Reading a System

1. The messy room

Stress is not dirt. It is the absence of mastery.

When I am asked to walk into someone’s messy room to clean it, I tense up. It is not the laundry on the floor. It is that I do not yet have mastery of the system. By mastery I do not mean deciding everything. I mean understanding the system enough to know what my gesture changes in it. As long as that grammar does not exist, every local gesture is a blow in the dark.

Very likely because of certain elements of my neurodivergence: walking into a system I do not own stresses me before the work even starts. School made me pay for that. It grades the micro: the exercise, the page, the task. It almost never requires the system to make sense. I always approached the problem from the top. That was taken for indiscipline, sometimes for arrogance. It was a matter of cognitive survival. Without a map, I do not start.

Case — the room

Two people receive the same instruction: “tidy this room.” The first folds what sticks out, takes out what smells, lines up what shows. After an hour the room looks clean and nothing has a place, but above all there is no system in place that will prevent the clutter in the future. The second stops first: what lives here, what circulates, what must be findable with one’s eyes closed. She loses twenty minutes. She gains a system.

The difference between the good and the mediocre, in complex work, is not energy. The mediocre settle for the task at the scale at which it was handed to them, without asking about the system. Mediocrity begins there: executing each gesture as if it belonged to no whole. But the risk for the systems thinker is to find himself paralyzed by the mess already in place: one must understand the stress and apply method in order to take back control of one’s mind, one’s body, and one’s environment.

Gesture. In front of an unknown system: stop before delivering. Draw the whole. Name what rubs. That time is not inaction. It is already work. It is the method.

2. The macro first

A well-made system is a harmonious system. One feels it before knowing why — and that is still only an inquiry.

The macro scale serves first to distinguish harmony from disharmony. One does not yet have the names of the parts. One already has a sensation of the whole — too many cogs, a hole, a force pulling against another. This reading is not a technical judgment. The harmony I am speaking of is not only an impression of elegance. It is a whole whose parts serve the same intention without contradicting one another. The sensation opens the inquiry. It does not close it.

Why start there. A heap of knots, seen up close, is always justifiable. Each micro has its excuse. It is only seen from above that the excuses cease to hold. The macro does not replace the detail. It prevents one from falling in love with a detail that serves a lie.

Case — a music video, no prep

I was replacing a cinematographer at the last minute on a music video. No prep. No meeting with the directors beforehand. The locations were not mine, nor the equipment list: I was discovering everything at once. The only choice still mine was my camera assistant. One day to shoot.

The tension was there on arrival. The two co-directors were lost. The house they wanted had not been obtained, the set had changed, the entire story had to be taken up again. The stress was not the site. It was that there was no system left.

I took the producer aside. An hour. He should go get a coffee, look after the crew: I needed to sit with the directors. We went back through all of it — their ideas, their method, the place as it was. We reappropriated the project in front of the system we had, not the one we had lost. After that the work held. We finished on time. It made sense.

Gesture. In front of the unknown system: time, if one can get it, is required. Not to deliver. To sit with those who carry the intention. To apply the method and rethink a system that does not always know itself to be corrupted.

3. Reappropriate

Even a system that works, one does not take as given. One would be working without understanding.

A functioning system has a misleading authority. It runs, so one believes one must inhabit it as it was received. One can be competent in it. One is not master of it — in the sense set out above: one does not yet know what the gesture changes. Without that reading, one will be able neither to serve it truly, nor to correct it the day the charge requires it.

To reappropriate is not to break everything at once. It is first to zoom until each step has a why. Each part, each cog, each nut. One does not leave that zoom until one can tell the system to someone who has never set foot in it, without reciting a manual. I can rebuild an entire system on paper and conclude that the existing one must be kept. What then changes is my understanding. Reconstruction in the head, the drawing on the page, and the modification in the real work are three acts. Only the third requires the charge.

Only then does one climb back to the macro. And one understands what one is looking at. Contestation becomes possible, because it is no longer a mood: it bears on a named part.

