Tom,
" You'd have to show that the laws of physics are not uniform over the universe, before you could even make your assumption."
And your assumption is that we know all the laws of physics, including what will affect light traveling trillions of miles, so in order to support that assumption we currently have to accept inflation, dark energy and the various other fixes. Yes, I know all about the controversies between 'tired light/steady state,' vs. big bang models, but at some point the larger debate will have to be reopened, as we keep finding evidence of structures that took longer than the assumed age of the universe to develop. We are pretty much up to the line now.
Did you read that paper by Christov I keep linking to?
1. Introduction
The propagation of waves in linear dissipative systems is well studied but most of the investigations are concerned with the propagation of a single-frequency wave. On the other hand, in any of the practical situations, one is faced actually with a wavepacket, albeit with a very narrow spread around the central frequency. This means that one should take a special care to separate the effects of dispersion and dissipation on the propagation of the wavepacket from the similar effects on a single frequency signal.
"5. Conclusions
In the present work, the effect of attenuation and dissipation on propagation of waves governed by the Jeffrey equation is addressed. When packets of small but finite breadth are considered the presence of dissipation changes the central wave number of the packet. The distribution of the wave length around the central length is assumed to be Gaussian which is the most frequently encountered case in cosmology when hot stars are observed. Dispersion relation for the damped wave equation is derived and the evolution of the packet density is investigated in time(or space). It is shown that the attenuation acts merely to decrease the amplitude of the shifts packed, while the dissipation damps the higher frequencies stronger than the lower frequencies and shifts the maximal frequency of the packet to lower frequencies (longer wavelengths), i.e., the packet appears redshifted upon its arrival. For Gaussian wavepackets, this kind of redshift is linearly proportional to the time passed or the distance traveled. The coefficient of proportionality contains the ratio of the dissipation coefficient and the initial width of the distribution which means that the thicker packets are redshifted more than the narrower ones for the same distance or for the same time. We call this liner relationship ''Hubble Law'' for redshifting of wavepackets."
Missing one little detail can throw everything out of whack.
Regards,
John M