[Editor's note: Are increasingly complex graphics (visual confections / data visualizations), aka chart porn & eye/brain candy, necessarily good at conveying essential information to readers? Alberto Cairo says no, see below. I have to wonder if he means just for daily minute-read journalism or for publications that have a tradition of offering readers, say, tear-out map supplements and posters they like to admire and reexamine. This might have been raised by the 2008 box office graphic from the NY Times, implementing a steamgraph, but the general question applies to all infographic professionals. We have awesome new tools and datasets, but we still need to focus on fundamentals, beyond interesting and visually appealing, to consider what as Lee Byron calls the "interplay between considerations of aesthetics and legibility."
From Edward Tufte (p121 in Visual Explanations: Images and Quantities, Evidence and Narrative):
A confection is an assembly of many visual events, selected ... from various Streams of Story, then brought together and juxtaposed on the still flatland of paper. By means of a multiplicity of image-events, confections illustrate an argument, present and enforce visual comparisons, combine the real and the imagined, and tell us yet another story.
Tufte actually shows a similar chart to the NYT box office example in the same work from the cover of "Rock 'N' Roll is Here to Pay: The History and Politics of the Music Industry". He says of the Rock and Roll example: "The multiple, parallel flows locate music-makers in two dimensions – linking musical parents and offspring from 1955 to 1974, and listing contemporaries for each year."
But what fundamentals and what audience? A later NY Times graphic on How Different Groups Spend Their Day resolves part of the box office problem with an ability to focus (isolate) parts of the stream into a standard axis chart and thus read and compare quantitatively. Much less complicated, but also on a standard axis: Jackson’s Billboard Rankings Over Time. I don't entirely buy the similar axis baseline argument as I'm a fan of graduated circle maps, but reading them is certainly not for the average reader and must be augmented with summary charting and text.
But the box office chart is more approachable (read fun and inviting) then simple charting per Why Is Her Paycheck Smaller. But I think simple axis presentation of Taking apart the federal budget is easier to read at first glance then the multi-axis Obama’s 2011 Budget Proposal: How It’s Spent, even if I use the same approach in the Washington Post's Potus Tracker to analyze Obama's schedule. At the core is trying to make increasingly complex datasets (2) grockable visually. Have fun figuring it out and in the meantime enjoy this XKCD comic ]
Republished from Visual Journalism.
Alberto Cairo is a former professor at Chapel Hill, where he taught online graphics. Before that he was the succesful graphics director at El Mundo Online. In short: This guy knows what he is talking about, when it comes to both printed and online graphics …
Today Alberto took a critical look at the current data-visualization-trend, which got a huge boost last year, when New York Times took home a gold medal and even the ‘Best of Show’-award for a certain Box Office graphic. Regular readers of this site will remember the discussion we had last year after the awards. Click here to read it again.
Emperor’s new clothes
What Alberto is saying out in the open now, has been floating around for a while. But apparently only a few dare to say that graphics are getting too complicated – for fear of being looked upon as stupid, if they dare challenge such abstract wonders. I can’t help to think of the Emperor’s new clothes, whenever such a situation takes place.