Should Data Visualizations Be Beautiful?

Data visualizations can be designed to look beautiful, if you possess the required visual design skills. The question is, “Should data visualizations be beautiful?” For years a battle has raged between infographic designers who emphasize the importance of aesthetics and data visualizers with a more practical bent who focus on the degree and quality of understanding that results. Those in the aesthetics camp argue that if an infographic is not eye-catching, no one will look at it, and that compromises in the quality of communication are justified as a means to capture the reader’s attention. Those in the optimal-understanding camp argue that the reader’s attention is wasted if the visualization does not clearly and accurately tell its story. In truth, most people have joined one camp of the other, not because of deep thinking on the topic, but because of preferences formed by their experience or lack of it. I’ve tried to occupy a middle ground, pointing out that visualizations can be both aesthetically pleasing and fully informative, without compromising either concern, but that this takes a high degree of visual design and communication skill. While the battle rages, however, fundamental questions are being ignored.

Should data visualizations be beautiful?

What qualifies as beautiful?

If you believe as I do that data visualizations, despite secondary variations in purpose, are always meant to inform, then their effectiveness is determined by the degree and quality of understanding that results. Therefore, a data visualization should only be beautiful when beauty can promote understanding in some way without undermining it in another. Is beauty sometimes useful? Certainly. Is beauty always useful? Certainly not.

What’s always required is that a visualization work for the human eyes, which means that it should not be displeasing to the eyes. A few basic principles of visual aesthetics can be followed—good color choices, legible fonts, proper placement and spacing, etc.—to achieve this result. Making a visualization beautiful is rarely required and it is usually not worth the effort unless your audience is huge and the information is really important. In addition, it can often work against the goal of informing. Making a data visualization beautiful in a way that compromises the integrity of the data always works against you. Even when the information is not compromised, however, beauty can work against you by drawing attention to the design of the visualization rather than the information that it seeks to communicate. Think back over your life and ask: “Were the people who influenced and taught me the most all physically beautiful? If they were wrapped in a different physical package, would that have affected their ability to influence me or my ability to listen to them? Did I ignore information that wasn’t delivered by stunningly attractive people?” Beauty is not the goal of visualization and it is usually not required to achieve the goal.

On those occasions when making a data visualization beautiful is truly useful, we must face the fact that beauty is indeed “in the eyes of the beholder.” What qualifies as beautiful for some is not beautiful to others, beyond the basic aesthetics that I referred to earlier that are rooted in visual perception. Most of what we deem beautiful is a product of culture and experience. If you love wine, as I do, you probably no longer prefer the wines that you found pleasing in the beginning. The fruit-bomb California Zinfandel’s that I loved in the past are no longer palatable to me. I now prefer wines that were crafted in the European tradition to produce greater subtlety and depth of character and to pair well with food.

To further illustrate this point, I’ve found that, when arguing the importance of beauty in data visualization, people often illustrate their position using works by infographic designers such as David McCandless. To my eyes, however, even when I ignore the fact that the information has been ravaged, I rarely find his work beautiful. Obviously, some people see his work differently than I do, but that’s the point that I’m making. Beauty is a fleeting target. What qualifies as beauty varies with the tastes of the audience.

If you’re a gifted graphic artist and communicator and have the skill that’s required to craft beautiful data visualizations when they’re needed, that’s wonderful, and I wish you well. Just don’t hinder the advance of data visualization by arguing that it must always be beautiful. Remember that the goal is to enlighten.

Take care,

35 Comments on “Should Data Visualizations Be Beautiful?”


By Greg. February 1st, 2012 at 5:18 pm

Just to add one point: communication is not only about effectiveness and clarity but also about trust. A plain (but effective) bar chart looks pretty cheap – mainly because it *is* pretty cheap to create them – and thus less people tend to trust plain bar charts. A rather beautiful or at least intended-to-be-beautiful chart is much harder to create. You need to hire an artist or at least someone who thinks of himself as an artist. And while all the efforts put into beauty might compromise the integrity of the chart, this effort itself is still visible to the target audience and thus might increase the trust level of the graphic. And that’s what you want if your goal is to communicate numbers through infographics and visualizations, may they be true or manipulated.

