Open Thread: What the **** is Visualization Anyways?

October 21, 2009  |  Discussion, Visualization

I think ever since visualization got started, people have been asking this question.

Some... okay, many describe it as purely an analytical tool. Others (i.e. me) are a little more liberal with their use of the term while the rest are somewhere in between. Some insist that the stuff we see on information aesthetics belong in an entirely different category and that that stuff isn't visualization at all.

As art, science, design, statistics, computer science, etc. start to melt together, the line between what is and isn't visualization grows more blurry.

What do you think? Is visualization only analytical? Can visualization be art? Are the infographics that frequent the front page of Digg visualization or are they just pretty pictures? Can visualization be just a pretty picture? Leave your thoughts in the comments below.

19 Comments

  • Okay, I’m in a hurry, but I’ll give it a try in one sentence ;)

    “(Information) Visualization is an umbrella term for any translation of a set of abstract information into a visual representation with the objective of increasing insight into this information.”

    IMHO it doesn’t really matter for a definition whether the data domain is purely scientific or rather banal (i.e. Gapminder vs. We Feel Fine), or whether the output is aimed at professionals of a certain field or at a general audience. Instead, the application of a “visual design toolbox” as well as the overall goal of “enlightenment” should be the primary criteria to decide whether a product or service fits into this category or not.

    Apart from that I think there’s still a long way to go simply to sort out the naming confusions in this field: http://tr.im/CvHw

  • I agree with Christian on the definition of Visualization and I would include Stephen Few’s differentiation between Scientific- and Information Visualization as two subsets of Data Visualization.

    I believe a lot of confusion arise around the usage of abbriviations (like datavis or infographics) and the lack of a general agreement about how to use the terms correctly.

    For me personally a picture generated by using data is indeed a visualization—the usage of such a visualization may determine it’s classification.

    • Definitions of visualization tend to be given by its practionners, who fall into several categories: academics & computer scientists, developers & hackers, information designers & artits, and users. Each group’s point of view may not take into account that of the others.
      so I’m in favor of a loose definition, where visualization is the field
      where all these groups interact.

      that being said there are quite a few people who fall into all categories.

      • Also, the term “infographic” seems to be used for things (we) information designers and data visualizers would never refer to as an inforamtion graphic.

        Just a few minutes ago, I watched a clip on youtube claiming to be an infographic. For me, the lack of complexity in the visually communicated information makes this clip more an animated illustation telling a story (not that infographics could/should do that too).

        This also happens in print: I often wonder if the visualized data is quantitatively enough to call the end product an infographic.

        In most of these cases, the often raised question pops up again:
        where are the boundaries between visualization and “data art”?

        BTW, this is the clip I was relating to (try watching it without sound): http://www.youtube.com/watch?v=WgwboxatZPw

  • A useful distinction can be made by considering the role of viewers: how are they expected to engage with the image? Some visualisations require a ‘driver’ – not just a viewer. At the other end of the spectrum, some infographics are designed to be merely ‘consumed’ by viewers – there is nothing for viewers to do and they are not expected to bring anything to the image themselves.

    For the pioneers of scientific visualisation in the 1980s what defined ‘visualisation’ and distinguished it from mere ‘pretty pictures’ was that visualisations are used to *do* science – they are not merely a summary of conclusions or the outcome of science. This generally meant that visualisation is necessarily interactive. It is very different from the way images in science were previously understood.

    If anyone is interested, there’s a fair bit on this topic in my PhD thesis (chapter 3): http://bit.ly/3ej2d3

  • I come from the non-purest, pretty-broad-church school. I’m not a data visualiser, or an information designer so I can only offer a layman’s crack at a definition …

    a representation that combines two or more things (concepts/data points/objects/other) in such a way as to help the viewer visualise their relationship or meaning.

    Adam’s driver/consumer points are interesting. If there is a scale with visualisation at one end and pretty pictures at the other, I’d draw the line at the point where relationship or meaning are secondary to the form.

    So (trying to get back to your questions) as soon as something is just a pretty picture, then it’s just a picture that’s pretty.

  • I’ve been puzzling over this same issue. I’ve been intrigued lately by research that performs the function of a visualization through displaying data patterns in auditory ways. What do you call a visualization when it isn’t visual? Maybe we need a better word for this.

  • I would simply say that visualization is “the transformation of information into a visual form”

    The information could be a massive set of data it could simply be the relationship between two entities.

    This is similar to Christian’s view however, I think sometimes people do this as an art form (art being something that is purely created for aesthetic purposes) and only informative visualizations are done in the pursuit of insight.

    I think this blog really specializes on informative and quantitative visualizations. Using numerical data and relationships to create visual representations that invoke insight or highlight a relationship.

    Using the term visualization in this broad way (an umbrella term as Christian puts it) is totally fine in my honest opinion.

    Cheers
    Matt

  • Information Visualization – allows a user to have a conversation with the data. This is my personal take on the term.

    • Exactly… learning by doing… kinsthetics. It’s not just about being the web 1.0 multi-million dollar ad campaing, it’s really communicating with the data, getting close to it and understanding the intricacies of the common, sometimes seemingly uniform, elements.