Case — the pipe of a team one joins

You are handed a bible: workflow, nomenclature, habits, tools. One applies it without reading it. The other scorns it at once. The just work is slower. One opens each box. One asks where this rule comes from, what it used to prevent, what it still costs. Some parts hold. Others are scars of an old accident. When one climbs back, one knows whether the whole is an architecture or an accumulation.

Gesture. For each received rule, one sentence: “it is there because…”. If the sentence does not come, the rule is not yet yours. Do not fight it by dogma. Look for the why. Then decide.

4. The blank page

Sometimes one reads first by making things break. Each breaking point names a part.

There are systems one does not read on paper. One reads them by testing them. I arrive at a machine. I run it. Dozens of trials, deliberately different. Each trial shows a point of weakness. That is not vandalism. It is mapping. The point that gives way says what the part was supposed to hold — and what it does not hold.

From these failures, one reconstitutes the pipeline that would allow the machine to function. In parallel, one writes the mapping of an ideal system: a revision from the ground up, from the lowest common denominator, the skeleton. The blank page does not arrive in place of the test. It arrives with it. The test names the parts. The page rebuilds. At each addition, one zooms in and zooms out to check that the harmony still holds. A modification of the macro that does not entail a revision of the parts is vanity: it creates losses at the moment when there is no time left.

Case — a machine already in place

One arrives at a system one did not design. The first thing is not always to redraw it. It is sometimes to run it, again, differently, until the breaking points appear. Dozens of trials. Each failure names a part. Then the mapping: the ideal pipe, rebuilt from the ground. Fine-tuning will come later, once the lock is set. Here the test precedes the lock. It serves to read.

Gesture. Test until each weakness has a name. In parallel, a page: the system that would no longer have those weaknesses. If a part has no parent in the skeleton, it does not enter. This is for me the best way to appropriate a system, but it has a significant cost in time and energy that not every project necessarily allows.

5. Lock, then Refine

Imagining the simple architecture is cheap. Refining it is the real work.

To lock the macro, here, is to freeze the intention and the architecture of the whole — not each micro. As long as the top is moving, every hour spent on a detail is an hour one will do again. If the tests reveal an error of structure, one reopens the lock. That is expensive. That is why one sets it late enough for it to hold, early enough for the adjustment to have a frame.

The secret of a somewhat serious system is not the starting idea. It is the human time spent testing each micro-part to its minimum breaking point: the smallest change that makes it lose its function. Hours. Hours. It is not glorious. It is the craft.

A robust system has no useless part. Fewer gears, fewer breaking points — when the part only serves to reassure. But redundancy is not always fat. Even the most worked system will break somewhere one has not mapped. Not for lack of hours. Not for lack of intelligence. One does not cover every circumstance.

In cinema one previsualizes everything, at every scale, in every segment, in order to allow oneself freedom on set. One prepares for the bulk of the ordinary cases — say ninety-five percent of what should break. The remaining five percent, one equips with a parachute. A backup source. A way back. That is not furniture. It is the share of not-knowing one accepts. That chance does not exist everywhere. Some domains have no possible parachute. There, the discipline stays dry. Here, on a set, the parachute is part of the system.

Case — twenty sources, or one

One can plant twenty lamps and believe one has planned for everything. One has stacked. Each source has its local excuse. One can also keep one, deliberately mute, for the case one did not see coming. The first pile is fat. The second is a parachute. Fine-tuning is not adding the twenty-first. It is knowing which one you are holding — and accepting the loss of some, except that one.

Gesture. Write the lock of the macro on one dated page. Until it is set, no optimizing the detail. Once it is set: test each part to its minimum breaking point. If the domain allows it: name the parachute. One part, not a pile. One will always have to return at some point to the macro: anticipate the back-and-forth as much as possible in order to future-proof the system upstream.

6. The charge and the rung

This mentality is not a right of everyone over everything. It belongs to those who have the charge — and, in the short version, to those who hold a section.

The system once read and set is also occupied by people. None of them necessarily owns the whole. If everyone redraws the film, there is no film left. The blank page and the lock hold for the one who carries a whole. Below, and beside: a strict hierarchy, able to run both ways on certain objects only.

