For anyone who wants to see the quasi-practical result of calibrating with room lighting turned on.
My computer is in a small room with red walls and a 7ft ceiling. There’s only one primary light, which is a 60w incandescent. Without moving the meter, I ran the test twice; once with the overhead light on, once with it off.
See the files for the impact made by the overhead light. I suspect that most of the variance is due to instrument unreliability, rather than the overhead lighting. For example, the blue channel shifts high at some points, and low at others.
Note: At this point, the monitor was running without a calibration.