diff --git a/chapters/algebra-moonmath.tex b/chapters/algebra-moonmath.tex
index baa87be..7992573 100644
--- a/chapters/algebra-moonmath.tex
+++ b/chapters/algebra-moonmath.tex
@@ -459,14 +459,14 @@ \subsubsection{Hash functions}\label{sec:hash-functions} Generally speaking, a h
 
 \begin{example}[$k$-truncation hash]\label{ex:k-truncation-hash} One of the most basic hash functions $H_k:\{0,1\}^*\to \{0,1\}^k$ is given by simply truncating every binary string $s$ of size $|s|> k$ to a string of size $k$ and by filling any string $s'$ of size $|s'|<k$ with zeros. To make this hash function deterministic, we define that both truncation and filling should happen on the highest bits, or ``on the left''.
 
-For example, if the parameter $k$ is given by $k=3$, $s_1=<0,0,0,0,1,0,1,0,1,1,1,0>$ and $s_2=1$, then $H_3(s_1)=<1,1,0>$ and $H_3(s_2)=<0,0,1>$.
+For example, if the parameter $k$ is given by $k=3$, $s_1=<0,0,0,0,1,0,1,0,1,1,1,0>$ and $s_2=<1>$, then $H_3(s_1)=<1,1,0>$ and $H_3(s_2)=<0,0,1>$.
 \end{example}
 
 A desirable property of a hash function is \term{uniformity}, which means that it should map input values as evenly as possible over its output range. In mathematical terms, every string of length $k$  from $\{0,1\}^k$ should be generated with roughly the same probability.
 
 Of particular interest are so-called \term{cryptographic} hash functions, which are hash functions that are also \term{one-way functions}, which essentially means that, given a string $y$ from $\{0,1\}^k$ it is infeasible to find a string $x\in\{0,1\}^*$ such that $H(x)=y$ holds. This property is usually called \term{preimage-resistance}.
 
-Moreover, if a string $x_1\in\{0,1\}^*$ is given, then it should be infeasible to find another string $x_2\in\{0,1\}^*$ with $x_1\neq x_2$ and $H(x_1)=H_(x_2)$
+Moreover, if a string $x_1\in\{0,1\}^*$ is given, then it should be infeasible to find another string $x_2\in\{0,1\}^*$ with $x_1\neq x_2$ and $H(x_1)=H(x_2)$
 
 In addition, it should be infeasible to find two strings $x_1,x_2 \in\{0,1\}^*$ such that $H(x_1)=H(x_2)$, which is called \term{collision resistance}. It is important to note, though, that collisions always exist, since a function $H: \{0,1\}^* \to \{0,1\}^k$ inevitably maps infinitely many values onto the same hash. In fact, for any hash function with digests of length $k$, finding a preimage to a given digest can always be done using a brute force search in $2^k$ evaluation steps. It should just be practically impossible to compute those values, and statistically very unlikely to generate two of them by chance.
 
@@ -615,9 +615,9 @@ \subsubsection{Pseudorandom Function Families in DDH-secure groups}
 As noted in \ref{def:Pedersen_hash}, the family of Pederson's hash functions, parameterized by a set of generators $\{g_1,\ldots,g_j\}$ does not qualify as a family of pseudorandom functions, and should therefore not be instantiated as such. To see an example of a proper family of pseudorandom functions in groups where the decisional Diffie--Hellman assumption (see \secname \ref{def:DDH-secure}) is assumed to hold true, let $\G$ be a DDH-secure cyclic group of order $n$ with generator $g$, and let $\{a_0,a_1,\ldots,a_k\}\subset \Z_{n}^*$ be a uniform randomly generated set of numbers invertible in modular $n$ arithmetics. Then a family of pseudorandom functions, parameterized by the \uterm{seed} $\{a_0,a_1,\ldots,a_k\}$ is given as follows:
 \begin{equation}
 \label{prf_in_cyclic_group}
-F_{\{a_0,a_1,\ldots,a_k\}}: \{0,1\}^{k+1} \to \G:\; (b_0,\ldots,b_k)\mapsto g^{a_0\cdot \Pi_{i=1}^k a_i^{b_i}}
+F_{\{a_0,a_1,\ldots,a_k\}}: \{0,1\}^{k+1} \to \G:\; (b_0,\ldots,b_k)\mapsto g^{\Pi_{i=0}^k a_i^{b_i}}
 \end{equation}
-\begin{exercise} Consider the multiplicative group $\Z_{13}^*$ of modular $13$ arithmetic from \examplename{} \ref{ex:Zn*} and the parameter $k=3$. Choose a generator of $\Z_{13}^*$, a seed and \uterm{instantiate} a member of the family given in \eqref{prf_in_cyclic_group} for that seed. Evaluate that member on the binary string $<1,0,1>$.
+\begin{exercise} Consider the multiplicative group $\Z_{13}^*$ of modular $13$ arithmetic from \examplename{} \ref{ex:Zn*} and the parameter $k=2$. Choose a generator of $\Z_{13}^*$, a seed and \uterm{instantiate} a member of the family given in \eqref{prf_in_cyclic_group} for that seed. Evaluate that member on the binary string $<1,0,1>$.
 \end{exercise}
 %\begin{example}[p\&{}p-$\F_{13}$-drop-hash]We can consider the same pen\&paper hash function from XXX and define another hash into $\F_{13}$, by deleting the first leading bit from the hash. The result is then a $3$-digit number and therefore guaranteed to be smaller then $13$, since $13$ is equal to $(1101)$ in base $2$.