From Art to Apps: Data Visualisation finds a purpose
27th August 09
Posted in creativity, data, design, guest
Author: Jim Carroll, Chairman, BBH London
I recently attended an excellent Made by Many event hosted at BBH which featured a re-presentation by Manuel Lima of his 2009 TED talk on data visualisation. Manuel is the curator of visualcomplexity.com and is an eloquent, modest, charming pioneer in this fascinating field.
As a novice myself, I could not help wondering why we are all so immediately and instinctively attracted to the best of data visualisation.To start with, I’m sure there is some fundamental truth that for most of us data become meaningful only when we can see scale, change, patterns and relationships. Seeing is understanding.
It’s also very reassuring to discover that complex, seemingly chaotic data sets and networks can be expressed as elegant, colourful, ordered maps and models. Perhaps there’s something akin to what the Enlightenment scientists felt as every new discovery revealed the endless beauty of nature.
Indeed the best examples of data visualisation have their own aesthetic beauty. (I felt a nostalgic pang as I recalled time spent with spirograph in my bedroom as a child.)
Like spirograph, but better: Email map by Christopher Baker
To some extent however this elegance, which makes data visualisation so immediately compelling, also represents a challenge. It’s possible that the translation of data, networks and relationships into visual beauty becomes an end in itself and the field becomes a category of fine art.
No harm in that perhaps.
But as a strategist one wants not just to see data, but to hear its story. And it can seem that for some visualisations the aesthetic overpowers the story. I spent many hours when younger staring at data tables, yearning for them to reveal a narrative. It is the prospect of bringing articulacy to hitherto cold, laconic facts that should be at the heart of the excitement around data visualisation.
The more compelling projects from Manuel’s archive did indeed seem to reveal some insightful truth about the relationships that they considered. Enron’s email patterns, the map of Segolene Royal’s supporters, the plotting of visitor eye traces in Barcelona, all looked extremely useful.
Enron Communication Graph, by Kitware Inc.
With this last instance in particular, one can start to imagine how understanding the dynamic patterns of tourist traffic around the city and its most photographed areas might enable the development of all kinds of helpful tools and services for both tourist and city.
Tracing the Visitor's Eye by Fabien Girardin
Manuel himself talked about ‘turning tools of curiosity into tools of functionality’. In this respect he quoted Chaomei Chen: ‘A taxonomy of information visualization is needed so that designers can select appropriate techniques to meet given requirements.’ And clearly this desire to enable greater utility is driving Manuel’s own research into the different methods and models of visual representation.
As a pioneer in his field, Manuel discussed the opportunities emerging in interactive data maps and he described a Californian experiment in which it should be possible physically to interact with a huge data set distributed about a six storey building. Blimey. I think I’ll leave that to the true data connoisseurs …
Finally, as a grey haired strategist, I found myself considering how the paucity of visual representation techniques had impacted the way we tackled problems in the past. I think we knew fundamentally that most events were precipitated by complex systemic pressures and relationships. But our limited power to disentangle the many elements in one system reduced us to characterising most strategic problems in rather monochrome ways.
So, this is progress indeed. Data visualisation has radically improved our understanding of these complexities. The real question is: what will we do with that understanding?
20 comments on “From Art to Apps: Data Visualisation finds a purpose”
or sign in using Facebook Connect
Enter your personal information to the left, or sign in with your Facebook account by clicking on the button below:
Patricia McDonald
ello!
guest post from a MASTER> good stuff.
Jim - if you’re reading - this is my proleptic response
http://farisyakob.typepad.com/blog/2008/06/ways-of-seeing.html
more data requires new ways to see/
Nice article!
Plenty more business visulaizatons found here:
http://orgnet.com/cases.html
http://www.thenetworkthinker.com
Wow. That is a wonderfully articulated rumination on Manuel’s speech. (It was our great pleasure to arrange the talk, and have people like yourself there Jim).
Indeed I agree that the aesthetics of many visualizations can have such an overpowering initial impact that the true meaning and insights of the data can be obscured. But invariably I find that like a bee, once I have been attracted and curiosity aroused, I then get stuck into the real business, in this case, of searching out the (finally not-so) hidden gems.
Thanks for such a thoughtful post.
Thanks Jim, for your article. Of late, I have become consumed by this field and spend hours trying to lay my hands on anything of interest. Increasingly, as I look visualizations, I find myself asking, “So What?” Did this manner of representing the data enable someone to do something differently, to take an action they wouldn’t have otherwise, to uncover a hidden pattern that might change their view on the topic at hand? Sadly, for many of the examples touted as the ‘best of the best’, I can’t even begin to answer the “So what?” question.
If we want this field to be sustainable, we need to consider that question and try to answer it. If we don’t then let’s just agree, it’s art and appreciate it for that.
Great questions, Peggy. The purpose of a visualization of data is to reveal the complexity of the dataset so we can better answer important questions about it. It is beautiful as an artifact partly because we love elegance.
I can imagine uses for these, and they are very cool, but these are far from the best of the best of data visualization examples I agree. Ed Tufte’s books show the beauty of useful quantitative visualization albeit low tech. There are a few high tech ones that are really good but most don’t measure up because the focus is on the tech and the interface and usually not the evidence patterns.
I hope we’re moving toward a high-tech version of real decision-making data visualization that show the artistry in revealing patterns of evidence and not just cool graphics that make art out of data. There’s a lot of potential for this field.
