5 Reasons Visualization Is Not More Prevalent
April 20th, 2008Why does it seem I have to look hard to find good data visualization examples? Why do few tech companies devote resources to visualization (Google’s the obvious exception)? Why are there relatively few job postings for visualization, with many of those there are requiring mainly graphic design skills and not data visualization skills? I was thinking about this today and I came up with a few possible reasons, some based on perceptions, and others based on marketplace realities.
Reason #1: People Don’t Know What Data Visualization Is
People don’t know what data visualization is. Don’t believe me? Read the Amazon.com reviews for the book Data Visualization by Ben Fry. They contain negative comments such as “One would expect a book with the title ‘Visualizing Data’ to be crammed with pictures”. The issue seems be that too much of the book is devoted to data and the mapping of data properties to visual properties.
Graphic design is different from data visualization. Graphic designers are largely free from having to deal with actual data, and from having their product emerge from data. Graphic design components and data visualization components are often mixed, and with great success. But they are different. Art is not visualization. And visualization is not art…unless it is 
The above visualization (which is, in fact, by Ben Fry) is driven by the properties of two underlying datasets. One dataset is the DNA of a monkey. The genes (the data) are represented as very tiny white text. A second dataset used is human DNA. It is only depicted after the difference of the two datasets has been computed. Then the genes that are different between the monkey and human are represented in red. Fry obviously didn’t choose which areas of the visualization would be red, the data did. What about the monkey pic? Even that is a visual representation of a property of the dataset…the type of the DNA dataset shown in white text.
Reason #2: Crappy Existing Visualizations have Polluted Perception
The visualization on the left is the interface for the search engine Kartoo. The visualization on the right is a feature CNET used to have called The Big Picture. Both attempt to visualize data usually shown as lists (search results, related news articles) as 2D networks. Its a nice idea, as pairwise relationship properties can be visually represented as edges. But these particular efforts both miss the boat. They don’t actually increase the amount of information represented by very much vs lists, while greatly increasing the mental load placed on the user trying to extract the basic information.
Reason #3: People are Unable to Mentally Separate the View from the Data
Here’s another Ben Fry work (I was watching a video/talk of his earlier today, which is part of the reason he is so prevalent in this post). It shows six different visualizations of the same dataset.
Many times data relates to physical objects. In such cases people may have trouble dealing with such data as visually represented in any other manner than that which includes those physical objects. Or another situation is one in which data has just always been depicted in a certain way, which interferes with any new depiction.
Reason #4: Visualization is Difficult to Create and Easy to Copy
This is somewhat irrelevant, but I have had a Yahoo mail account for about a decade. There was a good six year stretch where it never changed. If Gmail hadn’t come along, who knows.
When Google released Google Finance, it marked a number of firsts…the use of AJAX for stock charts (the chart itself is actually Flash), the overlay of events on the chart, and the dual time sliders. No doubt Google spent much time and effort designing this visualization tool. How long did it take Yahoo Finance to copy Google Finance’s chart once Google revealed it? Not long. Good visualization design is hard. It’s even harder when its object is to deconstruct very complex data. Reverse engineering a visualization is easy.
Reason #5: People Won’t Pay for Visualization?
I’m not so sure about this one, but our company’s CTO recently commented to me that he couldn’t think of any successful standalone visualization effort other than Processing.
Applications such as Google Maps don’t count both because its free, and, more importantly, because people wouldn’t have access to the underlying data without the visualization. I can think of a few commercial successful standalone visualizations such as this one, but surely the list is fairly short.
April 20th, 2008 at 5:55 am
Very interesting! I particularly like the Google Finance vs Yahoo Finance example, which raises an interesting point: What is the incentive to invest time/effort in a new visualization, if someone else can copy it rather easily? I assume a lot of experimentation / user studies go into coming up with new UI/visualization techniques..
Maybe company-internal tools are the right environment for new visualization innovations, as the result is somewhat protected.
Keep up the good work, nice blog! Discovered it recently, great stuff.
April 20th, 2008 at 6:18 am
Interesting text despite the initial erroneous assumptions. I think that there is not a lack of good visualization examples. In fact I think that never so many good visualization examples were available to the general audience.
Second, Google is not alone in their efforts to promote visualization. All the top web companies (MS, Y!, IBM, Sun) are actively betting on “visualization”. You just have to look for it.
April 20th, 2008 at 9:33 am
I agree with Miguel – the number of good visualisations out there is quite plentiful.
however, I think the key is why ‘Visualisation Is Not More Prevalent in the Minds of People’.
and graphic design *should* be a part of visualisation – of a good visualisation, at least! look at Jonathan Harris, for starters, who does amazing work. and his are examples where visualisation is art!
your point about crappy visualisations is quite true, and the 4th reason is a very good observation. like Maeda says, being complex is easy. it’s being simple that’s difficult.
April 20th, 2008 at 10:59 am
Very interesting question. I kind of agree with reasons #2 to #5. But #1 seems in conflict with the saying “seeing is understanding”. So must people should know what visualization is. Maybe visualization means something different here.
