Quartz 4

week1-introduction-art-of-making-data-speak

2 min read

Type: lecturenote

Up: 024_art_of_making_data_speak Next: week2-histories-of-data-visualisation

Histories of Data Visualisation

Reading Notes:

What is Visualization?

  • Core concepts & Definitions
    • Information Visualization (infovis) Provisional Definition: A mapping of data to a visual representation.
    • Computer Science Definition: Infovis is often equated with the use of interactive computer-driven visual representations and interfaces, such as “the communication of abstract data through the use of interactive visual interfaces.”
    • Infovis vs. Scientific Visualization:
      • Scientific Visualization often deals with numerical data and commonly uses 3D-shapes or volumes (voxels). It developed in the 1980s alongside 3D-computer graphics.
      • Information Visualization is sometimes distinguished by its focus on non-numeric, non-spatial, and high-dimensional data, though in practice, many Infovis projects still use numbers. It often uses 2D and vector graphics (points, lines, curves). It developed in the 1990s alongside 2D-graphics software.
    • Infovis vs. Information Design:
      • Information Design starts with data that already has a clear structure, and its goal is to visually express this structure (e.g., the London Tube map).
      • Information Visualization aims to discover the structure of a data set; this structure is not known beforehand.
  • Two Key Principles of Traditional Infovis (18th Century to Today)
    1. Reduction: Infovis uses graphical primitives (points, lines, simple geometric shapes) to stand in for objects and their relations. This involves an “extreme schematization” of data to reveal shared patterns.
    2. Spatiality: Infovis maps the most important dimension(s) of the data onto spatial variables like position, size, and shape (or layout). Other visual dimensions, such as color, tone, and transparancey, are typically reserved for secondary or categorical aspects (e.g., functioning as a label). This practice likely reflects human visual perception, which privileges the spatial arrangement of elements.
  • Direct Visualization
    • Definition: A newer approach (since the mid-1990s) that creates new visual representations from the actual visual media objects or their parts, rather than abstracting them into graphical primitives.
    • Core Principle: It preserves the original form of the data and builds the visualization directly out of it.
    • Examples:
      • Tag Cloud: Varies the font size of words (the actual media) to represent their frequency, rather than using a bar chart (a new visual representation).
      • Cinema Redux: Uses scaled-down still frames (actual media) from a film, arranged sequentially, instread of using a single color or shape to represent each frame.

Listener’s Notes:

  • Date:
  • Paper:
  • Main arguments and concepts:
  • Questions:
  • Notes:
  • **Key areas & problems:

Graph View

Backlinks