A data design process of five steps based on Ben Fry’s universal process, combined with the narrative-centered design concepts of Ricardo Mazza:
In addition, our general visualisation approaches include:
Filtering/highlighting/sorting filtering and/or highlighting and focusing can be used to select a subset of elements to reduce visual clutter; similarly, sorting of elements allows to restrict the focus on a subset, utilising a metric for the purposes of ranking.
Highlight highlight a subset of elements to identify, eg elements forming part of a policy violation.
Exploiting visual form and representations utilise visual form and well-known representations to allow quick and high-level recognition to humans.
Using abstractions use abstractions in the set of elements to allow grouping ‘similar’ elements and combine into fewer elements in order to visualise effectively.
Overview and drill-down give an overview of the total system, possibly starting with higher-level abstractions of subsystems, while allowing drill-down into individual subsystems to show more detail.
Multiple views show multiple views of the system from different viewpoints or ‘gazes’ to highlight different aspects of the system at the same time in a coordinated-visualisation (see also North, Chris and Shneiderman, Ben, Snap-together Visualization: Can Users Construct and Operate Coordinated Visualizations?).
However, we also need to consider the focus of the narrative and the nature of the target user communities. We have therefore further applied Riccardo Mazza’s, principles:
Problem: What is the main purpose of the visualisation? Is it needed for reporting purposes, used for exploring the dataset in order to find new information, or is the purpose to confirm an assumption or prove a hypothesis?
Data type: What type of data needs to be visualised? Is it nominal, ordinal, interval, or ratio data? (Mazza combines interval and ratio types under the label quantitative.)
Number of dimensions: How many dimensions need to be examined using the visualisation? These are defined as the number of independent attributes (the attributes that vary with respect to one or more independent attributes).
Primary structure of data: What is the structure of the data we need to visualise? Are there simple values, or are we primarily interested in temporal aspects of the data? Is it of spatial (physical extents), hierarchical, or network structure? Are we interested in a distribution of values?
Type of interaction: How much interaction is needed for the task? Can we use a static display? Does the user need to be able to transform the data prior to visualisation, or manipulate display attributes like colour or zoom-level?