Really enjoyed your take on this talk. I was there as well and was bowled over by the aesthetic beauty, and frustrated by the tip-of-the-tongue use and function that is still so embryonic. As brand researcher and strategist i’m not sure where we stand with data viz at this point, your right, the narrative is not yet clear in most examples shown. But, having studied subject/object networks that connect cities within the academic sphere of geography, i found the examples of mapping of global communities fascinating. Community and distance is obviously changing rapidly and is a long told story. However some of these visualisations i thought were fantastic ‘maps’ of the world we live in, which as yet haven’t been to great. Thus there are two points i took from the talk that i think are really important. 1) There is an aesthetic value of maps that creates perspective and stimulates a wonder which goes beyond a purely ‘art’ aesthetic. We still use hard geographies of metric distance all the time and i think these visualisations help us create relevant geographies of the 21st century.
2) And i guess, out of this first point will come a greater understanding and feel of the ‘communities’ through which which brands live and breath
Anyway fascinating stuff
Jim,
What a crisp, insightful piece. And one that hits particularly close to home for me — I recently wrote an article about data visualization for BusinessWeek which, to my surprise and sorrow, got a pretty negative response from readers. For the most part, people felt like the 20 works I had chosen to display were too “artsy.”
So the point you make about data viz maybe converging to fine art made me think: Consider the great paintings of the Renaissance. Fine art, yes, but they portrayed the social dynamics and captured the cultural patterns of that era — they were social observation that sparked public discourse.
And that’s precisely what data viz is, or has the potential to be, today.
As a strategist myself, I absolutely and wholeheartedly agree that data viz can aid a richer understanding of patterns and relationships and other mesh networks of factors that underlie strategic challenges.
The key, I think — and where all those people who yell “artsy!” get tripped up — is approaching it not with the *expectation* of a solution, but with the openness to one. It’s only then that the creative, “artsy” aspect of data viz can become a spark for your own fresh thinking, an eye-opener, not a distraction.
Great post Jim. I was at the Manuel Lima talk, and totally with you guys on tools of curiosity > tools of functionality. But for me, the point is that curiosity in this instance serves to drive functionality: it is a reciprocal relationship and that is what is important.
My thoughts here http://ow.ly/loaP
Cheers
Great comments and a fascinating topic. What strikes me reading the comments is that, building on Maria’s point, great art is that which makes us look at things differently and reconsider our perspective. Great art, from the beginning of time, has told us something about the world we live in. Conversely, great analysis has an elegance, form and structure to it. So perhaps the distinction is not art versus science but simply great versus not….
Top post Jim,
I feel what you are saying about the stories contained in data. I was discussing with Manuel that any data visualisation project should start with “What questions do I want to ask of my data” before decisions on integrations and aesthetics.
Only by asking the right questions first will we truely discover real insight and a story beautifully told.
Thanks to everyone for your kind words. I’m not an expert in this field so I’ve been busy following up on the various suggested articles. Fascinating and inspiring stuff.
Many thanks Jim, for deconstucting vague and obscure art forms(?). Had to share this teeny bit of info that may seem interesting. A friend of mine is into data viz of a similar -yet different nature. He turns Indian classical music, which by the way has no written notes(!), into art.
While the significance of turning sound waves into a painting were slightly lost on me, i couldn’t help marvel at the form they had taken. Immensely powerful. Take a look at http://www. orka-m.com…cheers
[...] [Artikel lesen] [...]
[...] From Art to Apps: Data Visualisation finds a purpose Kool Moves Simple flash animation and effects software. CrazyTalk 6 Neat lipsynce face animation [...]
[...] Kartik Mani: Many thanks Jim, for deconstucting vague and obscure art forms(?). Had to share this te… [...]
Pats,
Couldn’t agree more.
Thanks Jim, Justin, Manuel and everyone else who put on this thought-provoking talk.
I particularly liked how Manuel considered the convergence of different trends driving visualization. These ranged from the obvious, like social media and open data, to the less obvious (to me at least!) like cheaper storage costs and complex scandals like Enron.
With regard to the discussion about “useless” visualizations, I think Say’s Law will apply here. It will be interesting to see what uses evolve from visualizations that, for now, seem more aesthetic than functional.
[...] BBH Labs via Twitter [...]
Relevance is they key for useful vis. of any kind.
Until my developer DIED we were coding a geographically based data-agnostic decision tool as a plugin for NASA’s WorldWind 3d globe viewer.
This would allow layering of almost anything including input from SMS to enable comparison and contrast also linkages between global phenomena.
One great example illustrates immunization rates from UNMDG .csv over the past 20 years. It highlights where best improvement and declines occurred then we would layer funding networks and colorize conflict zones.
It would be just as easy to chart poor growing conditions against weapons sales to identify key regions for some ‘preventative maintenance’.
Many of Tufte’s examples are geographically based and it occurs to me that abstract data realms can be interesting but bringing this study back to real life has a much finer purpose.
I even think Enron’s email would yield some understanding if the physical locations were incorporated with localized information. How many of Ken Lay’s more aggregious decisions were made from Crawford Tx? Or some “undisclosed” bunker location under the US Naval Observatory?
I am eager to learn more and be involved in the practical application of data vis. Thank you for starting the conversation.
[...] wrap up this intro with a post by Jim Carroll, Chairman, of BBH London that was featured on the BBH Labs site, which also has a great follow-up [...]