At TU/e, we believe that every problem has an optimal solution. If something doesn’t work, you find the error, adjust the model, and you’re done. Politics doesn’t fit that logic. There’s no single outcome, no model with a minimal loss function. It’s about people, interests, and compromise. To many engineers, that feels like a poorly trained system.

Maybe that’s why politics hardly resonates here. Not out of disinterest, but because it’s hard to grasp. In project groups, you choose the model with the lowest loss, not the compromise that keeps everyone satisfied. In politics, it’s exactly the other way around. It’s beta versus alpha. Logic versus interpretation.

The international composition of our campus plays a role as well. Many students aren’t allowed to vote. For them, Dutch politics feels distant—much like European elections often do for Dutch students: important on paper, but hard to truly connect with.

That’s why conversations naturally shift toward global topics: fossil fuels, climate change, unrest in the Middle East. These are issues everyone can relate to, regardless of passport or voting rights. You can sense that in university discussions too, even within the Executive Board. But once the conversation turns back to Dutch elections, engagement becomes harder to find.

Yet politics affects us directly. Decisions on industrial policy, housing, research funding, and education shape how we study, live, and eventually work here. Maybe politics just feels too complex—like a model that never quite fits, no matter how often you try to fine-tune it.

So let’s give it a try—not with an algorithm, but with a ballot. You can vote on Wednesday in the MetaForum, where a polling station has been set up. Think of it as an experiment—an iteration in a model that’s far from fully trained.

Wob Knaap is a Data Science student at TU/e. The views expressed in this column are his own.

This article was translated using AI-assisted tools and reviewed by an editor