Tufte – Info Graphics From Galileo to Today (in 6 hours)
I am writing this to attempt to catch up to myself. I have been attending some great events over the past few months, and Edward Tufte’s one-day course in DC marked the first of three excellent conferences I’ll be trying to document here. (more on the others soon)
In the interest of time, I’ll be brief, but let’s start by saying that Tufte is an encyclopedic mind and a fantastic presenter. If you every had any doubt that a single person could cover the history and analysis of information graphics spanning from the time of Galileo (1613) to today, structure it for presentation in six hours in a meaningful way, attend his course. Secondly, like James Brown is the Godfather of Soul, E. T. is the Grad Daddy of information graphics. For info geeks like myself, he’s a hero.
From my notes:
9.28.09 | 8:45 AM | Crystal City Marriott
E. R. Tufte walks over and leans against a nearby credenza outside the ballroom as we open our books and start our “read-ahead assignments” while we wait to enter. He cracks a smile and jokes, “Nice books.” He then lets us know that we can take our time, but the room was open if we wanted to go ahead in.
I entered and sat in the second row, just beyond the right-most of the two projectors. I am one of 4 people in the room.
He walks straight up, reaches into my box of books, and grabs one out, beginning to unwrap it himself.
He turns to page three of The Visual Display of Quantitative Information and signs his name, while he asks two rapid-fire questions at once, “What do you do? Where do you work?”
Here are my take-aways in list form:
- Always Remember: Do whatever it takes to explain something.
- Don’t ask, “How can I use data visualization / charting to understand X?”
- Ask, “How can I understand X” (the method of explanation will make itself clear)
- In diagramming: linkages (links) = verbs. Annotate and label them clearly and simply.
- Avoid (and be on the lookout for) “Chart Junk” (Tufteism)
- “You never need boxes around things. Ever.”
- You need to include signs of credibility (data) – the reason to believe what is being presented.
- “Keep an open mind, not an empty head” (Tufteism)
- People are very good at finding relevant information in a complex visual field. (newspaper sports pages)
“Something is always relevant to someone”
- Assume your audience is smart…or smarter than you (they managed to find themselves in a position of authority somehow – how dumb can they be?)
If they aren’t, make them so.
- “There is no such thing as information overload, only bad design” (Tufteism)
- On Supergraphics: Supergraphics are important for many reasons, one of which is that there are many stories or “hooks” in them for the viewer.
- On figure ground vibration between design elements: 1+1=3 (first line + second line + the interaction between the two)
- On “fudging figures and data”: Local “Optimizing” = Global “Pessimizing” (Tufteism)
- On Interfaces: make them 93% content and 7% administrative (and make it minimal)
- Use Trebuchet for tables on the PC (if you have to) (Tufte is a Mac guy)
- Sort tables by Performance, not alpha.
- Don’t get it original, get it right.” (Tufteism)
- Develop a good routine for presenting performance data.
- Find a supergraphic you can use.
- The intellectual model for non-fiction reporting should reporting. Look at science and nature for inspiration.
- Add data to help with credibility…and DON’T LIE.
- For “serious stuff” it’s better to spread stuff across space, rather than stack it in time. (Spacial Adjacency)
- If you want to understand something more, increase the resolution. (Tufteism)
- On PowerPoint (don’t get him started!): PowerPoint slides represent a breaking point (interruption) within a flow of important information.
- Information resolution is key.
- Work to maximize viewer cognition time and minimize viewer “figuring out” time. (“Eliminate impediments of content”)
- A way to escape “flatland” (the confines of 2D paper space) is to show real object / 3D item whenever possible (Galileo’s Istoria e Dimostrazioni intorno alle macchie solari included an actual pop-up figure)
- As a consumer, be on the lookout for information “cherrypicking,” blocks to original sources, glibness, smugness, lack of backup information, comparisons, causal an multi-dimensional information. Look for documentation.
- E.T.’s law: 97% of content in a given field is junk. Our job is to find the other 3%. (Tufteism)
- On Small Multiples (diagrams using small representations or figures to explain): Small multiples are useful on many levels for comparisons, multivariate data and reliability.
- The idea of “Know your audience” can often lead to underestimating them. Instead think, “Know your content. Respect your audience.” (Tufteism)
- Remember Conway’s Law relative to design: BAD organizational structure dictates structure of design.
Tufte on “How to give a good presentation”:
- Only use PowerPoint as a tool, not a platform for presenting
- Give a handout (you can still use the screen for movies, charts, images, etc)
- Use sentences
- Handout: Side 1 = Supergraphic; Side 2 = State problem, proposal, solution, resolution, etc.
Tufte on “How to prepare for a presentation”:
- Eliminate Items that damage content
- Make content better
- Practice (use videotape if you have to)
- Show up early
- End early
The minute he said “end early,” he was done…three minutes early.
After the course, I spoke to him again at his merchandise table as I sorted through a pile of signed press sheets, from which I scored two excellent forms from the Envisioning Information print run.