Indeed, the last century was full of developments concerning the measuring of viscosity. In the Newtonian sense, oils are not ideal liquids, i.e., their viscosity is dependent on the shearing-off stress. Therefore, turbulences may not play a part in the viscosity testing of oils. This however, was precisely the case with the first discharge-measuring-vessels. Almost the same as the first phrasing from 1911, thereby, oil should, at a certain minus-temperature, 'flow freely from the oil-can'.
A hallmark of scientific testing, is the ability to reproduce the resuts. Therefore, there is a mathematical framework, with clearly defined specifics and units. When testing using discharge-measuring-vessels, the oils can, at best, be compared with each other. Were this the case, all the laboratories would have to agree on having a so-called reference engine oil, which would mean more impediment than progress.
Testing routines are a very important criterion for a quality-controll-laboratory. The above shown Ubbelohde-viscometer is, because of the air-pressure compensation, U-shaped and features a long capillary (green), which avoids the influence of possible turbulence-flows. One pours the oil in at the top and then measures the time it takes for the oil to travel from one red line to the other.
At this point, we would like to point out the rotational viscometer, with which the dynamic viscosity can be measured.