By Stephen Few. February 1st, 2012 at 5:51 pm

Greg,

It may be true that some people place more trust in charts that appear to exhibit a more sophisticated design effort, but the opposite is also true. Many people have learned to distrust designs of this type as potentially misleading. This is due, in part, to the fact that most of the charts of this type that they’ve seen over the years were used for marketing. I would hesitate to assume particular advantages related to trust based on the appearance of a graphic artist’s involvement.

By Steven Ng. February 1st, 2012 at 11:25 pm

Infotainment aside, I do think it is possible for charts to be both effective and beautiful. Beauty, as you mention, is subjective, but beauty does not necessarily mean ornate and complex. It is a delicate balance, but given two equally effective charts, I’d choose a tastefully beautiful one over an ugly one.

By Alan Kang. February 2nd, 2012 at 1:34 am

I usually don’t like chart-junks but thinks it’s too hasty to completely ignore aesthetic aspects:

First, beauty surely have subjective aspects but there is also some common ground which can be objectified in some degree. According to neuro-aesthetics, some aesthetic preferences have neural bases.

Second, in case of interaction/interface design, beauty affects on perceptual/cognitive effectiveness(so called Emotional Design). It’s pretty obvious that the same principle could be applied to IV.

By Andrew Marritt. February 2nd, 2012 at 3:54 am

Lots of well designed things look beautiful because they are designed to meet a functional need. You’ve discussed Dieter Rams’s work for Braun & Vitsoe (I have a Rams clock on my desk and his shelving lines my office – both act as inspiration). Rams uses colour occasionally to draw the eye to important functions.

Lotus’s 1958 Elite is a car that I count as one of the most beautiful of all time, yet it was designed by Kirwan-Taylor (an accountant) and Costin for aerodynamic effectiveness (it has a drag of .29 even without use of wind-tunnels). The Helvetica based signage of SBB (the Swiss trains) is quick to communicate a message and looks good. All are examples of beauty because of function, not beauty for beauty’s sake.

Effective data visualisation will often be beautiful. The use of techniques such as grid / golden ratio positioning, use of white space, care with typography, use of colour in a Rams-like manner can bring a certain beauty to the visualisation. I happen to think that this piece I’m working on at the moment is pleasing to the eye, even though still a work in progress: http://www.organizationview.com/example-dashboard

For me, a visualisation demonstrates a sense of pride in the work. It looks more professional, it demonstrates an attention and care for the details. As Donald Judd used to say about his work: “the simple expression of complex thought.”

Maybe we need to think of the principles set out in Adolf Loos’s famous ‘Ornament & Crime’ essay. “The evolution of culture marches with the elimination of ornament from useful objects”

By Jakob. February 2nd, 2012 at 4:11 am

Dear Stephen,

I was looking forward to you writing about aesthetic appeal and visualization, if only because I wrote about good design practices on visual.ly http://blog.visual.ly/what-makes-a-good-infographic/ and always considered your point of view a role model for a specific use of visualization. I hope you feel appropriately portraied in the article.

Appropriateness, btw, is a key element of design that I feel is missing from your argument when you speak about visualization in general. You are creating a false dichotomy in dividing designers into those who “emphasize the importance of aesthetics and data visualizers with a more practical bent who focus on the degree and quality of understanding that results.”

I must say that this statement feels somewhat parochial, too. Do you really mean to deny designers who care about aesthetics that they care just as much about the efficacy with which their work communicates? This being a false dichotomy I’d actually say that there is no middle ground for you to occupy.

The tradeoffs you speak of do exist, but they need not work the way you imagine them to work. You do have to compromise on one element of communication when you make a choice for or against noise/embellishment. I find decoration to be a more helpful property by which to classify visualizations than beauty, btw., because it is less ephemeral and can be tested against. Incidentally, decoration can be meaningful.

With meaningful decoration we enter the area where tradeoffs become a conscious choice of leveraging different perceptual and cognitive properties of the brain. Helping the brain frame information comes at the expense of cognitive load, but that can be a good thing, if you do indeed want to help the brain contextualize the visual input.

There is a whole body of research I feel you may have overlooked, which justifies aesthetic appeal – more importantly it justifies ornamental design choices. Just two examples I came across recently:

The efficacy of narratives:
http://onlinelibrary.wiley.com/doi/10.1111/j.1468-2958.2009.01367.x/full
The efficacy of aesthetic appeal through positive affect:
http://www.psych.ucsb.edu/~ashby/rev10635.pdf

But regardless of whether we agree with these studies the mere existence of the priming effect from psychology alone should tell us that we can obviously influence the uptake of information by introducing priming triggers into our designs. That may very well be why some chartjunk does work after all.

