My Operating System
A look under the hood, AKA: Why every article I write is actually about the same thing.
Every once in a while, I finish an article, lean back, and realize I’ve jumped tracks entirely. One week it’s child protection. The next it’s institutional trust, corporate governance, or engineering infrastructure. On the surface, it looks disjointed, like someone aggressively collecting unrelated hobbies.
But someone recently pointed out a pattern that I hadn’t fully articulated myself: none of these pieces are actually about their stated subjects. They are all running on the same underlying operating system.
I never consciously set out to construct a unified worldview or earn a “systems thinker” label. I was just chasing whatever question wouldn’t leave me alone. Eventually, though, those questions started converging on the same patch of ground.
When I was working as a Linux systems engineer, the day-to-day mechanics of fixing a broken server didn’t interest me nearly as much as the architecture behind the failure. Why did this specific configuration collapse while another survived? Why did a single outage cascade into three others? Why did two engineers look at the exact same telemetry and reach completely different conclusions?
Years later, when I began researching child exploitation, I expected the terrain to be entirely different. It wasn’t. I found myself asking the exact same structural questions. The core problem wasn’t just “who are the bad actors?” It was: Why do so many environments consistently manufacture opportunities for them? Why do these patterns replicate across completely different platforms? What hidden variables allow one coure of action to drastically reduce risk while another unknowingly compounds it?
It didn’t feel like a career pivot. It felt like pointing the same flashlight into a different dark room.
The pattern kept repeating. When I looked at election integrity, politics was the least interesting layer. The real question was trust architecture: how do you design a system where participants don’t require blind faith in one another and a black box process? What happens when the foundational assumptions of that system quietly expire? When I started sketching out ideas for a crypto-based voting system, the technology changed, but the ledger of questions remained identical: Who holds agency? Where do the incentives actually point? What happens when a participant decides to cheat, and can the architecture be set up so that cooperation is simply more efficient than malice?
Looking back at everything I’ve written, the common thread is a mechanical one. I almost never start with an opinion; I start with a model. Before deciding if an outcome is good or bad, effective or ineffective, I need to see how the machine works. What are the moving parts? What assumptions are buried in the foundation? What incentives are quietly driving behavior while everyone else argues about intentions?
That is probably why my conclusions rarely fit into standard ideological buckets. Most public debates start by asking, “Who is right?” I usually start by asking, “What is this system currently optimized to produce?”
Those two questions don’t live in the same universe, and they never yield the same answers.