Here's an attempt to put MiHsC in context. A long time ago Galileo performed carefully timed experiments rolling balls down inclined planes and found that as height was lost, speed was gained in a particular way. This was later modeled using two interchangable kinds of energy: kinetic (speed) and potential (height) and their sum was found to be nearly conserved: PE + KE = constant (some energy leaks to smaller scales as friction, and that is where thermodynamics appears). Then Einstein explosively predicted that one can convert a tiny bit of mass to lots of energy, and reversewise (E=mc^2) so that now mass-energy was conserved: mass + energy = constant. What I think MiHsC is telling us, is that what is actually conserved is:

Energy + Mass + Information = constant

This sum is a 'property' consisting of energy, mass and information that you could call 'EMI', so that EMI is conserved. What might this mean for a real experiment?

Consider a ring in a cryostat, that is suddenly spun. Tajmar et al. found that a nearby gyroscope moved slightly to follow the ring, despite there being no frictional connection. MiHsC predicts this exactly, since the sudden acceleration increases the inertial mass of the gyroscope and to conserve the momentum of the gyro+ring system the gyro has to move with the ring. Can we interpret MiHsC as EMI conservation? Perhaps. When you accelerate the ring, the Rindler horizon seen by the gyroscope, which sees a mutual acceleration, comes closer to it, so it looses information about its environment (I'm not sure how to calculate this yet). To conserve EMI it must gain mass-energy, or inertial mass (this agrees qualitatively with MiHsC).

Conversely as an object's acceleration reduces as it moves away from concentrations of mass into deep space, the Rindler horizon it sees moves away and it gains information (I), so EM must be lost. Therefore inertial mass decreases and the object is more sensitive to external forces and accelerates again. The minimum acceleration of 6.7*10^-10 m/s^2 occurs when the Rindler horizon coincides with the Hubble horizon since then no more information can be gained. The thing now is to see if the maths of this idea predicts the right sort of behaviour..

Consider a ring in a cryostat, that is suddenly spun. Tajmar et al. found that a nearby gyroscope moved slightly to follow the ring, despite there being no frictional connection. MiHsC predicts this exactly, since the sudden acceleration increases the inertial mass of the gyroscope and to conserve the momentum of the gyro+ring system the gyro has to move with the ring. Can we interpret MiHsC as EMI conservation? Perhaps. When you accelerate the ring, the Rindler horizon seen by the gyroscope, which sees a mutual acceleration, comes closer to it, so it looses information about its environment (I'm not sure how to calculate this yet). To conserve EMI it must gain mass-energy, or inertial mass (this agrees qualitatively with MiHsC).

Conversely as an object's acceleration reduces as it moves away from concentrations of mass into deep space, the Rindler horizon it sees moves away and it gains information (I), so EM must be lost. Therefore inertial mass decreases and the object is more sensitive to external forces and accelerates again. The minimum acceleration of 6.7*10^-10 m/s^2 occurs when the Rindler horizon coincides with the Hubble horizon since then no more information can be gained. The thing now is to see if the maths of this idea predicts the right sort of behaviour..