The problem lies with people who do not make conscious decisions about their designs. Those who motivate their choices surely should be allowed to make things beautiful, no?

By Naomi B. Robbins. February 2nd, 2012 at 5:18 am

Greg,

I need to believe people to trust them and I need to believe a graphic to trust it. How can I believe or trust something that has been manipulated?

By Hector. February 2nd, 2012 at 6:12 am

It would be wonderful to see some examples as to wether a graphic is both artistically beautiful and functional to its most, to clarify at least just a bit that frontier.

By Andy. February 2nd, 2012 at 6:53 am

Stephen,

I met you at TCC11. I’m the person that asked you to sign the front cover of your book. Surely you remember.

Anyway, like you, I despise most of the infographics produced. I have no particular beef with them, as long as they communicate well and most effectively to speed interpretation and insights.

David McCandless created a word cloud mess a few weeks ago which I critiqued. I thought you might find my take interesting.

http://vizwiz.blogspot.com/2012/01/information-is-beautiful-only-if-you.html

Andy

By Laura Davies. February 2nd, 2012 at 7:53 am

I agree that design beauty should be used as a means to enhance the quality of communication, and that it should not be used where it in any way detracts from point of the grpahic: to accurately and helpfully convey information.

However, I would like to make the case that data vizualisations should also be ‘not ugly’? As a market researcher I have encountered many representations of data that make me wince unwittingly due to their choice of colour palatte or generally lack of aesthetics. Certainly, for me, this detracts or at least distracts from the message.

I completely appreciate that this also rests on making the case that there is some universality to aesthetic principles – a whole other debate. :-)

By Laura Davies. February 2nd, 2012 at 8:00 am

I should also add that while I like to think that someone’s appearance would not affect my judgment as to whether I should listen to or be influenced by them, I might get a poorer impression if someone, say, was very shabbily dressed speaking at a business conference. It is a question of professionalism, which I think is similar to having a well-design chart. Also, unfortunately, the evidence suggests that the beauty of a person does indeed have an effect, with more attractive people earning more, even amongst university professors! http://www.businessinsider.com/beauty-pays-daniel-hamermesh-2011-10

By Jamie. February 2nd, 2012 at 8:48 am

Design, like many other things, is most often at its best when the audience reaction is to not initially notice the design.

When the reaction is to become immediately absorbed in the content the designer has done their job. When the reaction is to immediately focus on how beautiful it is, the content is already hindered.

I believe that the distinction between attractive and beautiful is important to your point here Stephen. You touch on it, but the distinction may well be lost on some readers.

By Andrew. February 2nd, 2012 at 9:25 am

@Greg “A plain (but effective) bar chart looks pretty cheap … and thus less people tend to trust plain bar charts.”

I find it hard to believe that people are untrustworthy of something because it appears plain, and I think you’ll find such an argument rather difficult to substantiate.

“And while all the efforts put into beauty might compromise the integrity of the chart, this effort itself is still visible to the target audience and thus might increase the trust level of the graphic.”

If the integrity of the chart is compromised, the target audience may (and often does) get the wrong message. That their trust level might be increased in such a situation is not a good thing.

By Stephen Few. February 2nd, 2012 at 10:40 am

Steven — You and I are of the same mind. I am making a distinction between visual beauty as a goal in and of itself, which is not useful in a visualization, and pleasing aesthetic qualities that contribute to a visualization’s ability to communicate, which is what I teach and attempt to inspire.

Alan — I’m not arguing that aesthetics should be ignored. What Donald Norman advocates in his book “Emotional Design” and I advocate in my books and courses is an aesthetic quality that makes something pleasing to look at and interact with. As you point out, our response to aesthetics of this variety is built into our senses and brain. This is what I was saying when I wrote, “What’s always required is that a visualization work for the human eyes, which means that it should not be displeasing to the eyes.”

Andrew — Amen!

