What is 'visualisation' and why is it important?

"The field of computer-based information visualization draws on ideas from several intellectual traditions: computer science, psychology, semiotics, graphic design, cartography, and art. The two main threads of computer science relevant for visualization are computer graphics and human-computer interaction. The areas of cognitive and perceptual psychology offer important scientific guidance on how humans perceive visual information. A related conceptual framework from the humanities is semiotics, the study of symbols and how they convey meaning. Design, as the name suggests, is about the process of creating artifacts well suited for their intended purpose. Cartographers have a long history of creating visual representations that are carefully chosen abstractions of the real world. Finally, artists have refined methods for conveying visual meaning in sub-disciplines ranging from painting to cinematography."
Tamara Munzner, Interactive Visualization of Large Graphs and Networks

What is 'visualisation'?

The word 'visualisation' is generally used in either one of two senses:

  • as an abstract noun, it is understood as the activity (and, often, the actual techniques) for creating visual representations (images, diagrams, models, animations) for the purpose of effectively communicating a message, such as an abstract idea or concept, that would be difficult to communicate by any other means (e,g, through verbal description)
  • as a concrete noun (and hence with a preceeding article), a visualisation is any such representation

Examples of graphical representations that correlate closely with linguistic (and hence conceptual) categories:

  • maps
  • music scores
    music score
    Why are notes 'high' or 'low'? why do we use spatial terms to describe differences in pitch? why are the 'high' notes high on the stave and the 'low' notes low? Language and graphical transcription together help us visualise music: the 'higher' the note acoustically, the higher graphically its position on the page.
  • organogram
    organogram
    An organogram (or organisational chart) is a graphical representation of the hierarchical structure of formal relationships between roles within an organisation. Note that seniority (rank) within the organisation is indicated by how high up in the chart a person is. X may have a "top job" within the organisation; Y may "work under" Z.

Why are we looking at 'visualisation' at this point of the course?

Because, in essence, all virtual reality is precisely about visualisation. In the first week of this course, we looked at 3D desktops, window managers, and filesystem browsers.

Why is 'visualisation' important?

The standard argument for visualization is that exploiting visual processing can help people explore or explain data.

Visualization today has ever-expanding applications in science, engineering, product visualization, all forms of education, interactive multimedia, medicine etc. Typical of a visualization application is virtual reality, using computer graphics. The invention of computer graphics may be the most important development in visualization since the invention of central perspective in the Renaissance period. The development of animation also helped advance visualization.

Types of visualization

  • scientific visualization uses a spatial layout that is implicit in the data. Examples: cartography; meteorology; molecular chemistry.
  • information visualization entails finding a spatial mapping of data that is not inherently spatial. Examples: spreadsheet, database, and in general any n-dimensional information (where n > 3).

Node-link graphs: any domain that can be modeled as a collection of linked nodes can be represented as a graph. Graphs (as mathematical constructs) have a natural corresponding visual representation as nodes and connecting links arranged in space. (NB: “the data in a relational database forms an undirected graph”, Ronald Bourret, ‘Defining XML Views over Relational Data’)

Cognitive plausibility: visual representations of small graphs are pervasive – people routinely sketch such a picture when thinking about a domain, or include pictures of graphs in explanatory documents. Visual depictions of graphs and networks are external representations that exploit human visual processing to reduce the cognitive load of a task. Endeavors that require understanding global or local graph structure can be handled more easily when that structure is interpreted by the visual processing centers of the brain, often without conscious attention, than when that structure has to be cognitively inferred and kept in working memory.

Domain Node Link
Cartography Locality Road
WWW Web page Hyperlinks
Computer File Path (in directory structure)
Genealogy Person Relationship
Taxonomy Entity ISA, AKO, PART_OF, etc
Lexicography (semantic network, dictionary) Word "a kind of", etc
Organisation Organisational role Reporting, c-of-c, etc
Museum Artefact n-dimensional context
Rich picture (SSM) Stakeholders Labelled relations
Linguistics (syntax) Word, phrase Labeled arcs in phrase-markers
Ethnography (kinship) Person Labelled kinship links

Issues in spatial representation of graphs: [1] proximity (length of arc), [2] size (bigger is more?), [3] cognitive metaphors.