... waitdavid137, if you can indeed show me why I should abandon this, I will tip my hat to you. ...
I'm not saying that you should abandon this at all. What I am saying is why I doubt the feasibility of the model and that its on you if you want to convince me of it.
If you can show me the usefulness of isotropic coordinates and how to make clear transformations between coord. systems, I will appreciate it.
The transformation between the standard RN coordinates and isotropic coordinates is equation 7.1.12a-c in my chapter on general relativistic electromagnetism
I doubt isotropic coordinates exist for the Kerr solution as I think in order to truly be considered isotropic they would have to at the same time diagonalize the metric.
They are useful in that they leave the coordinate speed of light c' independent of direction as
c' = (1 - G²M²/4r²c⁴ + Gkq²/4r²c⁴)c/[(1+GM/2rc²)² - Gkq²/4r²c⁴]²
If the radial and tangential polarizations should be affected differently then why does these coordinates exist allowing the effect on light in both directions to be the same?
If a PV model is correct why don't we see a wavelength dependence on the index of refraction of space in this equation here just like you see in materials?
Consider also the Schwarzschild solution in Kruskal-Szekeres coordinates. The coordinate speed of radial moving light is everywhere c. How does your model cope with that?
If your model prefers Schwarzschild coordinates, you need to demonstrate why there really is a preferred frame. If you can demonstrate that there really is a preferred frame, then you truly have demonstrated something profound.