLACK BOX THEORY
Our designs ought to be made with the expectation that at least one person can completely (white-box) understand its internal mechanism, and the design must remain robust, impressive, and functional regardless. Obfuscation cannot, and must not be used as an element of design, as we can expect our “secret” elements to become entropically revealed.
Security
While designing secure systems, it is ideal and necessary that we imagine our systems to be white-box models. A system can be considered secure when full knowledge of its internal workings are revealed to threat actors and they still are unable to find any exploits. This means that in the presence of continual improvement, a complex system that begins as open-sourced will become stronger than a closed-source equivalent, given they are placed on a long enough timeline. In other terms, any chronically unexamined Prophet can be called a liar.
We have achieved this openness in cybersecurity with HASH encryption function SHA-32. The code is open-source (white box, completely known internal mechanism,) has clarity of inputs and outputs, and yet remains undecryptable.
Communication
In communications, there ought not be excessive pedantry. In fact, for the most effective communication, we ought to model ourselves after a white-box. That is, the goal is to have a known internal mechanism combined with clarity of inputs and outputs.
I cannot find an English word for what I wish to express. To anthropomorphize means to apply human-like characteristics to non-human entities. I am looking for a word similar: a word that means to apply computerized characteristics to humans. I will call it automationization, derived from the word automaton. It is important not to automationize people.
Until there are vast improvements in the study of the subconscious mind, we can only observe inputs and outputs of other people, with the exception of ourselves, and even then a large portion of our consciousness remains inaccessible. The staticity of the black-box-value of human beings means we must fit our designs to them, and never reverse the function and form. Only completely reveal that which is possible. A system may, a person cannot. Note: this does not mean a person should not strive for “white-box-ness”, working to reveal certain subsystems of themselves, but rather that due to inherent constraints of consciousness, complete revelation is impossible.
Therefore, a core tenet of my design philosophy:
PEOPLE ARE BLACK-BOXES. DO NOT ATTEMPT TO AUTOMATIONIZE.
Language is a form of self-organization with network effects. At a certain point in our development, we stop screaming and instead have new linguistic concepts explained to us in the same language we are learning. This cannot occur without a baseline autochthonous production of communication. In design, do before you refer. It’s taken me 12 years to figure out how to word this. Those last sentences are a message to my Grandma about why a dictionary couldn’t teach 7 year old me a new word if I didn’t already know it. This process of linguistic acquisition by internal compulsion never stops.
Recently, we are seeing a scaled reversal of form following function in communications. We are training intelligent AI tools on datasets based on things previously written by human beings, proceeding to becoming reliant on them, and therefore losing our time-honored gift of misspelling and inventing words.
It is important that Avery Magazine remains a source of information written only by humans. Grammar tools and the vast majority of editors are largely a method of deferred low self-esteem. Like all anxiety, they are gnats. A facade of freedom, at the loss of autonomy. A unique voice, with cracks, is all important.
Art
In the fine arts, there ought not be any reliance on “aura,” “hype,” or “mystique,” unless inextricably related to the work (ex. Performance Art.) Works of art ought to stand on their own, as, with time, they will. Furthermore, a secretive process is never a measure of quality. Non-intuitively, neither should time spent. Artists may choose mystery, but this should only be for external reasons, (ie. to prevent a leak of an upcoming album,) never artistic ones.
If humans did not expire, the preferable choice would be a complete allegiance to artistic experimentation. However, a successful work of art lives on much longer than a person will, so, eventually a person must choose a “stopping point,” where they feel comfortable publishing. Another referral to my design tenet: People are black boxes, a design does not have to be. Do not reverse these roles.
The black box heuristic is an effective mental model for reasoning about complexity, via a negativa. In many cases, when making certain determinations, it is not important to completely elucidate the algorithmic root. Restated, (this time, in English,) sometimes all that matters is that you get what you want. Sometimes things can simply be, and we can observe as they are. Is it useful to ask a child to explain their thoughts? Maybe, certainly eventually, however, introducing abstraction to unknowingly complex systems (a child) may be ineffective and inefficient when we are sure outputs are unsuccessful. While you are a guest at a party, enjoy the meal and the warmth of the home rather than drill on about the nutrition facts of the dish. Do not let a word about “Carbon Footprint” leave your mouth. We each learned communication, in our mother tongues, based on survival needs, far before studying grammar.
"...scientific, [personal, and] technical work is made invisible by its own success. When a machine runs efficiently, when a matter of fact is settled, one need focus only on its inputs and outputs and not on its internal complexity. Thus, paradoxically, the more science, technology, [and art] succeed, the more opaque and obscure they become."
Bruno Latour (1999). Pandora's hope: essays on the reality of science studies.
additions made by Avery Magazine in brackets.
Conclusion
Is it important to understand the psychological basis of your marketing campaign if you are ethically making money? Is it important to analyze word frequency of a poem anthology if one line made you cry? Are their credentials important if their advice made you honestly happier? How clearly do you want your doctor to tell you your cancer is terminal?
In this article, I have outlined the black-white-box spectrum and showed three examples above in topics of my interest. The possible applications of the model are limitless and I encourage you to spend time with them, seeing what sorts of ways you can find application in your own life. I’d like to hear all about it, and I hope it does you good.