diff --git a/chapters/ethernet.tex b/chapters/ethernet.tex
index 38637dd..75fa4ce 100644
--- a/chapters/ethernet.tex
+++ b/chapters/ethernet.tex
@@ -37,7 +37,7 @@
 \begin{figure}
 \centering
 \includegraphics[width=.8\textwidth]{images/mac-address-format.png}
-\caption{The \acs{MAC} address consists of two parts with the first part having two bits with special meaning}
+\caption[The format of a \acs{MAC} address]{The \acs{MAC} address consists of two parts with the first part having two bits with special meaning}
 \label{fig:mac-format}
 \end{figure}
 }
diff --git a/chapters/introduction.tex b/chapters/introduction.tex
index b9e88fc..ad95e32 100644
--- a/chapters/introduction.tex
+++ b/chapters/introduction.tex
@@ -172,7 +172,9 @@ \subsection{Wide-area networks}
 \begin{figure}
 \centering
 \includegraphics[width=.5\textwidth]{images/thompson.png}
-\caption{Ken Thompson (sitting) and Dennis Ritchie working on a PDP-11, circa~1970. The picture was taken by Peter Hamer and the line drawing was made by \href{https://www.truecable.com/blogs/cable-academy/a-brief-history-of-network-technology}{truecable.com}.}
+\caption%
+   [Thompson and Ritchie working on a \abbr{PDP-11}]%
+   {Ken Thompson (sitting) and Dennis Ritchie working on a \abbr{PDP-11}, circa~1970. The picture was taken by Peter Hamer and the line drawing was made by \href{https://www.truecable.com/blogs/cable-academy/a-brief-history-of-network-technology}{truecable.com}.}
 \label{fig:thompson}
 \end{figure}
 
@@ -273,26 +275,54 @@ \subsection{Wide-area networks}
 
 \Paragraph{unicast}
 \mode<article>{
-Unicast is a one-to-one transmission from one point in the network to another point; that is, one sender and one receiver, each identified by a network address.
+Unicast is a one-to-one transmission from one point in the network to another point (\cref{fig:unicast}); that is, one sender and one receiver, each identified by a network address.
+
+\begin{figure}
+\begin{minipage}{.4\textwidth}
+\includegraphics[width=\textwidth]{images/unicast.png}
+\caption[Unicast routing scheme]{Unicast routing}
+\label{fig:unicast}
+\end{minipage}
+\hfill
+\begin{minipage}{.4\textwidth}
+\includegraphics[width=\textwidth]{images/broadcast.png}
+\caption[Broadcast routing scheme]{Broadcast routing}
+\label{fig:broadcast}
+\end{minipage}
+\end{figure}
 }
 
 \Paragraph{broadcast}
 \mode<article>{
-Broadcasting is a method of transferring a message to all recipients simultaneously.
+Broadcasting is a method of transferring a message to all recipients simultaneously (\cref{fig:broadcast}).
 Broadcasting can be performed as a high-level operation in a program, or it may be a low-level networking operation, for example broadcasting on Ethernet.
 }
 
 \Paragraph{multicast}
 \mode<article>{
-Multicast is group communication where data transmission is addressed to a group of destination computers simultaneously.
+Multicast is group communication where data transmission is addressed to a group of destination computers simultaneously (\cref{fig:multicast}).
 Multicast can be one-to-many or many-to-many distribution.
 }
 
