I'm having trouble with this definition of the higher order differential that I'm presented:
If $f: U \subset \mathbb{R}^n \rightarrow \mathbb{R}^m$ is k-times continuously differentiable in a neighbourhood of $x_0 \in U$, then the differential of order k
$d^{(k)}f(x_0):\mathbb{R}^n \times \mathbb{R}^n \times ... \times \mathbb{R}^n \rightarrow \mathbb{R}^m$
is explained as symmetric k-linear mapping through:
$d^{(k)}f(x_0)(v_1,...,v_k) = \partial_{v_1},...,\partial_{v_k}f(x_0), (v_1,...,v_k \in \mathbb{R}^n)$
The definition continues by providing examples for different orders:
$d^{(1)}f(x_0)v = \partial_vf(x_0) = lim_{t \to 0} \frac{f(x_0+tv)-f(x_0)}{t} = \sum_{i=1}^n \frac{\partial f}{\partial x_i}(x_0) v^i$ for $v = \left( \begin{array}{c} v^1\\ \vdots\\ v^n\\ \end{array} \right) \in \mathbb{R}^n$.
finally arriving at:
$d^{(k)}f(x_0)(v_1,...,v_k) = \sum_{i_1,...,i_k = 1}^n \frac{\partial^kf(x_0)}{\partial x_{i1}...\partial x_{ik}} v_1^{(i_1)}...v_k^{(i_k)}$ for $v_i = \left( \begin{array}{c} v_i^1\\ \vdots\\ v_i^n\\ \end{array} \right), 1 \leq i \leq k$.
Im confused by the very first definition, as I can't imagine what $d^{(k)}f(x_0)(v_1,...,v_k)$ is even aiming for. I know the following lines look like the directional derivatives for different $v$, but why would we bring them up in the definition of the higher order differential? Any explanation of the definition is highly appreciated. Also, I think an example (or a link to an example) would really help me.