  • I agree with others above that the terms ‘visualization,’ etc. are all quite broad and apply to more artistic works as well as informative ones. What they have in common is, as Matt puts it, “the transformation of information into a visual form.”

    Interestingly, I didn’t see any terms in Christian’s graphic (http://tr.im/CvHw) that separated the artistic and informative uses of visualization. Like the other terms, there would be a lot of grey area, but there’s enough distinction to warrant some new ones.

    • @Daniel, the graphic isn’t mine. It’s been made by Dan Saffer some time ago when he tried to shed some light on the confusing terminology of Interaction Design and its surrounding disciplines. Information Desgin/Visualization plays only the role of one sub-discipline among many, but in my opinion it goes into the right direction. I think if you would sit down and draw a similar map only for the terminology used for information design, data visualization and everything in between and around, you would end up with a similar mess. :)

  • I think the definition of sonification give us a great start (http://en.wikipedia.org/wiki/Sonification): “Sonification is the use of non-speech audio to convey information or perceptualize data”. Visualization could be defined as the use of non-text image (mainly) to convey information or perceptualize data.

  • Tim Buckingham October 21, 2009 at 8:15 pm

    This from Wikipedia. If it is on wikipedia then it must be true :)

    Visualization is any technique for creating images, diagrams, or animations to communicate a message. Visualization through visual imagery has been an effective way to communicate both abstract and concrete ideas since the dawn of man. Examples from history include cave paintings, Egyptian hieroglyphs, Greek geometry, and Leonardo da Vinci’s revolutionary methods of technical drawing for engineering and scientific purposes.

    Visualization today has ever-expanding applications in science, education, engineering (e.g. product visualization), interactive multimedia, medicine, etc. Typical of a visualization application is the field of computer graphics. The invention of computer graphics may be the most important development in visualization since the invention of central perspective in the Renaissance period. The development of animation also helped advance visualization.

  • visualisation is everything! it’s way too broad a term, I think, to even be mentioned on flowing data, because you post within fields of visualisation.

    there’s scientific visualisation, information visualisation, information aesthetic visualisation, visualisation art, information graphics…and they’re all different and special =)

    so: no, yes, yes, yes.

  • A key component of visualization I haven’t seen mentioned on here is: analysis.

    ‘Visualization’ (in its many disparate but commonly bound forms) is not only a way to share insight and context… but to develop deep thoughts, and really work through hurdles and problems: a form of non-physical kinesthetic learning.

    Interactive visualizations play this role for the end user, but the act of creating a visualization plays this role solely for the creator.

  • Exactly!. Ever since I encountered that musical “visualization” of research data I started to think about how senses interact with the basic concept of manipulating data into a form that our human senses can understand and synthesize better than that numbers. In theory, this can be done for any of the senses, it is just that we usually use vision. Thinking you could have data displayed in smooth or rough textures of different levels/heights or intensity, or sour/sharp/sweet odors. You could even use levels and types of pain, although I would hope not.

    The point is that visualization is a tool for helping us understand and communicate data in ways that are not intuitive for most people. To convert the counterintuitive to intuitive. ;) Some people are visual learners and others have a preferred learning modality of hearing or touch, so the idea of converting data displays to accommodate the preferred learning modality of the user would seem to indicate that “data visualization” might be more accurately described without the embedded word ‘visual’. Perhaps data art?

    There is a fascinating image in Gordon Dickson’s book, The Final Encyclopedia, in which the encyclopedia is an interactive databank that displays the shifting information structures in a 3D holographic sensory barrage or light, movement, sound (texture?) that is so complex most people either can’t perceive it or it drives them nuts. The librarian was the person who could understand the “visualization.”

    There are people who combine data displays across a variety of senses. I have in mind Iannis Xenakis who converted mathematics, computer programs, and architectural designs into orchestral performance pieces. I once conceived a piece of music where the score was derived from a line drawing. Only the conductor could see the lines combine into the visual display, and the performers would see only the individual lines. Remember the message sticks from Australia or the kente/adinkra cloth from Ghana?

    “Is visualization only analytical? Can visualization be art? Are the infographics that frequent the front page of Digg visualization or are they just pretty pictures? Can visualization be just a pretty picture?” It seems to me that the answers are implied in the framing of the questions. Data visualization (or data art?) by definition is a combination of analysis and communication, both facilitating discovery on the part of either the creator or the audience. A data art creation (whether visual or auditory) can lean more towards analysis or more toward communication, but must contain elements of both.

  • I’m going to sit on the fence here. Visualisation is a communication tool, it tells us information in ways that provide insight and connection.

    BUT

    Visualisation that does not reflect analysis (as per the visualisation manifesto http://bit.ly/1GTu8n) can’t deliver insight. It’s like a picture book story without the story. Pretty, but without a story to keep me turning the pages.

  • David Palmquist November 2, 2009 at 11:55 am

    In sales & marketing analytics affective data visualization is ultimately about driving decisions. Each ‘view’ needs to contain enough context and delineation between different courses of action, as well as a clear path to take the hoped-for action.

Copyright © 2007-2014 FlowingData. All rights reserved. Hosted by Linode.