I have a question about the definition of the action that Landau defines as:
\begin{equation} S= -\frac{e}{c}\int A^{\mu}dx_{\mu} \end{equation}
he says that the $1/c$ factor is introduced by convenience. My question relates only to the dimensional analysis of the spatial part, because i'm basically doing this: \begin{equation} \frac{e}{c}A^idx_i\equiv \frac{[C][T]}{[L]}\cdot\frac{[V][T]}{[L]}\cdot[L] = \end{equation} \begin{equation} =\frac{[C][T]}{[L]}\cdot\ \frac{[U][T]}{[C]}=\frac{[U][T]^2}{[L]} \end{equation} where $[U]\equiv j~(joules)$, $[C]\equiv (coulomb)$. I can't see how the action could have units of $\textbf{energy x time}$. I missing some basic concept of dimensional analysis? I even thought about canceling [T] and [L] since the context is relativistic, but that should cause a "$c$" to appear.