Jakob — I have not taken the positions that you are arguing against. If you reread what I wrote in this blog piece, you’ll see that I have not ignored “appropriateness” as an essential aspect of design. The goal is to inform. Whatever is required to inform a particular audience of a particular message is appropriate and useful. The dichotomy that I made between “infographic designers who emphasize the importance of aesthetics and data visualizers with a more practical bent who focus on the degree and quality of understanding that results” is quite real.” I’m not arguing that all those whose work focuses in infographics fall into the former category—far from it—but many do, and the Web and print publications are filled with examples of their work. I have not argued that form and function, aesthetics and usability, cannot be married without compromise to either. In fact, I argue that they can and challenge infographic designers to strive for this high level of skill. What I am arguing against is that visual beauty in and of itself is not the goal of data visualization. It should only be part of the goal when it’s useful, when it contributes to the visualization ability to inform.

I have not overlooked the research that exists regarding the usefulness of aesthetics. I teach the importance of aesthetics in data visualization. No credible studies have demonstrated the usefulness of superfluous chartjunk. If an embellishment of any type contributes to a visualization’s ability to communicate information and help the viewer think about the information more effectively, in my opinion it is not chartjunk.

Hector — For an example of a visualization that is appropriately beautiful and fully functional, take a look at my blog post “Data Blooms in Beauty and Truth” (http://www.perceptualedge.com/blog/?p=1044).

Andy — Yes, I remember you. Thanks for joining the discussion. Regarding McCandless’ tag cloud, they are rarely useful, even in the hands of better skilled designers.

Laura — Absolutely. Data visualizations should not be ugly. They should be pleasing to the eye in a way that draws viewers into the data.

Jamie — Whether we use the term “beautiful” or “attractive”, we need to define our meanings, because the distinction is not obvious, nor will it be understood the same by everyone without definition. What I’m arguing is that, beyond a few basic principles of visual aesthetics, one’s goal when designer a data visualization should not be to make it beautiful (or attractive). The goal is to communicate. If additional focus on visual aesthetics is required for a data visualization to communicate, then it’s appropriate.

By David Gerbino. February 2nd, 2012 at 5:34 pm

Many years ago I was at an all division team meeting. A new senior executive was presenting business data via Power Point. He bragged how he learned all the Microsoft tools by Microsoft employees while he lived in Microsoft country. After his presentation, they’re was such a buzz. Everybody was talking about what they saw. Chart types and data tables like you never saw before. Embedded audio and video was used for dramatic effect as was all the cinematic effects Power Point offers. A lot of important information was presented. When I asked people specific questions about the data, not one person could recall more than a few facts. My take away from this was I would never over load a presentation with so much sensory noise, instead I would focus on what was germaine to the presentation and only use cinematic effects when it added to the data being presented.

Just my humble two cents.

@dmgerbino

By Fabio Alamini. February 3rd, 2012 at 6:33 am

Well,

I´m all in favor about the data beautiful. Sincerely, here in Brazil we need to draw the attention about our tools, and else, many people don´t give nothing about all the technology behind the chart, they only care about the beauty. The David´s post say about it.

I believe too, that the “beauty data” engages users to see the data, and after all, if the tools are all customizable, why not try and make something visually pleasing ?

Lastly, is very important maintain the visual pattern of the company, bacause the dashboard is a part of her. Would be somewhat inconsistent to develop a “blue dashboard” for HSBC, not?

@alaminif

By Jamie. February 3rd, 2012 at 11:21 am

Stephen – “Whether we use the term “beautiful” or “attractive”, we need to define our meanings”

Yes, good point

“What I’m arguing is that, beyond a few basic principles of visual aesthetics, one’s goal when designer a data visualization should not be to make it beautiful (or attractive).”

I am in complete agreement. By attractive I simply meant adhering to those basic principles of visual aesthetics.
My point was merely that it could be easy enough for some to misinterpret (or ignore parts of) your post to mean that the very ugly but undecorated charts that many of us are used to seeing are a-ok.
I very often see people making straw man arguments against the principles you promote by misinterpreting them in such a way.

By Stephen Few. February 4th, 2012 at 11:43 am

Jamie,

I appreciate your efforts to prevent misinterpretations of my argument. Vigilance is warranted.

