topics

Who Cited Envisioning Information by Edward Tufte?

Tufte's three books are now widely cited by those involved in the design of information, especially statistics, that are presented on screen or printed. It's not easy to work out whether they are cited more because the books themselves are beautiful objects and their contents inspiring or because of the analysis, theory and techniques described therein and the examples they reproduce.

• Who in HCI has been citing Tufte?
• What kinds of papers are citing him?
• What is he cited for - What makes someone cite Tufte?

Here's a first attempt at creating a picture that explores some aspects of those questions. The diagram represents papers published between 1990 and 2009 that are listed by the ACM Portal as citing Envisioning Information.



Clicking the image above will open a new page with magnified views.

The diagram attempts to show:

1. An icon for each paper citing Envisioning Information
2. The author's name for each paper
3. The year that each paper was published
4. A dot for each reference in the text
5. A coloured dot for a reference that uses one of the following keywords:
   • colour
   • chartjunk
   • data-ink ratio
   • small multiples
   • micro / macro
   • layering and separation
6. A pale yellow banner when the citation is in a section called 'Introduction;
7. Papers citing Graphic and Graphic Information Processing by Jacques Bertin are labelled Bertin'81
8. Papers citing La Semiologie Graphique (or a translated version) are labelled Bertin'83
9. Papers are organised so that they appear in the following groups:
   • Those that reference one or more of the chosen keywords
   • Those with references in the introduction
   • Those with references (neither keywords nor in the introduction)
   • Those with no specific reference in the body of the text
   • Those that are not available to the public for free and thus have not been checked

Paperless Office

This fantastic study by Sellen and Harper first inspired me in 2003 and I still consider it to be a great place to be reminded about the power of 'affordances'. It is particularly poignant for anyone who is involved in the design of a software application that attempts to emulate or replace tasks that are currently carried out on paper.

Affordance

If you are new to the term 'affordances', here is a brief introduction. The term was coined by the perceptual psychologist James J Gibson and then introduced to the HCI community by Don Normam in his book The Psychology of Everyday Things.

My favourite way to think about affordances is to consider opening doors. Say you encounter a door with a handle that suggests you should pull to open it and you discover that you actually have to push it. You have encountered a poor use of an affordance. A door with a handle suggests 'pull', whilst a door with a flat plate suggests 'push' hence: the handle 'affords' pulling. Similarly, in a software interface the furniture (such as links and buttons) suggests to the user what function it represents and implies how one should interact with it.

Edward Tufte

Mister Tufte refuses to fit into the traditional boxes. That makes him both troublesome and powerful. He has assimilated a vast collection of beautiful examples of successful representations of information. They include two-dimensional, paper-based designs, as well as objects that involve interaction. Tufte continually stuns the academic world because the pieces in his collection are beautiful and they are presented in the most aesthetically pleasing way.

Tufte can teach us a lot for another reason. His research forms something akin to (or more like a grandfather to) a study of 'best practice' from which he has drawn, not rules but new terms (he has coined a few), principles, guidelines. You won't suffer a '3-click rule' from Tufte.

The 3-click Rule

The 3-click rule states that a website should be designed so that a user finds their content within three clicks. The rule has been used and abused and even though those in the usability community are well aware of its limitations, it is still around. The rule is powerfully pervasive because it is easy to communicate and to apply. Rules are great, they turn us all into experts who can speak with conviction and authority in meetings because we know the 'rules'.

The three click rule is a simplifcation of a perfectly sound principle put forward by Jakob Nielsen, the man in HCI whose name is most frequently mis-spelled. Those of us who learned the rule: " 'i' before 'e' except after 'c' excepting 'seize' and 'protein', were all doing fine until we learned there are more exceptions and that a 'rule' isn't really a 'rule'. So it is with the 3-click rule.

For a first time user, a click is a choice or a guess about the information that will be presented as a result of clicking. Choices are complicated things. There's a wealth of literature in psychology to help decipher what might be going on in a choice. When making a choice you'll need to consider how many things there are to choose from, what you think each thing represents and what influence everything else surrounding the choice may have. Thus you begin to consider the cognitive weight of each choice. Or click.

In the world of websites, finding your way to a particular bit of information might involve a large number of clicks. If those clicks are really easy choices that are completely expected, you'll go right ahead, click-click-click-click-click-click and you're there. If on the other hand, the clicks you need are not where you expect them to be or the thing you're looking for could be in more than one of the choices presented to you then you'll be much slower with each click. Of course this is just one of the parameters and the way that you might measure the cognitive weight of a click will change depending on a number of parameters. However you look at it, the 'Rule' certainly doesn't 'rule'.

About me

Sarah Parker: freelance user experience architect based in the Thames Valley, UK

Contact me