Nature of Visualisation
October 7, 2011 § Leave a comment
Visualisation is “the process of forming a mental vision of important aspects of some situation that is not naturally visible”.
By visualising data, we harness the vast capabilities of our human visual cortex, arguably our most powerful and naturally accessible computing system. We can process images and audio quickly, however, writing reduces our process time significantly. Three core benefits to visualising data:
• Communication – quick and effective transmission of complex messages
• Insight – clearer understanding of what is already known
• Discovery – reveal new facts and relationships
There are four different types of data, each with different optimal representations:
• Nominal – Unordered (simple labels e.g. Artefact Type)
• Ordinal – Ordered, Unequal Intervals ( e.g. Condition )
• Interval – Ordered, Equal Intervals (e.g. Dates)
• Ratio – Interval Variables (values expressed as multiples e.g. measurements)
Charles Minard’s (an engineer) map of Napolean’s 1812 Campaign is especially famous because it shows several things at the same time.
It is: – spatially correct – represents size of army – represents temperature.
The map simplifies a complex multi-dimensional process into a compact and insightful visual representation. Through the represented relationships, a story is told. (The complexity involved in the initial abstract of significant features and compacting such multi-dimensional data should be acknowledged, a fact that is often forgotten by novice visualisations.)
Co-Ordinate Pairs, were becoming commonplace by the 16th Century.
Data Graphics, and many modern graphical designs were created by mathematician Johann Heinrich Lambert and political economist William Playfair, in the 18th Century. Some of Playfair’s works;
There are many different types of graphs ranging from illustrative and organisational diagrams, to maps and plans, and statistical graphs (of which there are subcategories).
Illustrative Diagrams are commonly used to show a complex object broken down into its component parts. They graphically portray an object in a simplified form, trading between realism and abstraction.
Organisational Diagrams emphasise the relationships between objects or parts of a single object (regular real or abstract). There is no clear boundary between illustrative and organisational diagrams.
Maps and Plans can be representational of spatial locations, and include additional numeric information encoded in some graphical fashion.
Statistical Graphs represent one of the largest greatest levels of abstraction in visualisation. William Playfair’s great achievement was to introduce entirely abstract forms of graphical data display, graphs that are so ingrained in our culture we hardly notice them nowadays.