The model I know is that of cinema, modeled in certain points on the army. Leads at every level. Real transversality. Not a mass of deaf executants. The gaffer does not question the acting. He does not question the art direction of the film. He can, however, tell the cinematographer that the mood chosen on this shot no longer holds the intention of the scene. A hierarchical superior is not omniscient. The good collaborator is an ally. That does not make him a second director.

There are three levels, which do not have the same mandate. See a problem. Formulate a note. Decide a change. One may think outside one’s section. One does not intervene everywhere. One decides only within one’s charge.

As far as possible, one must avoid pure executants. One needs heads able to think a system at their scale — and able to take a direction, not to question everything, not to want to do everything themselves. One reduces the number of heads, not the level of the heads. In each section, a person who understands the micro and the macro of what they must do, and who has understood exactly the vision the rung above asked them to apply.

Case — the gaffer

A gaffer who reopens the casting is a hole in the system. A gaffer who executes a mood he knows is false without saying anything is a dead part. A gaffer who comes to me, one rung, with a reason, forces me to review my system. That is how I recognized the good technicians: they did what they were asked, without reopening everything; they saw anyway; their notes were accurate enough that I could question myself.

Not skipping a rung is not a matter of ego. It is a matter of system. Hierarchy is itself a part of the mechanism. A hierarchy founded on ego is fat. A hierarchy founded on the system is architecture. Those who do not understand this distinction will later carry a problem larger than their grade.

Example — lighting chain on a set. Not the map of an entire crew.

Understanding three hundred and sixty degrees is not a permit to remake the whole. It is the condition for occupying one’s box well. According to the rung I am placed on, I adapt the perimeter. I have been able to produce, write, direct actors, do the light, set a camera, plug in lamps. That is not a reason to do everything the moment I see everything.

Gesture. Write the four lines of one’s box: what I see, what I send up, to whom, what I decide. Reread them the morning of a project. Hold them in the evening, especially when one is right too early. Ask the question of one’s field of responsibility and above all take its full measure and importance, integrate the hierarchy of the system and of the posts.

7. Two wirings

Some people are made to follow pipes. Others to remake them. This is not a moral hierarchy.

Both are needed. Corporate and school often forget this, and they forget above all time and the power to question: people are not given the time to absorb a pipeline nor the freedom and the safety to challenge it. Without that time and that freedom, even the right profile does bad work. No one does a project from A to Z alone; he needs a team able to shake his priors. Elon Musk does not work alone: he is surrounded by extremely talented people with high IQ and EQ as he says himself, people who have emotional intelligence and not only fluid intelligence. A bad product very often starts at the top: a lead unable to delegate, to transmit, to receive feedback. At the bottom, people who were not given the time to make the pipe theirs, or who do not feel legitimized by the superior or by their post to make constructive notes.

My wiring is not that of the career technician. I was a good technician, but not the best. I was a technician who was bored, and who was unhappy. What feeds my brain is collaboration at the most macro scale I can hold. That is my reward system. If I am put in a system that is not mine, without being given the possibility of making it mine, and above all without being given the means to understand the decision and the system I must serve, I will do it badly. The problem is not that a proposal of mine is refused. It is that the reading is closed to me. The parts I judge stupid will not be done with love or with force.

The technician must still be named without contempt. Doing the micro without putting it into the macro is his work. It is also, if artificial intelligence and robotics keep their promise, the gesture they will take. The proper power of human intelligence, here, is to put systems in place. Among the technicians I trained, the difference was not cable talent. It was the one who did the micro while having a vision of the macro, an opinion on the macro, and the capacity to make discreet notes, without skipping a rung.

These two tendencies — follow, remake — can show at several levels. A gaffer can already increase a system. A lead can merely administer a pipe he has not read. Grade does not say the wiring.

Gesture. Name one’s wiring without romance. If one is made to follow, demand the time to absorb and the right to flag. If one is made to remake, do not settle into a box of pure gesture: one will be competent there, but certainly less productive than someone else and, above all, the soul will be bad there.

8. Taste and the climb

From the bottom I was a problematic technician. Not from incompetence. From taste.

