Lessons from Tufte

Edward Tufte is a leading thinker on information display. His courses and books offer many insights for creating effective presentations and slides. I pulled from a couple of sources (Al Globus at NASA and Signal v. Noise) to create this list of Tufte principles:

General Principles

  • Quantitative thinking comes down to one question: Compared to what?
  • Good design is clear thinking made visible, bad design is stupidity made visible
  • Never harm the content — the design should be based on the content, not the other way around
  • Frame your presentations: What’s the problem; who cares; and what’s your solution
  • Be content focused

Presentation of data

  • Try very hard to show cause and effect
  • “It’s better to be approximately right than exactly wrong"
  • Visualization excellence is nearly always multivariate
  • Tell the truth about the data
  • Encourage the eye to compare different pieces of data
  • Reveal the data at several levels of detail, from a broad overview to the fine structure
  • Display an accessible complexity of detail

Valuable advice for creating good slides

  • Don’t break up evidence by accidents of means of production
  • The way you reduce clutter is to clarify the design and then add information
  • 1+1=3… Two elements in close proximity can create a third “ghost image" from the negative space between the two elements
  • If a chart or table or object needs a label, label it inline — don’t use legends/keys that require “back-and-forths"
  • Visualization excellence is that which gives to the viewer the greatest number of ideas in the shortest time with the least ink in the smallest space
  • Induce the viewer to think about the substance rather than about methodology, graphic design, the technology… or something else
  • Serve a reasonably clear purpose: description, exploration, tabulation, or decoration
  • Use words, numbers, and drawing together
  • Reflect a balance, a proportion, a sense of relevant scale
  • Have a narrative quality, a story to tell about the data
  • Avoid content-free decoration
  • Comparison rather than mere description