Skip to content
 

The pretty picture is just the beginning of the data exploration. But the pretty picture is a great way to get started. Another example of how a puzzle can make a graph appealing

Ben Hyde sends along this appealing image by Michael Paukner, which represents a nearly perfect distillation of “infographics”:

Here are some of the comments on the linked page:

Rather than redrawing the picture to make the lines more clear, I’d say: leave the graphic as is, and have a link to a set of statistical graphs that show where the different sorts of old trees are and what they look like. Let’s value the above image for its clean look and its clever Christmas-tree design, and once we have it, take advantage of viewers’ interest in the topic to show them more.

P.S. See my comment below which I think further illuminates the appeal of this particular tree.

6 Comments

  1. Adam says:

    I got a headache tracing those lines back. Pretty nonetheless.

    • Andrew says:

      Adam,

      The headache is, I believe, part of the point. First, if the lines were direct you wouldn’t get the pretty Christmas tree pattern. Second, the investment required in following the lines makes you appreciate what you’ve learned. Third, the curvy lines are themselves a puzzle; as you trace them, you gradually learn the meaning of the y-axis.

      • Jessica H. says:

        I would agree that it’s a nice example of an aesthetically appealing design (through the symmetry and low contrast palette) and apt metaphor (the tree shape created by the crossing lines) being used to balance what might otherwise be a relatively mundane data set, at least to users without any specific prior interest in trees. I would bet that if you took away either of these first two characteristics you’d have a less engaging, if more direct, graph, but one that might get less views, likes or shares online

        The ‘puzzle’ aspect brings to mind Laura Novick’s work on visualizing biological trees. She has a few studies that help demonstrate how the perceptual difficulty can actually be a good thing if it’s in service of a metaphor that characterizes the data (kind of like your observation about the y axis being understood as one traces the crossing lines). In the case of visualizing phylogenies in biology, her work has offered some evidence that while ladder formats use the Gestalt principle of good continuation, cladogram formats which go against this principle actually lead to better comprehension in students. The increased learning from the perceptually disjoint format is higher when the student has less background knowledge on biology, getting at the way that the expertise of the user is clearly going to matter (you wouldn’t want to give this graph to an expert in ancient trees, for example). I wouldn’t be surprised if this Christmas-tree graphic also led to better learning of the tree types and origins than a less puzzling version that lacked the metaphor.

      • Jeremy Miles says:

        I gave up before I learned the meaning of the y-axis. :(

  2. Jerzy says:

    If “the investment required in following the lines makes you appreciate what you’ve learned,” what do you actually learn by following the lines?

    If you squint carefully you MIGHT be able to learn:
    * the location of an old tree with a particular name or age
    * the old trees closest to where you live

    But adding some simple interactivity (e.g. bolding the line on mouseover) would make these tasks easier, and I don’t think it’d make you appreciate the knowledge less.

  3. John Mashey says:

    Puzzles are fine, but drawing a graph that looks nice but actually makes some relationships hard to see isn’t so fine. This is all too reminiscent of architects that build beautiful buildings where thsoe who work there end up finding them miserable to be in. For sure, http://moyhu.blogspot.com/2011/07/more-proxy-plots-with-java.html“>mouseivers like this are helpful, but if the image really is static, there were any of several ways to make it easier to follow the lines, especially in the dense areas.