 \Paragraph{anycast}
 \mode<article>{
-Anycast is a network addressing and routing methodology in which a single destination IP address is shared by devices (generally servers) in multiple locations.
+Anycast is a network addressing and routing methodology in which a single destination IP address is shared by devices (generally servers) in multiple locations (\cref{fig:anycast}).
 Routers direct packets addressed to this destination to the location nearest the sender, using their normal decision-making algorithms, typically the lowest number of \gls{BGP} network hops.
 Anycast routing is widely used by \acf{CDN} such as web and \acs{DNS} hosts, to bring their content closer to end users.
+
+\begin{figure}
+   \begin{minipage}{.4\textwidth}
+   \includegraphics[width=\textwidth]{images/multicast.png}
+   \caption[Multicast routing scheme]{Multicast routing}
+   \label{fig:multicast}
+   \end{minipage}
+   \hfill
+   \begin{minipage}{.4\textwidth}
+   \includegraphics[width=\textwidth]{images/anycast.png}
+   \caption[Anycast routing scheme]{Anycast routing}
+   \label{fig:anycast}
+   \end{minipage}
+   \end{figure}
 }
 
 \mode<article>{
@@ -325,7 +355,7 @@ \subsection{Wide-area networks}
 \begin{figure}
 \centering
 \input{images/network-models}
-\caption{The three network models}
+\caption[The \acs{OSI}, \acs{TCP}/\acs{IP}, and hybrid networking models]{The three network models}
 \label{fig:network-models}
 \end{figure}
 }
@@ -372,7 +402,7 @@ \subsection{Wide-area networks}
 \begin{figure}
 \centering
 \input{images/data-encapsulation}
-\caption{%
+\caption[Encapsulation and decapsulation of packets]{%
 Data encapsulation at the source and decapsulation at the destination.
 Each layer adds its own header, but the data link layer also adds a trailer.
 }
@@ -443,7 +473,7 @@ \subsection{Wide-area networks}
 The network and transport layers are host-to-host.
 
 
-\begin{figure}[ht]
+\begin{figure}
 \centering
 \input{images/data-flow}
 \caption{Encapsulation and data flow in the network}
@@ -463,7 +493,7 @@ \subsection{Wide-area networks}
 \Vref{fig:icons} shows the three basic network icons used most often in network drawings.
 Many different variants exist for these icons, depending on taste or function (e.g.~a core switch versus an access switch).
 There are more icons available for devices such as wireless \acs{LAN} controllers, acccess points, load balancers, and many more.
-\begin{figure}[hb]
+\begin{figure}
    \centering
    \begin{subfigure}[b]{18mm}
    \includegraphics[width=\textwidth]{images/router.jpg}
diff --git a/chapters/ip.tex b/chapters/ip.tex
index 7c9d518..f6c83a7 100644
--- a/chapters/ip.tex
+++ b/chapters/ip.tex
@@ -310,7 +310,7 @@
 \begin{figure}
    \centering
    \usebox\subnettingOne
-   \caption{With subnetting you can divide a block of 256~\acs{IP} addresses into four smaller blocks of 64~addresses each}
+   \caption[Dividing a block of \acs{IP} addresses in four equal parts]{With subnetting you can divide a block of 256~\acs{IP} addresses into four smaller blocks of 64~addresses each}
    \label{fig:subnetting}
 \end{figure}
 }
@@ -348,7 +348,7 @@
 \begin{figure}
    \centering
    \usebox\subnettingTwo
-   \caption{When using variable-length subnet masks you can further divide a subnet into smaller segments}
+   \caption[With \acs{VLSM} you can divide an \acs{IP} block in unequal parts]{When using \acfp{VLSM} you can further divide a subnet into smaller segments}
    \label{fig:vlsm}
 \end{figure}
 }
@@ -650,7 +650,7 @@
 \begin{figure}
    \centering
    \input{images/metric}
-   \caption{A router can learn a prefix from several other routers. The metric decides the best route.}
+   \caption[The metric decides the best route in a network]{A router can learn a prefix from several other routers. The metric decides the best route.}
    \label{fig:metric}
 \end{figure}
 }
diff --git a/chapters/physical.tex b/chapters/physical.tex
index 5a69190..f8b975a 100644
--- a/chapters/physical.tex
+++ b/chapters/physical.tex
@@ -82,9 +82,21 @@
 \mode<article>{
 An Ethernet crossover cable is a crossover cable for Ethernet used to connect computing devices together directly.
 It is most often used to connect two devices of the same type, e.g. two computers (via their network interface controllers) or two switches to each other.
-By contrast, \emph{straight through} patch cables are used to connect devices of different types, such as a computer to a network switch.
+By contrast, \emph{straight-through} patch cables are used to connect devices of different types, such as a computer to a network switch.
 Intentionally crossed wiring in the crossover cable connects the transmit signals at one end to the receive signals at the other end.
-Many network devices today support auto \acs{MDI-X} (aka `auto crossover') capability, wherein a patch cable can be used in place of a crossover cable, or vice versa, and the receive and transmit signals are reconfigured automatically within the device to yield a working connection.
+Many network devices today support auto-\acs{MDI-X} (aka auto-crossover) capability, wherein a patch cable can be used in place of a crossover cable, or vice versa, and the receive and transmit signals are reconfigured automatically within the device to yield a working connection.
+
+To make a straight-through patch cable, either use the order defined by \abbr{T568A} on both ends of the cable, or the order defined by \abbr{T568B}.
+To create a crossover patch cable, use \abbr{T568A} on one side and \abbr{T568B} on the other.
+
+\begin{figure}
+\centering
+\includegraphics[width=.7\textwidth]{images/t568.jpeg}
+\caption{The \abbr{T568} standard for twisted-pair cabling}
+\label{fig:t568}
+\end{figure}
+
+% TODO: explain (with a drawing?) that pair 1/3 is used for sending and the other for receiving data.
 }
 
