Visualize This!

 Posted by on Sep 1, 2009  Add comments
Sep 012009
 

A few days ago, a management consultant I follow on Twitter commented critically about a certain software vendor rating method, dubbed the “Magicke Quadrante” in this satire on the Web (shhh… like Voldemort’s, certain names shall not be uttered in this forum). He asked: “Were x and y coordinates so firmly etched in our brains that we can’t get away from it?”. It was hard to reply meaningfully in 140 characters, but I wrote: “it may be fairly simple: 1 rating dimension is rarely enough, >2 are hard to visualize, so we converge on 2, add thresholds = MQ”

Not coincidentally, another Twitter correspondent pointed the next day to the “Periodic Table of Visualization Methods,” a remarkable classification work that lists 100 ways to display data, by Prof. Martin Eppler, of the Swiss-Italian University in Lugano, who leads the “Visual Literacy” e-learning project. When you look at the Web page, don’t go away too quickly: mouse over the boxes, and see how examples of each type of diagram pop up. Eppler’s work is somewhat reminiscent, and is perhaps consciously inspired by the work of Edward Tufte (especially his famous book “The Visual Display of Quantitative Information”). And sure enough, one of the”elements” in the periodic table is a simple one-dimension continuum chart, and another one, which the author calls “cartesian coordinates,” is the (in)famous four-quadrant partitioning of the space being analyzed.

Representing more than two dimensions on paper requires, one way or another, the recourse to some trick to map the excess dimensions down to the flatness of the paper or screen. Color, of course, can be used to add a dimension. But when animated holograms become good enough and cheap enough (for example, as a result of Michael Bove’s work at the MIT Media Lab) then we may start to routinely view with ease three- or four-dimensional data, and then we can send the old trusted quadrant to meet the abacus at the Smithsonian Museum.

Caesar’s “trip report” on one of his campaigns was purportedly the concise “veni, vidi, vici” — I came, I saw, I conquered. With the second word, he was certainly not thinking of visualization the way we do today, but it is clear that choosing how to display data meaningfully is a very important step before being able to make a winning move. Quadrants may be serviceable for some purposes, but Prof. Eppler’s work does us great service by pointing out that there are at least 99 other ways to view information for decision-making purposes.

avatar

Claude Baudoin

Claude Baudoin is a Senior Consultant with Cutter Consortium's Business & Enterprise Architecture and Data Insight & Social BI practices. He is a proven leader and visionary in IT and knowledge management (KM) with extensive experience working in a global environment. Mr. Baudoin is passionate about quality, knowledge sharing, and providing honest and complete advice.

Discussion

  2 Responses to “Visualize This!”

  1. Good post. Because of the rampant spread of the magic quadrant syndrome (excessive and blind reliance on “The Quadrant”), I have come to deeply distrust the number 4. I have also come to distrust the belief that all good things are upwards and to the right. I now tend to associate anything in that quadrant with things that float away. I think I am understanding how lefties feel (I am right-handed).

    To this day, as I watch vendor after vendor slideware, I am anxiously looking for anything based on the number 5, 7 or 9. These days, spider charts are a welcome relief.

  2. Great link to the Periodic Table of Visualization Methods — very interesting. But I didn’t see the work of Hans Rosling and Gapminder (http://www.gapminder.org/) in the table. I wonder if you’ve seen some of their novel visualization methods? http://www.ted.com/index.php/talks/hans_rosling_shows_the_best_stats_you_ve_ever_seen.html

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

(required)

(required)