By Ladi Omole. February 4th, 2012 at 6:21 pm

Beauty is surely in the eyes of beholder. To me, distracting decoration in data visualization is like adding a spoiler to a well designed beautiful car like Porsche. Should spoilers be added to cars primarily for styling purposes? Steve you are right – what we deem beautiful is product of culture and experience. I will vote for spoiled Porsche years ago without much thinking compare to now. So also executives that need to make a critical decision will not have time for any distracting decoration.

Ladi

By Janett the Admin. February 8th, 2012 at 12:43 pm

Stephen,

I’m forever trying to unlearn the idea that my charts have to be pretty in order to be functional and accepted. Thankfully, after attending E. Tufte’s seminar in Philadelphia last year, I’ve gained allies in my plight to cease with the pretty colors (which by the way – my department/company is known for) and go with data that means something. I really appreciate that your blog is here — and I always have my Tufte books nearby for inspiration to stay on the right course.

By the way, I thought of you last night while participating in my new hobby – pining (aka: Pinterest). I saw someone had a board on data visualization – with some nice and not-so-nice examples of data in art form. One of the pictures the user had collected was on a project a gentleman did regarding the closeness he feels to his Facebook friends via colored, wax busts of his head. Very interesting stuff! I just love that there are people out there who data-visualize for sport :)

By Richard. February 10th, 2012 at 9:14 am

Stephen,

I am relativley new when it comes to your work and I am enjoying your book “Show me the Numbers”.

My own thoughts are by presenting visualizations that are simple, concise and illustrate data in a meaningful way, is inherently beautiful.

the samples you have provided in your book and on this blog are simple and elegant, which to me is “beautiful”. Very subjective topic and I do not think we will ever have definitive answer.

I also think if you could take a chart off a white page onto a nice powerpoint background or select a complimentary colour scheme which does not distract from the message…great go for it! You may enhance the visual interest and thus garner more attention to your data’s message.

I think the lesson so form what I have seen in your work is that less is more, though I could be totally wrong..lol

All the best to you.

By Larry Keller. February 12th, 2012 at 7:19 am

Hello Stephen,

The last words I want to hear when a visualization is being described is “that’s cool”. That is not unlike my opinion of visualizations being described as beautiful. After recovering from that’s cool, ask the question “so what?” Has the author presented something with insight in the most simplistic way? Does the visual drive a question that had never before been asked? Oh, and about trust? Was the source a rogue Excel file or a trusted system of record. Visualize facts….LK

By DR. February 28th, 2012 at 10:02 am

I take issue with the premise of the question.

“Should food taste good”? Of course! First it should nourish, but then – absolutely yes – it should taste good.

“Should buildings be beautiful?” Of course! First they should withstand a rainstorm, but then – absolutely yes – they should be beautiful.

If your food is healthy, but unpalatable – you’re not done. Keep cooking. (Made a donut? You went too far.)

If your building stands up to hurricanes but is an eyesore – you’re not done. Keep constructing. (Built a glass house? You went too far.)

My point being: there’s an ideal here. Settling for “beautiful” but not “informing” demonstrates a lack of understanding. Settling for “informing” but not “beautiful” demonstrates a lack of vision.

By Frank. February 29th, 2012 at 1:21 pm

I think the answer to this question is easier reached if you ask the same about resumes. When an employer has a stack of 300 resumes, asthetics (paper stock, alignments, breaks, fonts, etc.) are one way of gaining their attention in the crowd. Once you have their attention, then the message needs communicated as efficiently as possible. To much asthetics will drown out the message. It’s a fine line that depends on the audience and how much it takes to draw them into raw data. I don’t need much to present to accountants, but I’ve had more than a few marketing people call a well designed graph “ugly” and go off chasing squirells.

By Stephen Few. February 29th, 2012 at 8:34 pm

DR,

Your argument exhibits the “straw man fallacy.” A straw man is a component of an argument and is an informal fallacy based on misrepresentation of an opponent’s position. To attack a straw man is to create the illusion of having refuted a proposition by replacing it with a superficially similar yet different proposition (the straw man), and refuting it, without ever having actually refuted the original position.

Read my argument again and you will see that I am saying that data visualizations should indeed be aesthetically pleasing (yes, the food should taste good), but that there is seldom a reason for them to be beautiful (you don’t need Thomas Keller to cook every meal).

