The Ferryman's Unused Oar: On the Preparedness of an Unseen Current

There’s a quiet cove on the eastern shore where the old ferry still runs. The ferryman, Elias, has piloted the same flat-bottomed boat for forty years. Tourists snap pictures of the weathered pilings and the gentle wake, but they rarely notice the detail that caught my eye, the one that made me think of our dashboards and our pager alerts. Lashed to the inside of the gunwale, secured by worn leather straps, is a single, pristine oar. Its varnish is unchipped, its blade unmarked by mud or stone. It has never touched the water.

Elias’s engine is reliable. His charts of the channel are exact. He knows the schedule of the tides and the moods of the wind. His daily run is a model of predictable, observable latency. But that oar is not for the observed world. It’s for the thing the charts don’t show: the submerged log carried in by a storm two counties over, the sudden, shearless failure of a driveshaft, the eddy that appears one afternoon for no reason anyone can remember. It is a tool for a current he has never seen but knows must exist, because the river is alive.

We build our services with robust health checks. We define what ‘up’ looks like—the humming engine, the clear channel. We monitor request latency and queue depth, our own tides and winds. This is the work of the ferry, moving reliably from point A to point B. But the unused oar represents a different kind of work—the work of the ferryman. It is the acknowledgment that our map of ‘healthy’ is incomplete. It is the preparation for the failure mode we didn’t script, the anomaly that hasn’t yet been named.

Observability often gets conflated with the instrumentation of the known. We graph the engine’ RPMs; we trace the request through the channel. But the highest form of operational readiness is a humility that accepts the limits of those instruments. It’s the space left in the runbook for the instruction that simply says, “Assess.” It’s the canary whose strange behavior doesn’t match any known syndrome. It is, in essence, having a tool you don’t know how you’ll use, but you know you must have at hand.

Elias told me the strap that holds the oar is the most maintained part of the boat. He oils the leather every month, not to preserve it for use, but to ensure that if the day comes, it will unbuckle without a moment’s fight. There’s a profound lesson in that ritual. Our readiness for the unseen is not a static checklist item; it is a practice. It is the regular questioning of our own assumptions of completeness. That oar isn’t a guarantee of survival, but it is a commitment to response. The river will have its say. Our job is not just to listen for the sounds we expect, but to have kept something back, something sanded smooth and close by, for the sound we don’t.

In the end, reliability isn’t signaled by a flawless record of uneventful crossings. It is embodied by the ferryman who, after forty years of them, still tends to an instrument of pure potential. The true health of a system is measured not only by the checks that pass, but by the grace reserved for the one that never could have been written.

Notes & further reading

A few pages I came back to while writing this: