Last month I got some travel funding from work to fly to Atlanta and attend a one-day seminar about data presentation, graphics and charts. It was conducted by Edward Tufte, who wrote the seminal book The Visual Display of Quantitative Information, which I recently named as one of my most influential books.
Tufte has been touring and conducting these seminars for several years now, and he’s become something of a rock star in information geek circles. Attendees receive copies of all 4 of his lavishly illustrated books which is an added bonus, and is also important to his presentation style. One of his central presentation concepts is that computer displays and powerpoint projections are extremely ineffectual in displaying complex multivariate information simply because they are of such poor resolution. He’s comparing the resolution with that which is possible on a printed page or large sheet of paper, and the capacity of the human eye and mind to perceive and absorb and analyze visual information. Therefore, his presentation style involves minimal use of projection, and he instead asks attendees to refer to specific illustrations in his books to support his lecture.
He also brought out and shared centuries old book editions by Galileo and Euclid to illustrate some of his concepts.
As you can imagine, it was a very rich and detailed seminar. Tufte started the seminar with reading assignments, and held ‘office hours’ during breaks – which were really book signings, although there was some substantive discussion during these breaks.
After I returned, I invited UCF colleagues to a 1hour presentation of my own where I debriefed and led a discussion about some of his concepts and ideas. One good suggestion I brought back was to create a supergraphic. That’s a large scale printed handout that meets several criteria. A supergraphic should:
- engage the reader, it should contain detailed information of particular interest to the audience
- contain lots of detail, Tufte says “to clarify, add detail, its not information overload if it is well designed”
- provide information from multiple sources
- show multivariate analysis and comparisons
- show causality, mechanism, structure or explanation
- be credible, and include documentation of source material
A supergraphic is turning out to be a technical and a conceptual challenge, but I’ve got some ideas. If nothing else, I know this has changed the way I’ll use powerpoint.