It is easy, given basic skills, to make data visualizations aesthetically pleasing. It is quite difficult, and rarely necessary to make them beautiful. When we focus primarily on making a visualization optimally informative and secondarily on making it pleasing to the eye, without expending the time and effort to make it beautiful, we don’t lack vision, as you suggest, but are demonstrating intelligence and a realistic assessment of what matters.

Frank,

Your resume analogy doesn’t seem to fit most uses of data visualization. People don’t usually interact with data visualizations as they interact with resumes. When people use data visualizations to support decision making, they have an interest in the data and don’t need anything entertaining or decorative to get their attention. They seek meaning and clarity.

Your analogy to some extent does fit infogragraphics that are designed for public consumption, which appear in newspapers, magazines, or on websites, when the reader’s attention much be captured. When an employer is faced with a stack of 300 resumes and lacks time to read them all, she uses some quick filtering criteria to pare the stack down to a few. If she’s an intelligent employer, she isn’t going to look for the resumes that are beautifully typeset on pretty paper, however, but those that include the required qualifications and appear well organized, neat, to the point, and clearly written.

Even those in the marketing department who rely on data to do their work appreciate simple, clear, and accurate presentation. Those who produce marketing displays for the public, however, usually care most about getting their brand to stick in the reader’s mind, not about clear and accurate communication.

Appropriate aesthetically-pleasing design never drowns out the message—it enhances it by drawing the reader’s attention to the data. Only when meaningless embellishments are added to data visualizations in an attempt to make them beautiful do aesthetics distract from the message.

By DR. March 1st, 2012 at 12:28 pm

Stephen,

First, thanks for the condescending definition of a well-known term.

But more to the point, I guess I just feel differently. Why should Thomas Keller not cook every meal? Or put differently, why should I not try to cook like Thomas Keller every time I make dinner? Why lower my ambition?

The great *can* be done, so why discourage it?

I suppose you don’t view your argument as encouraging mediocrity, but that’s how it jumps of the page to me.

By Stephen Few. March 1st, 2012 at 7:47 pm

DR,

Being familiar with the straw man fallacy, you should avoid committing it. Once again, in your latest comment, you’ve created another straw man. I haven’t spoken of greatness. I certainly haven’t encouraged mediocrity. I combat it every day.

Data visualizations are meant to enlighten. When you create them, do everything that is necessary to make them fully informative. Just as people don’t need to be physically beautiful to communicate effectively, data visualizations rarely require beauty. When visual beauty is required to communicate your message, take all of the time and effort that’s necessary to exceed the norm of visually pleasing visualization to achieve beauty.

I work diligently to promote the best possible uses of data visualization among those who strive to support better decisions in the world. By suggesting that presenting data in visually stunning ways is always the goal, you undermine its use and effectiveness. Only a few visualizers of data are capable of making visualizations beautiful. That’s appropriate, because doing so is rarely necessary. If you are one of those rare talents, the last thing that I want to do is discourage you. Give us beauty, just don’t tell us that information isn’t useful unless it is visually stunning.

Given your position, both in this particular exchange and in previous exchanges when you’ve argued that data presentations should be sexy, perhaps you could show us an example of your work so we can judge whether it is indeed beautiful and/or sexy, and then determine if those qualities made it more effective.

By DR. March 2nd, 2012 at 10:17 am

My last statement was not an argument (strawman or otherwise) against your position, it was an expression of my opinion.

“…just don’t tell us that information isn’t useful unless it is visually stunning.”

You’ll never hear me argue against *effectiveness* by itself. It is the baseline requirement.

But I will forever suggest, that when done correctly
*effectiveness* + *beauty* = *greater effectiveness*

Given this belief, and assuming *greatest effectiveness* is the goal, then I need to argue for beauty.

By Stephen Few. March 2nd, 2012 at 10:59 am

DR,

You suggested in response to my argument that I was discouraging greatness, and then proceeded to promote greatness as the goal as if I opposed it. Calling your statement an opinion doesn’t change the fact that you used it as a straw man.

I think that your equation illustrates our disagreement: “effectiveness + beauty = greater effectiveness.“ In data visualization, beauty is not something that you add to effectiveness to achieve greater effectiveness; beauty is either a component of effectiveness (i.e., something you incorporate as a integral component of effectiveness) or it is separate from and a distraction from effectiveness. In other words, if you view beauty as something to add to effectiveness rather than something that is a component of effectiveness, it will not achieve greater effectiveness, but instead a tension between form (beauty) and function (effectiveness). To achieve greater effectiveness, beauty must be incorporated to serve effectiveness, not sought as a separate goal.