This text remind me a script of P. Dirac I read few weeks ago (here http://www.atomicprecision.com/Other/Paul%20Dirac%20Talk%20-%20Projective%20Geometry%20(2).pdf). In the script Dirac compared gemoetric and algebraic thinking from phillosophic and practicle perspectives. It has suprised me that he actually perfered geometrical thinking, but just for technical reason most of his puplications seemed more algebraical.
I think most researchers and scientists are still favoring analytical or algebraic presentations; and this might be another reason for the lack of interest for visual data presentation.
April 20th, 2008 at 12:49 pm
Todd,
Nicely put. I especially agree with your first two reasons for data visualization’s slow progress. Much of my time is spent trying to alleviate people’s misunderstandings about data visualization and the harm done by many visualizations that simply don’t work. Data visualization has recently been growing in popularity, but much of what people see is the silly stuff–more more graphical glitz than true and meaningful data display.
Take care,
Steve
April 20th, 2008 at 2:15 pm
Great article. I think perhaps one answer to your question lies in your CTO’s comments – he’s looking for successful standalone efforts, while we think that visual representations of information rarely stand alone. There are always concerns of the sort you call our in your comments on Ben Fry: you have to be intimately familiar with the data, and willing to look beyond stock solutions like treemaps or force directed “stick-n-rock” graphs. Processing itself is a programming language for artists rather than an effort specific to visualization.
Based on our own work, I’m seeing infovis creep into the wider world in two ways: standalone set pieces like our work for Digg Labs or the NYTimes’ occasional interactive piece, and augmentations to existing interfaces in the form of reactive charts, indicators, and sparklines kneaded into the flow of more traditional information.
April 24th, 2008 at 1:41 am
Great comments…thanks for contributing!!
Cheers,
Todd
May 23rd, 2008 at 6:58 am
I agree to your arguments and with Steve’s refinement. (I’m reading your books now Steve and find them very well-written.)
but I think of a different reason for which data vizualisation isn’t more popular. It has to do with attention.
data visualization presentations which work are effective because they are instantly recognized, and that is because they have been used abundantly by mainstream media. a map, a pie-chart, a time series: those are governed by centenial laws and don’t need to come with a handbook.
now more novel forms of visualization need more explanation. this also means they will require a stronger commitment from a potential viewer to get the message.
for that reason, why would mainstream media use advanced visualization techniques rather than spend their money on beautifying (read: chartjunking) existing ones?
unfortunately there are quite a few cases where a standard chart won’t work and where a more sophisticated approach would be not only more impressive visually but also clearer, easier to use and more effective. but who outside of the NYTimes understands that?
September 22nd, 2008 at 4:30 pm
Have you seen FYI Visual? The original program was created in 1989 and was way ahead of its time.
Which once was GIFIC (Graphical Interface For Information Cognition) is now FYI Visual…
http://www.youtube.com/watch?v=l5th0i6UcQg
November 12th, 2008 at 6:06 am
What do you think of this: http://www.cloudtuner.com
November 27th, 2008 at 1:21 pm
From my experience, info graphics are not taught often in journalism schools.
February 9th, 2009 at 5:31 pm
Solid analysis Todd. In my experience, data visualization gets it’s exposure more and more these days. A factor of tremendous importance is the relation between data, visualization technic and audience. Too often over-complex visualization confuses the unexperienced user although the visualization itself is well suited for the represented data. It’s my belief that in the relationship between viewer and visualization lies great potential to help data visualization gain expsure and get understood.
October 4th, 2009 at 10:50 pm
[...] Todd provides 5 reasons why data visualization isn’t more prevalent: [...]
January 7th, 2011 at 12:30 am
[...] Todd provides 5 reasons why data visualization isn't more prevalent: [...]
May 12th, 2012 at 6:45 am
Was just thinking about this subject, and found your blog. I enjoyed your post – enough that I wanted more.
I think the Google example is important – specifically because the data was relevant – and the visualization helpful for making investment decisions – not just because it was plagiarized. When people understand the relevance of the data to a decision, then the visualization is perceived as “useful,” otherwise, the potential to to make tools to visualize something is simply “interesting.” Most money making entities (which tend to drive innovation), spend their money on useful things, and often don’t want to pay, or wait, for all the “interesting” things to develop – unless they are an enterprise that makes tools – and wants to spend money on R&D. It makes sense, because for every useful revelation (whether its data based or some other observation) there’s a large proportion of irrelevant ones – probably more than the Pareto principle suggests.
So, right now, the open source allows individuals to do things for themselves. Maybe your just not seeing all the visualizations. If I created an investment visualization tool that actually revealed a profitable buying strategy, that’s the last thing I would share freely with the general public, and you would never see it. Maybe that’s why processing.org is the public refrigerator, where all the kids art gets posted – and why Fathom.info is more interesting work often backed by sponsors, and not every example is fully operational.
An aside, I find it irritating when someone says that “Processing is for artists.” Processing is just a sketching tool open to anyone – non-technical doesn’t mean “artist” nor does “artist” mean non-technical – though it does give non-technical people an accessible and rewarding entry point for learning more about Java, and the principles of writing – anyone who is interested in collecting, or understanding data might see the value in learning to create tools for doing so – not just artists.