Taste, here, is not a title. It is a judgment of coherence — the faculty of sensing whether the parts serve the same intention — and the capacity to adapt that judgment to a project. It makes isolated micro uninhabitable. It forces the climb. My progression was not linear. The learning itself was not optional. I had to climb fast, not from impatience for a credit, from the need to feed the reward: set electrician, then gaffer beside cinematographers who had much more experience than I did. That rung allowed me to learn their system and to be a force of increase: bring my thinking into their mechanism, not in place of theirs. When the cinematographer said no, I folded. I had no choice. That was the price of the rung.

One learns a great deal from that. Taste, confidence, mastery of a new tool can be taught. One must not burn the stages of learning. Winning and ambition, I do not believe those are taught. One has them, or one does not. If artificial intelligence and robotics keep their promise, they are a power-up for the productive: all those who can will become executives. They do not spare the climb of the rungs. They only make more visible those who have nothing but the gesture.

Case — the gaffer

I was no longer in the box of the cable alone. I was not yet in the box of the lock. I was in contact with an intelligence more formed than mine, close enough to see how it thought, far enough not to take the film from it. My notes served his system. His nos sent me back to my part. That is the rung where one learns collaboration without confusing it with the taking of power.

Gesture. If isolated micro makes you unhappy while you do it well and you understand the holistic whole, that is not a whim: it is information about the place you can hold. Look for the rung where one is close enough to a larger system to increase it, held enough to fold when the lock says no.

9. The gymnastics

One does not serve an idea at full power if one has not made it workable in one’s head.

When I am cinematographer, the film is not mine. It belongs to the director. The relation I look for is not that of a silent executant, nor that of a clandestine co-author. It is an exchange between two people of taste, two artists. When I disagree, I give a note. I explain it. It is received, or it is not.

If it is not received, it is not war and it is not half-service. It is on me to take up what the director says, and to discuss enough — not to be right, to understand. Observe. Spend time. The dead time. Agreement, disagreement, sometimes the shouting match. Once the mental gymnastics of understanding is done, I can serve the new idea with all my power. Without that conversion, I have no force to give. That is why I spend a great deal of personal time with the directors: I need their chain of thought, not only their instruction. “Explain to me how you arrived there, so that I can make your vision mine and serve you better.” That is not a shame. It is the craft.

Absorption has a tighter mechanic than one usually says. Part of it is mimicry. In my case it has never been a pose: I study the person until I sometimes speak like them, sometimes dress like them, in order to be able, unconsciously, to think like them. That is not flattery and that is not a loss of self and above all it is never conscious. Experience made me see it. It is a tool. Once the rhythm, the lexicon, the hold of the body are taken, the chain of thought ceases to be a speech one listens to. It becomes a state one can inhabit long enough to serve. It comes to me naturally. Perhaps three years of theatre in Paris refined it: the actor’s work, at bottom, is already that — entering someone else’s system far enough for the thought to stand. The danger is the same as in the theatre: staying in the character after the scene. Mimicry stops when the idea is workable. Then one serves, with one’s own force, inside the other’s vision. But one comes out grown in every case.

One may give a view outside one’s exact executive field. A cinematographer may speak of something other than light. If the director says no, it is no longer my arbitration. His no does not make my idea false. It sets the decision I must serve. I step back. I work my part. Work for work has no value. Directed, thought work has one. No one is the best everywhere. One needs delegates. The first capacity, above the tools, is to learn: a new technique, and also to change one’s vision, to adapt one’s taste to a project, to land on one’s feet thanks to systems.

Case — the director’s no

I propose a mood. He refuses. If I stop at the refusal, I serve by folding the gesture and keeping my idea as a grudge. The light will be correct and without force. If I stay in the conversation until I see what he sees, his idea ceases to be an obstacle. It becomes an object I can carry. That is the only honest way to fold. But one must also know when the time for discussion stops, and where the moment of action begins. The digestion, between the two, is not always as long as one would like. Experience compensates.

