62
Arctic Yearbook 2015
From structured information to visualization
To begin we needed to find a platform upon which to develop the flexible and user-friendly tool. Mike
Bostock6 created D3.js, which is unlike classic visualization libraries in that it allows information to be
communicated in much more powerful ways. The almost endless options for data representations
make the library itself a tool. From the D3.js library we chose a visualization based upon hierarchical
edge bundles (Holten 2006). The data are represented in a wheel format, which is composed of nodes
and edges all coloured in light grey. The data are grouped into eight bundles: seven represent the
stakeholders organized by the scale(s) at which they operate (see Table 2) and the eighth represents
the interests. The nodes within each bundle are the stakeholder groups that operate at those particular
scale(s), and each is linked to its interests (Figure 5).
Figure 5. The foundation of the visualization tool showing all connections between stakeholder groups, and between
groups and their interests.
Each node (SG or interest) acts as both a source and target. When the user passes their cursor over
either a stakeholder group or an interest (the source), it becomes highlighted in black. The colour of
the edge (connecting line) and the target(s) will be different, though, depending on whether the
selected source node is an interest or a SG. If the user passes their cursor over a SG, the edge(s) and
the target node(s) become highlighted in red if it is an interest, or in purple if it is another SG (Figure
6a). In contrast, if the user passes their cursor over an interest, the edge(s) and the target node(s) will
become highlighted in blue (Figure 6b).
Maritime Activities in the Canadian Arctic