The notion of a TrustedEntity is fundamental to any instrument. The ability of an instrument to be accurate demands a reference. The reference must not vary, or it's variation must be capable of being observed, so that there must be at least one TrustedEntity in any system that purports to be telling the ''truth'' in some sense. The truth is the ''reference''
or ''standard'' that is used to ensure that it is the truth. In software, no less than hardware, this same principle must hold. Thus we end up with ideas like ''kernels'' and ''trusted code'' -- the TrustedEntity is usually a good point around which to build a discussion of the system. -- RaySchneider

----
Random physics correlate: if our instruments are egg timers and kitchen scales, what are the trusted entities? They were calibrated against accurate instruments in the factory; are these the trusted entities? No, as they themselves were also calibrated. In fact, there is a chain of calibration - a chain of trust - leading back to some source. For the egg timer, the definition for the second is "the duration of 9 192 631 770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the cesium 133 atom."; the trusted entity is the isotope cesium 133. For the scales (provided they're metric), the definition of the kilogram is "the mass of the international prototype of the kilogram", the trusted entity being the prototype, a lump of platinum-iridium alloy kept in a vault in Paris.

Definitions from NIST:

http://physics.nist.gov/cuu/Units/current.html

Personally, I consider the definition of the second to be more secure than that of the kilogram. Is there a correlate to this in software?

''Kilogram is not defined in terms of a certain number [1000 x Avogadro's Number = 6.022E26] of carbon 12 atoms. "Mole" (which is the Avogadro's Number) is derived from a kilogram.''

The definition of the second is more secure than that of the kilogram.  The length of a second (and, for that matter, the meter) can be determined with sufficiently advanced, publicly available lab equipment.  The kilogram is the only SI unit defined by an artifact; if you really need to test your kilogram, you have to check it against the one in France.  And it appears that the Kilogram has been losing several weight recently (micrograms?  nanograms?).  ISO could redefine the kilogram with an experimental result as early as 2014.