Agency as a Mandelbrot Set

Every week the stack has a new mascot. The question underneath does not change: who is running the loop? Philosophical agency is the capacity to initiate action under your own standard—to be (in part) a cause, not only an effect of incentives, shame, or mimicry. A creative agency (the company) is a legal and economic instrument that concentrates skill, reputation, and process so a market can buy outcomes on purpose. Both require a sense of self. If you are reading this, you are probably plural in market, singular in standard. We all have a way of doing things. The true illusion, lately, is the belief that we must constantly *replace* how we work the moment outputs spike—as if a winning loop were a liability. Strip the noise, and a simpler question remains: what do *I* stand for? AI is not a season; it is weather. Back in 2023—only three years ago—my outreach was one thread at a time: cold calls, emails, texts to people I already knew, each one costly in focus. Today, bots scrape ICPs and APIs warm the inboxes. The same tide fills the spam folder on the other side. Our prospects are not wrong to flinch. I do not think society has settled on a single mood; I think we are hungry for craft and allergic to theater at the same time. That tension is the air we work in. Henry Ford asked how people wanted to move faster; they asked for faster horses. The Italian Renaissance argued, for centuries, about whether theology or humanism should organize the age. Steve Jobs said, “People don’t know what they want until you show it to them.” I believe that—and I still want to know *where* Ford and Jobs located their philosophical agency, not only their product sense. Read their biographies and you will find different answers. Here is the thread I care about for this essay: all three stories are really about who gets to set $c$—the starting assumption—before the iteration runs. The fractal is the cleanest image I have found for what happens next when *you* become the person who must answer that question. Being an owner and being a consumer are two postures most of us oscillate between. Four months ago I crossed hard: from employee inside someone else’s creative agency to owner of my own. The map did not change. The *operator* did. As an employee, I could run the loop beautifully and still treat the standard as borrowed. The iteration was real; the *ownership* of the bound was not. A brief could wobble, a timeline could compress, and my job was to execute with grace inside parameters I did not set. That is honorable work. It is also a different physics. As an owner, the same sequence—clarify, make, expose to feedback, adjust—carries a second question I cannot delegate: **does this stay in the set?** When scope creeps, when fear asks for novelty without purpose, when a client wants “something viral” instead of something true, the escape velocity shows up in *my* calendar, *my* payroll, *my* name on the door. I learned that in the same week I was proud of a cut and ashamed of a timeline—both true, both mine. **I say agency is a Mandelbrot set** not because spreadsheets are sacred geometry, but because the *behavior* of durable creative work rhymes with how that set is built: a trivial rule, repeated, producing outcomes that are either coherent or catastrophic—and the most interesting territory is never pure safety and never pure noise. ### Reasons the metaphor holds The iteration $f(z) = z^2 + c$ is almost insultingly small on paper. That is the point. A real agency, at its best, runs a short loop: clarify the problem, make something true to the standard, expose it to feedback, adjust. You are not reinventing a new methodology every Monday; you are re-entering the same dependable process. The outputs still surprise you, because the *client*, the *moment*, and the *medium* are new each time. Simplicity of rule, richness of result. In the set, a sequence either stays bounded or escapes toward infinity. Projects obey the same binary under messier names: *on-brief*—recognizable, shippable, defensible—or *divergent*: scope, ego, novelty for its own sake, panic. Your job as the operator is iteration management: keep the work bounded without suffocating it. Zoom the Mandelbrot in and you keep finding echoes of the whole. That is brand and character at scale: the voice in a caption should feel like the voice in a treatment; the quality bar on a one-off deliverable should feel continuous with the flagship film. The fingerprint persists; only the magnification changes. The famous pictures are not the boring interior and not the empty exterior. They are the *boundary*—the razor edge where order and chaos negotiate. Great creative work lives there too: not so safe that it could be anyone’s template, not so wild that the audience has nothing to hold. The boundary is where the detail explodes. Finally, $c$ is just a starting point in the plane. Tiny changes in $c$ flip membership entirely. That is the brief, the positioning call, the first visual direction. Garbage in does not guarantee garbage out—it guarantees *unpredictable* out, which is worse for a business that has to ship. Sensitivity to initial conditions is not mysticism; it is leverage. Name it carefully. ### Where the metaphor breaks (and why that matters) No client is a complex number. People change their minds, lie kindly, discover budget, fall in love with the wrong round of revisions. Escape to infinity is not always failure—sometimes a project *should* blow up early so you do not marry the wrong strategy. And unlike the set, you can often *re-pick* $c$: a new brief, a reset conversation, a braver edit. So the Mandelbrot is not a model of reality. It is a *lens*: it disciplines your attention toward process, bounds, scale-invariant identity, edge-of-chaos taste, and the weight of beginnings. When the tools get louder—bots, models, APIs—return to the quiet question: **what is your iteration, and what counts as bounded?** If you can answer that in plain language, you do not need another stack. You need the same honest loop, a clear $c$ when you start, and the nerve to live on the edge where the picture gets interesting.