Wes,

There is so much more to this story that you could have communicated. Important information has become buried in the design, and the little that can be seen cannot be seen clearly. If all you wanted to say was that “our government spends a massive amount of money on Defense in relation to other agencies,” this could have been said much more elegantly and clearly.

There’s plenty of room for new ways of visualizing quantitative information, but there is no room for new ways that don’t work. Innovations that work less well than what already exists aren’t useful. Attempts to innovate that ignore what we already know are misguided and wasteful.

You wrote:

We are entering a new era of exploring data that calls for new ways to visualize this data. And this new era includes visual artists as the engineers developing the visualizations. I embrace this and am sorry to see you do not.

As someone who wishes to support the new era of exploring data, you should know better than to draw this illogical conclusion. The fact that we have found your design ineffective does not mean that we don’t embrace the work of visual artists. I welcome contributions to the field of data visualization from any source, especially those involved in other areas of visual design. The criteria of effectiveness that I use to evaluate all data visualizations are the same, however, no matter whose contribution I’m reviewing. Artists who venture into the field don’t get to be evaluated on different terms. To produce effective data visualizations, you must understand how people see and how they think when interacting with graphics. You must take advantage of this understanding to create visualizations that inform. If they also entertain, all the better, but they must inform.

You wrote, “We believe there are many complex relationships in data and that traditional methodologies simply do not illustrate.” You’ll get argument no from us. Your design, however, does not address a complex relationship. You’ve unnecessarily complicated a relationship that is quite simple.

You agree that circular designs are overused in data visualization, but believe that they “work really well” in many visualizations that you create. That’s certainly possible, but I’ve seen no evidence of this. I invite you to post an example here and let the people who read this blog—primarily people who care a great deal about data visualization and know a thing or two about it—assess its effectiveness.

It is incredibly important that we find ways to engage people with data visualizations, but the engagement only matters if people are actually engaged with the data. Looking at the pretty circles of color, no matter how much they appeal to us, means nothing if we don’t walk away from the experience better informed. Not just better informed, in fact, but accurately informed as much as possible about the things that matter.

Steve

You suggested we use bar graphs and scatter plots. This and everything else you say suggests that you don’t understand the point of this visualization. I’m not disagreeing with what you say. There is a much more simplified way to represent this data. The point is that we explore compelling ways to represent information and this is one of those endeavors. I don’t think you understand the meaning of this. So I will leave it there and will let time take its course.

Wes,

You must not be familiar with my work, otherwise you would know better than to think that I don’t understand what it means to “explore compelling ways to represent information.” I’ve dedicated my life to this pursuit. Important information like the media’s misaligned priorities should be presented in ways that promote the clear understanding that is needed to compel informed action. Experiment all you want, but test your designs, not by asking people, “Do you like it?”, but by determining if they actually understand what they see and find it useful. And don’t be satisfied with saying little when you could say much more that people need to hear.

My first thought upon seeing this visualization was that I’d like access to the raw data so I could make a simple bar chart. This visualization prioritizes design so far above the information that most of the information is completely lost.

Beyond some imbalance between defense spending and reporting, I can’t tell what’s going on here at all. I’m not compelled to do anything (like re-tweet it, or write a letter to an editor, or write my representative) by this visualization because…well, there’s hardly any information I can draw from it.

I’m reminded of David Ogilvy’s section on “the cult of creativity” from the 2nd chapter of Ogilvy on Advertising. “If it doesn’t sell, it isn’t creative” and “Originality is the most dangerous word in advertising” are two adages I keep close at all times. Or, as Stephen put it a while back, “Most of what we can do to make the world a better place involves, not doing the unprecedented, but doing what matters and what works.”

We can learn a lot from advertisers who dedicated their lives to compelling their audiences to change their behavior (i.e. take out their wallets). It doesn’t involve dazzling people, it involves effective communication.

Rob,

A stacked bar graph is not an effective solution. Bryan of Perceptual Edge will post examples of useful solutions tomorrow.

The contest called for designers to make “government data more accessible and comprehensible to the American public.” Compelling and engaging designs that do not work do not help the American public to comprehend the data and should not have won this prize.

I hate to say it, but I believe that Wes’s mission is entirely different from the staff and disciples of Perceptual Edge.

In his post, he says his “…primary goal is to explore compelling and engaging ways that break away from traditional forms of visualizations.”, As if traditional forms are somehow flawed, outdated in some other way obsolete….

While this “exploration” is potentially eye-catching, it blatantly ignores accuracy and precision, arguably more important objectives of data visualization. I am sure some would find this graphic engaging and/or compelling, but it doesn’t function to present a decipherable representation of reality.

Unfortunately, it seems that (like all too many others things in the world) this is designed to create value for the reader – by substituting flash for substance.

Simple-mindedly distorting a stacked bar graph (itself a potentially poor choice) 360 degrees will no doubt capture the attention of innocent and uninterested bystanders, but the important story contained within the data has also most certainly been twisted beyond intelligibility. And I for one fail to see any value in that.

You know, I think Jonas hit this one on the head. The ‘rectangular’ area without the distraction of the pinwheels is informative, impactful and beautiful.

Try masking the pinwheels out, the graphic becomes immediately clearer without losing any of its aesthetic impact.

There is so much overlapping action it might be helpful to make the graphic slightly interactive (a highlight of path on mouseover for instance) to accentuate the dimension in focus.

My graphic is rather small. But it looks like to LPs colored by a 1970’s hipster feeding into each other lyrics from a band named Dept. of Defense. I suspect that it is a criticism of defense department spending; but basically, I have no idea what the graphic is saying. As well, a peculiar squared area shows which is at odds with the circular flow implied by the basic idea. It appears that there are two minds working here; one has an interesting fluidity; the other this jarring block structure and dowdy font.(Me: not a designer.)