Science Quote by Edsger Dijkstra

Dijkstra’s line lands like a koan with teeth: testing feels like proof, yet it’s structurally incapable of delivering the one assurance managers, users, and sometimes developers secretly want - that the code is clean. The wit is in the asymmetry. A single failing test is an existence proof: here is a bug, right now, undeniably. But a thousand passing tests only say you didn’t trip over anything in the rooms you bothered to walk through, in the lighting you happened to turn on.

The specific intent is corrective, almost disciplinary. Dijkstra is pushing back against a cultural habit in engineering: treating testing as a moral alibi. In the 1960s and 70s, software was scaling faster than the methods used to reason about it, and “just test it” became a pragmatic creed. Dijkstra, an evangelist for rigor and formal reasoning, points out the philosophical hole at the center of that creed: programs have near-infinite behaviors, and tests are finite samples. Absence-of-evidence can’t be promoted to evidence-of-absence.

The subtext is also a critique of optimism-as-process. Testing flatters our desire to ship and our faith in checklists. Dijkstra reminds us that reliability isn’t a vibe; it’s an argument. Good engineering, in his worldview, means designing systems whose correctness can be reasoned about, not merely hoped for. Testing remains essential, but demoted: a bug-finder, not a truth machine.