Gesture. After a no: one more conversation, enough to restate the chain of thought without caricaturing it. Then know how to stop. Discussion has an end; action begins. Digestion is not always as long as one would like — experience reduces the need for it. If needed: take the other’s rhythm, then leave the costume. Serve at full power. One comes out grown in every case.

10. The two tuning forks

The gymnastics has an end. That is not a failure. It is a criterion.

It happens that the disagreement is no longer a gap in the chain of thought. It holds to the working system. The methods do not agree. In that case one does not work together. It is not magic. Not everyone can work with everyone, under every condition. A break is not a failure. That is how one learns, and the more days of experience one has, the more one knows in advance which projects one is ready to take on.

It has often happened that I talked with a director and understood that his way of working was not compatible with mine. We were not going to create a power-up relation. The sum of the two of us was not going to make something larger than us. No harmony. Two tuning forks that do not come into unison. That is decided in the discussions, the respect, the time, the rejections, the shouting match, the questioning of oneself. Knowing how to withdraw from a project is part of the method. It is not a question of who. It is a question of how.

One fights ideas. One also fights methods. Two methods in opposition do not say that one is good and the other bad. They say a compatibility. In the crews I have worked in, neurodivergent people have often tuned better with one another than in a forced mix with ways of functioning that do not have the same relation to the system, to time, to the messy room. That is not a clinical conclusion. It is a regularity strong enough, in my craft, to stop blaming oneself when the pair does not take.

Case — the project one does not sign

Everything is seductive: the subject, the name, the money, the wish to be likable. Two or three conversations are enough to see that the other will not leave the time to absorb, or that he takes every note as an attack, or that his hierarchy is ego. One can force it. One will make a product. One will not make a better. Withdrawing here is keeping the method intact for the next lock.

Situation

Gesture

What it is not

Disagreement of idea, methods compatible

Note, gymnastics, service at full power

A war of territory

No from the rung above, after explanation

It is no longer my arbitration. I step back.

A humiliation

Incompatibility of working system

One does not associate. One withdraws.

A moral failure

Skipping a rung

One does not do it: one breaks the system

Courage

Gesture. Before signing: three conversations at least — that is my marker — enough to hear the method, not only the idea. If the tuning fork does not come, refuse. Note why. That is how the filter sharpens.

11. The sequence

What one can apply without waiting to be cinematographer.

Here is the same text, reduced to a course of action. The order of reading holds. Two branches separate according to the charge. The last lines are attitudes, not steps.

Read

1. Stop in front of the unknown system. Do not deliver. Read the harmony and the disharmony.
2. Zoom to the why of each part. Fight nothing as long as the why is missing.
3. Climb back to the macro. Only then judge.

According to the charge

4. If one has the charge: test until the weaknesses are named, then blank page, skeleton, reconstruction. Check the harmony at each addition. Lock the intention and the architecture. Fine-tune after. Find each part’s minimum breaking point. If the domain allows it: name the parachute — one part, not a pile.

5. If one does not have the charge: hold one’s section, see the whole, flag one rung up, do not skip a rung, do not remake the film.

Serve, or withdraw

6. Ask for the chain of thought. Take the time to absorb — including by mimicry, then leave it. Make the vision one’s own in order to serve at full power. One comes out grown.

7. On a no: gymnastics, then action. Know where discussion stops. On an incompatibility of method: a break without shame.

8. Learn the tool and learn to adapt one’s taste. Do not burn the learning. Winning, I do not believe it is taught.

12. What this method does not say

It does not say that everyone must think as I do. The opposite is written above: both wirings are needed. It does not say that a hierarchy is a humiliation. It says that a hierarchy of ego is a dead part, and that a hierarchy of system is the only way to make an object larger than one head. It does not say one must love disorder. It says one does not enter disorder without giving it a grammar. It does not say that all redundancy is fat: it distinguishes the pile and the parachute. It does not say that discussion has no end: experience decides where action begins.

It says one simple thing, and that is enough. One does not do good micro in a system one has not read. One does not read a system by taking it as given. One does not redraw what one does not have the charge to hold. One does not serve at full power an idea one has not understood. One does not stay in a collaboration whose methods will never make a better. The rest is time, discipline, and the refusal to work for work.