Can You Hear Me Now

Pip: Can You Hear Me Now, where today’s episode asks the question every developer over forty has been quietly screaming into their IDE: what exactly are we trading away?

Mara: Andrew Prokop makes that case directly — covering AI in education, the calculator comparison that keeps getting made, and what it means to lose the actual process of learning. Let’s start with the argument at the center of it all.

Get Off My Lawn — AI, Debugging, and What We’re Losing

Mara: The post opens with a provocation: AI is producing developers who cannot debug their own code. That’s the frame — not whether AI is useful, but whether dependence on it hollows out the skills that make the work meaningful.

Pip: And the post names that loss directly. Here’s the line that anchors the whole argument, drawn from a Substack piece by Noah Brier that a friend shared: “I don’t believe the job of a high school English teacher is to develop great writers, but rather to convince teenagers that devoting time and energy to writing is a worthwhile pursuit.”

Mara: So the upshot is that the goal of education isn’t the output — it’s the orientation. Teaching someone to love the process is the thing AI quietly skips over.

Pip: The debugging nostalgia here is real, not decorative. There’s a specific memory of tracing a race condition deep in the Motorola Unix kernel — unfixed for years, found by someone who actually had to look. That’s the texture of what gets lost.

Mara: Right, and the post draws a careful distinction around the calculator comparison. The argument is that a calculator removes tedium but preserves understanding — you still know what multiplication is. AI, by contrast, does the research, the discernment, the creation. The student’s role shrinks to writing a prompt.

Pip: Which is why “show your work” existed long after calculators did. The work was still yours.

Mara: There’s also a side note about a musician friend — someone who once built original recordings in a homemade studio and now posts only AI-generated songs. It’s offered as a parallel: the same substitution, a different domain.

Pip: The post is careful to say this isn’t Luddism. There’s genuine hands-on AI work described — personal projects, familiarity with Ollama, years inside the tooling. The wariness comes from proximity, not ignorance.

Mara: The closing position is precise: “I firmly believe that a tool is only useful if it allows us to preserve our humanity, values, dignity, and compassion. Anything that takes away from the above is suspect and should be questioned.”

Pip: Debugging your own code as an act of resistance. There are worse hills.

Mara: The question underneath all of this is whether the process of struggling toward something is itself the point — and what we become when we stop.

Pip: Next time, we find out what else we’ve quietly handed over. Stay with us.