 \Paragraph{Gigabit Ethernet}
@@ -113,17 +125,25 @@
 Higher speeds or longer lengths require fibre optic cables to be used.
 }
 
-\Paragraph{direct-attach copper}
+\Paragraph{\acl{DAC}}
 \mode<article>{
-This is a copper 10~Gigabit Ethernet cable which comes in either an active or passive twinax (twinaxial) cable assembly and connects directly into an \acs{SFP+} housing.
+A \acf{DAC} cable is a copper 10~Gigabit Ethernet cable which comes in either an active or passive twinax (twinaxial) cable assembly and connects directly into an \acs{SFP+} housing (\vref{fig:dac}).
 \abbr{40GBASE-CR4} and \abbr{100GBASE-CR10} physical layers using \SI{7}{\metre} twin-axial cable are being developed as part of the 100~Gbit Ethernet specifications by the \acs{IEEE}.
+
+\begin{figure}
+\centering
+\includegraphics[width=.7\textwidth]{images/dac-cable.jpeg}
+\caption{A direct-attach copper cable}
+\label{fig:dac}
+\end{figure}
 }
 
 \Paragraph{optical fibre}
 \mode<article>{
 An optical fibre is a flexible, transparent fibre made by drawing glass (silica) or plastic to a diameter slightly thicker than that of a human hair.
 Fibres are used instead of metal wires because signals travel along them with less loss; in addition, fibres are immune to electromagnetic interference, a problem from which metal wires suffer.
-
+The concept of a direct-attech copper cable also exist with optical cables.
+These are then called \acf{AOC}.
 
 \begin{table}
    \caption{Fibre optic categories}
@@ -245,6 +265,13 @@
 \mode<article>{
 The \SI{2.4}{\giga\hertz} frequency supports eleven channels in the United States and thirteen channels in Europe and most of the world.
 The \SI{5}{\giga\hertz} frequency supports over fifty channels, depending on the bandwidth used.
+
+\begin{figure}
+\centering
+\includegraphics[width=.7\textwidth]{images/wifi-channels.png}
+\caption{Non-overlapping wireless channels in the \SI{2.4}{\giga\hertz}-range}
+\label{fig:wifi-channels}
+\end{figure}
 }
 
