I am currently in the process of studying special Relativity and I keep stumbling over a concept I can't make consistent for myself. It is about the fact which index of a Lorentz transform and the Minkowski metric denotes a column and which one denotes the row.
My thoughts so far: If I take a look at the Matrix multiplication $\mathbf{x}^{'\nu}=\mathbf{\Lambda}^\nu\,_\mu\cdot\mathbf{x}^\mu$, then the upper index $\mu$ must be the Index indicating a row as $\mathbf{x}$ is a column vector and therefor it's Index must specify the entry (row). Same holds true for the upper index $\nu$ of $\mathbf{x}^{' \nu}$. Additionally I know that the product describes a matrix multiplication with a vector and due to the Einstein sum rule the expression is summed over $\mu$. So if I am interested in the first entry of $\mathbf{x}^{'}=\mathbf{x}^{'0}$ I have to multiply the first row of $\mathbf{\Lambda}=\mathbf{\Lambda}^0$ with the column of $\mathbf{x}$. Therefor the upper index $\nu$ of the Lorentz transformation describes the row of the matrix and the lower index $\mu$ the column. So far so clear.\
Now I encounter some difficulties. For example our professor writes the following: $x_\nu=\eta_{\mu\nu}x^\mu$. (first question: strictly speaking and also written here https://en.wikipedia.org/wiki/Raising_and_lowering_indices (example in Minkowski spacetime): is $x_\nu$ a row vector?). The same reasoning as above cannot be used here as we have no upper/lower index. But after the same logic in $\mathbf{x}^\mu$, $\mu$ describes a row index and therefor in $\eta_{\mu\nu}$ $\nu$ must describe a row again and $\mu$ a column. So now the latter index is the row index (after my understanding). Now the inconsistencies start: In the book here equation (5.12)+(5.13) the authors say that $\Lambda^\mu\,_\alpha \eta_{\mu\nu}\Lambda^\nu\,_\beta$ is not a matrix multiplication as the index $\mu$ is a column index in the first Lorentz transformation as well as in the Minkowski metric.
I also know (very little to be honest) about Tensors and that they play an important role in the lowering and raising of indexes, but there has to be a self consistent answer to my Index problem somewhere. I have yet to find a satisfying answer to my problem and would be grateful for any help you can provide.
Edit: I have found a post where in one answer a reference to another question is given where the author says the left most index indicates the row. That would at least support my claim for the index label of the Lorentz transformation, yet the problem with the Minkowski metric remains.