- 24,488
- 15,057
I would consider the switch from Gaussian units, which are much more natural for electromagnetism, to SI units a (slight) minus. The SI is a desease in theoretical electromagnetism. I was shocked to see that even Jackson switched to SI units in the 4th German edition (I guess also in the newest English edition) only to switch back to the Gaussian units when it comes to the real beef, i.e., the relativistic formulation. Of course the use of any kind of units is as valid as any other, but the SI makes E&M unnecessarily a bit more complicated than the use of Gaussian units. The best units, in my opinion, are Heaviside-Lorentz units, which are rationalized Gaussian units. They are used in high-energy physics (relativistic QFT/QED). That's the ones I use always. If necessary, one can easily switch to SI units at the end of the calculation ;-).