With regard to the accuracy, the NIST scientists were reporting an uncertainty of 89 parts per billion. On the NIST website, it describes the significance as follows:
I think you misunderstood me. I wasn't questioning the accuracy of the measurement. They measured it a whatever frequency that the test concerned, whatever frequency of the coils involved. That doesn't prove it is that accurate for everything across the spectrum, from micro waves to visible light to xrays to cosmic rays. We can assume that it is, we already are, but if somebody finds a way to test PC for say xrays and we find PC is inaccurate at that frequency, we have big problems. This will probably never happen, but I can see why it is a possible problem.