Week 1 through Week 8: Normal. The route is 380 kilometers of cleared road through what used to be Ohio. I've been driving it for two years. The road is straight where it can be, curved where the terrain demands it, and maintained by automated systems that keep the surface smooth and the edges defined. The drive takes approximately five hours at convoy speed. The land on either side is trees and grass and whatever else grows in Ohio now. I don't look at it more than I have to. The briefing says not to look at it. I look at it anyway, because eight hours in a cab with nothing but forward is its own kind of problem. It looks like land. It looks fine. Weeks 1 through 8 were fine.

Week 9: The drive took five hours and forty minutes. I logged it. Dispatch said GPS calibration. I said nothing because GPS calibration is what dispatch always says. But the road felt longer. Not in a vague, subjective sense. I know this road. I know where the curves are, where the straightaways are, where the old exit signs are that haven't been removed because nobody maintains anything that isn't the road surface. The exit sign for what used to be Mansfield was in the wrong place. It was approximately three kilometers farther east than it should have been. I noted this. Nobody else in the convoy mentioned it. I checked with my co-driver. She said the sign was where it always was. I did not argue.

Week 14: The landmarks are wrong and I have stopped reporting it because the responses I get are variations of "GPS calibration" and "fatigue-related perceptual shift," and I am neither miscalibrated nor fatigued. The bridge over the Muskingum River — a fixed, physical structure of steel and concrete — was 200 meters longer this week than last week. I measured it by time at constant speed. My co-driver measured it independently and got the same result. We reported it jointly. Dispatch thanked us for our diligence and noted that bridge length measurements from a moving vehicle are inherently imprecise. We are not imprecise. The bridge is longer. Next week I will measure it again.

Week 19: A bridge appeared that was not there last week. It spans a river that was not there last week. The river is approximately 30 meters wide, flowing north to south, crossing the corridor at a point between the 240 and 250 kilometer markers. There is no river at this location on any map. The bridge is steel, double-lane, structurally sound. I drove across it because the road led to it and the convoy was behind me and stopping is not permitted. The bridge held. The river below was clear and moving. On the far side, the road continued as normal. I reported the bridge. Dispatch said they would send a survey team. The survey team found no bridge and no river at the coordinates I provided. I drove the same route the following week. No bridge. No river. The road was continuous. The asphalt where the bridge approaches had been showed no seams, no patches, no evidence that anything had ever interrupted it.

Week 23 through Week 26: I am keeping this log because the official reports do not reflect what is happening and I want a record. The road is changing. Not all at once and not dramatically, but consistently. Each week, something is different. A hill that wasn't there. A curve that's new. Trees on the left side that are a different species than the week before. The transit time varies between four hours forty minutes and seven hours twelve minutes with no explanation. My co-driver sees some of what I see and not all of it. The things she sees that I don't, she doesn't tell me about anymore, and I extend her the same courtesy. We drive the corridor. The corridor is reliable. The corridor takes us where we need to go. What the corridor is — whether it is a road through stable territory or a path that creates its own stability as we drive it — is a question I have stopped asking because the answer, whatever it is, will not change the fact that the convoy runs weekly and I am the one driving it.