Takeaways from Presenting Data and Information (Edward Tufte)
January 26th, 2009 by
Russell Wilson
I just finished Edward Tufte’s one day course, Presenting Data and Information, and here are some key takeaways:
He hates Powerpoint! (okay, everyone knows this)
Keep an open mind but not an empty head.
Talent imitates genius; genius steals.
Annotate everything!
The two (2) biggest challenges/considerations in presenting data
- any interesting problem is multivariate
- information resolution
The three (3) key factors with regard to content
- quality
- relevance
- integrity
The seven (7) fundamental principles of analytical design
- Show comparisons, contrasts, differences
- Show causality, mechanism, explanation, systematic structure
- Show multivariate data
- Completely integrate words, numbers, images, diagrams
- Thoroughly describe (document) the evidence
- Analytical presentations ultimately stand or fall depending on the quality, relevance, and integrity of the content
- (Not in book) – Use adjacency in space – don’t time stack data when you can display it side-by-side
Many, many more pearls of wisdom and insightful information that has had a strong impact on my work. This is meant to be a teaser to lead you to explore his writing more (if you haven’t already).
Great course. I took it several years ago. Highly recommended for anyone involved in data presentation.
[...] thoughts here. How/where/do these [...]
Yes, great lecture.
A few things I wanted to add:
(1) “Whatever it takes” attitude:
Using the best process to fit task. Many times requires
multiple processes
(2) Don’t deny info:
One of the critiques of Powerpoint….
(3) USE the phenomenal abilities of the eye to compute lots of data
at once:
Don’t take info away to make a design readable. We can handle
lots of information… However, may require a more thought-
through design of the multivariate info
(4) Maximize content reading time, minimize content figuring out time
(5) Do Inside out design, not outside in design:
What is required to make a product work should be the biproduct
of what is needed. Don’t use given solutions and work a product
to fit those solutions.