Decades of static data, collected without the controls that turned out to matter

A triboelectric measurement is a statement about two surfaces and everything touching them at the time: the humidity, the contact history of the samples, the separation speed, whatever invisible film had settled on them. If the number of factors in play is as large as the note says Conover found at the meeting ^[4], then a charging figure recorded in one lab describes that lab's afternoon rather than a property of the material pair. That is the part that matters to anyone who has copied such a figure into a work instruction.

Practice in the three domains the note names runs on transfer ^[1]. You assume a measured charging tendency for a glove against a lead frame holds in your building, at your line speed, in your climate. If the original data were gathered without controlling variables that later turned out to dominate, transfer is where the error enters, and it enters quietly. The discharge that kills the part or lights the vapour does not report which of the study's uncontrolled conditions failed to reproduce.

Notice the shape of the claim, though. It arrives in an editor's note introducing the issue ^[8], and it says earlier research looks "less than reliable" ^[5] without naming one variable, one study, or how far off any single measurement was ^[7]. Two very different worlds are consistent with that sentence. In one, controlling humidity moves a result by a few per cent and existing safety margins already swallow it. In the other, it moves the sign of the charge. An operator reading the note cannot tell which one they are in, and the note does not claim to say.

The same piece concedes something older and larger: after centuries of work, the mechanism by which charge leaps between objects is still not identified ^[2]. Engineers have shipped against that hole for the whole of the electronics era, substituting empirical charging tables and generous grounding for theory. A reliability problem in the tables does not widen the theoretical hole. It does raise the question of whether the empirical layer covering it was ever load-bearing.

The durable thing here is a reporting mechanism rather than a finding. Science News says it has spent decades paying to put writers in rooms where work has not yet gone public ^[6], and this is the sort of thing that turns up there. Members of a field telling a visiting reporter that their predecessors measured with the wrong controls does not appear in the abstract of the paper still citing those measurements.