Now, after the redefinition of SI, the elementary charge $e$ and the reduced Planck constant $\hbar$ (and also $k_B$) are exact quantities in the SI units, as is the velocity of light $c$. The magnetic and electric constants became non-exact quantities (having the uncertainty of order $10^{-10}$), as the fine-structure constant $\alpha$ always was.
The Gaussian units (and also units in other variants of CGS) are connected with the SI units via exact factors involving $\pi$ and $c$. Neither BIPM nor any other organization regulates the Gaussian system as BIPM regulates SI. So it seems that fixing $e$, $\hbar$, and $c$ in SI automatically fixes these constants in Gaussian unit system.
Generally, $\alpha=k_Ee^2/\hbar c$, and the Coulomb potential is $\phi=k_Eq/r$ (Coulomb law). In the Gaussian unit system, the units were chosen in such a way that $k_E=1$, nicely.
So apparently this means that the Coulomb constant $k_E$ in the Gaussian unit system now becomes an experimentally measured quantity approximately equal to $1$ with the same relative uncertainty. Is that true?
Another option is to make the conversion factor between the charge units in SI and the Gaussian system a non-exact quantity but keep $k_E=1$. But who will determine what quantity becomes non-exact ($k_E$ or the conversion factor)? Is it possible that different authors (in particular, textbook authors) will use different conventions?
