Pie Are Squared

Anil Dash and Kieran Healy explain that some publications have taken a shine to square pie charts. Here's the example they've both posted (the left chart taken from the New York Times; the right, from Wired):

Stylish, maybe, and the pie format certainly has its shortfalls, but Healy explains why square doesn't take the circle:

The main problem with this [square] style of presentation is that it uses two dimensions to display unidimensional data. As the graphic on the right, especially, makes clear, the layout of the subcomponents of the graph is arbitrary.

My first thought was about the use of color, and how color (and of course, pattern) could be used by scheming editors and their nefarious art departments to sway in subtle ways a reader's appreciation of the graph. If you were to remove the data tags, after all, you would have forms that would read in appreciable ways to a viewer.

Josef Albers, White Line Square VII, 1966.

What would Josef Albers say? Here are some potentially pertinent lines from a 1964 concrete essay, "The Origin of Art":

THE ORIGIN OF ART:
The discrepancy between physical fact
and psychic effect

THE CONTENT OF ART:
Visual formulation of our reaction
to life

THE MEASURE OF ART:
The ratio of effort to effect

THE AIM OF ART:
Revelation and evocation of vision

Wordy guy, that Albers. (That's the long and short of that essay.) In another essay from the same year, titled "The Color in My Paintings", Albers expands on the function of color within the set pattern of his "homage to the square" series:

[Colors] are juxtaposed for various and changing visual effects. They are to challenge or to echo each other, to support or oppose one another. . . .

[ . . . ]

Such action, reaction, interaction—or interdependence—is sought in order to make obvious how colors influence and change each other: that the same color, for instance—with different grounds or neighbors—looks different. . . .

[ . . . ]

Such color deceptions prove that we see colors almost never unrelated to each other and therefore unchanged; that color is changing continually: with changing light, with changing shape and placement, and with quantity which denotes either amount (a real extension) or number (recurrence). And just as influential are changes in perception depending on changes of mood, and consequently of receptiveness.

All this will make [us] aware of an exciting discrepancy between physical fact and psychic effect of color.

Of course, all the concrete abstract painters working at this time were thinking experimentally and focusing on the theatricality of abstract painting, though I'd say that Albers was one of few to do the science right and narrow that focus down to a single variable. Given his belief that color value is contextually determined, I doubt Albers would agree that color could be arbitrarily assigned, which has perhaps not a whole lot to do with data presentations so long as you are trying to have a very serious Friday.

Posted by Kriston at August 3, 2007 11:59 AM