 \Paragraph{bandwidth}
diff --git a/images/anycast.png b/images/anycast.png
new file mode 100644
index 0000000..c720ad4
Binary files /dev/null and b/images/anycast.png differ
diff --git a/images/broadcast.png b/images/broadcast.png
new file mode 100644
index 0000000..41c9450
Binary files /dev/null and b/images/broadcast.png differ
diff --git a/images/dac-cable.jpeg b/images/dac-cable.jpeg
new file mode 100644
index 0000000..3b69dc3
Binary files /dev/null and b/images/dac-cable.jpeg differ
diff --git a/images/multicast.png b/images/multicast.png
new file mode 100644
index 0000000..b2f9f69
Binary files /dev/null and b/images/multicast.png differ
diff --git a/images/t568.jpeg b/images/t568.jpeg
new file mode 100644
index 0000000..ff16810
Binary files /dev/null and b/images/t568.jpeg differ
diff --git a/images/unicast.png b/images/unicast.png
new file mode 100644
index 0000000..8102f79
Binary files /dev/null and b/images/unicast.png differ
diff --git a/images/wifi-channels.png b/images/wifi-channels.png
new file mode 100644
index 0000000..5d12d13
Binary files /dev/null and b/images/wifi-channels.png differ
diff --git a/main-book.tex b/main-book.tex
index 37311cc..25f0433 100644
--- a/main-book.tex
+++ b/main-book.tex
@@ -8,8 +8,8 @@
 \usepackage[noamssymb,noamsthm,envcountsect]{beamerarticle}
 \setjobnamebeamerversion{main.presentation}
 
-\setfloatlocations{figure}{hbp}
-\setfloatlocations{table}{htp}
+\setfloatlocations{figure}{hbtp}
+\setfloatlocations{table}{htbp}
 
 % Lists
 \usepackage{multicol}
diff --git a/main.tex b/main.tex
index 2c0b405..9e1dc68 100644
--- a/main.tex
+++ b/main.tex
@@ -73,6 +73,8 @@
    \input{tex/titlepages}
    \clearforchapter
    \tableofcontents*
+   \listoffigures
+   \listoftables
    \input{chapters/preface}
    \printglossary[type=abbreviations,style=mcolindex]
    \mainmatter
diff --git a/tex/abbreviations.tex b/tex/abbreviations.tex
index d085e74..1d744d8 100644
--- a/tex/abbreviations.tex
+++ b/tex/abbreviations.tex
@@ -1,4 +1,5 @@
 \newabbreviation{8P8C}{8P8C}{8 position 8 contact}
+\newabbreviation{AOC}{AOC}{active optical cable}
 \newabbreviation{AP}{AP}{access point}
 \newabbreviation[
    longplural = {automatic private \acs{IP} addresses},
@@ -20,6 +21,7 @@
 \newabbreviation{CLI}{CLI}{command-line interface}
 \newabbreviation{CRC}{CRC}{cyclic redundancy check}
 \newabbreviation{CRT}{CRT}{cathode ray tube}
+\newabbreviation{DAC}{DAC}{direct-attach copper}
 \newabbreviation{DAI}{DAI}{dynamic \acs{ARP} inspection}
 \newabbreviation{DARPA}{DARPA}{Defense Advanced Research Projects Agency}
 \newabbreviation{DEC}{DEC}{Digital Equipment Corporation}
@@ -123,6 +125,7 @@
 \newabbreviation{UTP}{UTP}{unshielded twisted pair}
 \newabbreviation{VID}{VID}{\acs{VLAN} identifier}
 \newabbreviation{VLAN}{VLAN}{virtual local-area network}
+\newabbreviation{VLSM}{VLSM}{variable-length subnet mask}
 \newabbreviation{VPN}{VPN}{virtual private network}
 \newabbreviation{VSTP}{VSTP}{\acs{VLAN} Spanning-Tree Protocol}
 \newabbreviation{VTAM}{VTAM}{Virtual Telecommunications Access Method}