Also, we disagree in that you believe that beauty always makes a data visualization more effective. I believe that beauty sometimes is a useful component of effectiveness, but often not, and that people too often attempt to incorporate beauty in a way that undermines the effectiveness of a data visualization. And finally, in the real world where every component that is incorporated into a data visualization bears a cost, the cost of beauty is usually not justified.

I’m still interested in seeing an example of your work. It would also be useful for you to share your identify with us. Knowing the source of an opinion (or argument or evidence) is always useful in assessing its merits and motives. The anonymity of the Internet often undermines the quality of discourse.

By Stephen Few. March 2nd, 2012 at 2:12 pm

DR,

I have not allowed your latest comments to appear here because you have ignored my request to share your identify with us. You know who I am; polite discourse requires that you now share who you are as well. I and readers of this blog have a right to assess your comments in light of your role, expertise, and affiliations. Software vendors whose products I’ve critiqued sometimes post comments on my blog anonymously. This is something that I don’t allow. Share your identify and you may state your case, whatever it might be.

By DR. March 2nd, 2012 at 2:44 pm

Stephen,

I understand your request and respect your right to moderate the comments.

I am neither a vendor nor a data-vis-consultant, however given the field I’m in, sharing my work is simply something I prefer not to do. Making my work public is not in my (or my clients’) best interest.

I’m passionate about the data visualization field as it instructs everything I do, and should we cross paths in person (perhaps the eyeo festival?), I’ll readily introduce myself. However google knows all, and keeping my business unrelated to any other node is more valuable to me than is arriving at a satisfying conclusion to this conversation. I hope you understand.

Thanks for providing the platform for discussion.

By Stephen Few. March 2nd, 2012 at 3:15 pm

DR,

I’m afraid that our paths won’t cross at the Eyeo Festival. I don’t attend events that focus on data art yet call what they do data visualization. Events such as this contribute to the confusion that is currently undermining the field of data visualization. If Eyeo and other similar events differentiated what they did from data visualization–the use of visual representations of abstract data to explore, make sense of, and communicate information effectively–then I might attend and enjoy them as I would other art events. For the time being, you will only see me at Eyeo if they invite me to give a keynote to explain the differences between data visualization and data art and why these differences matter.

By Michael Cytrynowicz. April 23rd, 2012 at 12:08 pm

Stephen,

Let me start by saying I entirely agree with you. When the purpose of visualizing data is to facilitate the discovery of trends, relationships, patterns, surprises, etc., then the overriding factor is how to design the visualization in order to maximize (satisfice?) these.

That said, and not unexpectedly, data visualization can easily be hijacked for visual porn purposes. OK, Let’s call it visual titillation. That’s when the content matter itself starts being relatively unimportant, and the designer’s (and client’s) rhetorical intent is quite different from the first case. Well, sometimes, the content matter is used as a bait-and-switch device. We see this commonly in advertising: borrowed interest. The goal is to visually titillate – and to get the viewer to come for more, talk about it, tweet, blog, etc.

There is a third way, one that starts with John Dewey and goes through Kenneth Burke, which refers to that kind of beauty that mathematicians or physicists talk about (or for that matter, to, as Dewey exemplifies) a certain baseball movement executed in a particularly satisfying way by a certain player. This one is associated with having an experience. However, I would pose that (with Dewey) these experiences are all fused, and that it is quite hard (at least for me) to look at a visualization (however attractive it is) and not be disappointed when the data underneath is silly, insufficient, unsatisfying, irrelevant, inappropriate or even misleading. In Burke, decoration has a serious purpose. Empty visualizations, where the goal is pure titillation, are visual porn to me. When the visualization is teaching me something I didn’t know, or unexpected – and it is visually arresting – THEN we are talking about data visualization that NOT ONLY does what it is MEANT to do, but does it in an extraordinary, memorable way.

By Stephen Few. April 23rd, 2012 at 1:11 pm

Michael,

You have stated your case beautifully.

By Michael Cytrynowicz. April 26th, 2012 at 4:33 pm

Much obliged, Stephen!