A map is worth a thousand pictures…when animated

It is no secret that the newspaper industry is in serious financial trouble. Many argue that the age of newspapers as we know it is over. The only interesting part of the debate seems to be what “newspapers” could become if they fully embraced digital reality.

One of the early indicators of that future has been the coverage of the 2008 Presidential elections by newspapers’ online extensions. The New York Times, for example, has been consistently exploring various novel ways of bringing visual clarity and richness online to what would otherwise be dry data in newsprint:

string.jpg

analysis.jpg

namingnames.jpg

The online versions of newspapers have moved far from the dead-trees-archived-in-microfiche mindset by attempting to become digital centers of research for ordinary readers. As the “newspaper of record” that’s a good thing for the New York Times. The web offers infinite space and flexible time, two items conspicuously absent from print designers’ toolbox. But as gifted as the Times graphic designers are, the editorial approach to information visualization seems still mired in a static, 2D print world.

A recent election results map depicts “difference in support for major parties between 2008 and past elections”:

map1.jpg

By clicking on the map, readers can drill down from national to state to county level results in a few clicks:

map2.jpg

While this is infinitely better than what would otherwise be endless pages of mind-numbing statistics in newsprint, it’s still quite linear and vertical. What’s missing is a timeline, a “narrative” on the horizontal axis of time. Yes, there’s a handy slider to switch the map to four-year intervals to see snapshots of electoral change, but that requires readers to stitch together the narrative by themselves, which they likely won’t do to full effect.

The analytics problem here is how to visually present the county-level swing between the Republican and Democratic axis across four election cycles, comparing 1992, 1996, 2000 and 2004 each to 2008. While the election-over-election differential is interesting, the real story is the transformation:

transformation-anim.gif

[I simply extracted and sequenced Times' four maps with a Year/President designation at the bottom. Red: Rep; blue: Dem.] There’s much to improve here, but a single, animated map frames the transformation in an unmistakable manner with an unforgettable conclusion. Research becomes insight…all without the need for any clicks, slides or drill-downs by the reader.

Maps are not enough

On the web, there are many venues to get the facts. What’s scarce is the narrative, the perspective, the insight that ties it all together in a time-effective way. In analytical terms, what’s missing is trend spotting, rendered with clarity that only digital technologies can make possible online. Every day, an unimaginable number of colorful reports and charts are printed to gain “business intelligence” in corporate America. What follows that is an unimaginable number of hours spent daily by armies of analysts to make sense of these reports. Insight is gained only, and very rarely, when the dots are connected in disjointed graphs, reports and maps to detect a trend along the time axis so an action can be taken in the future.

The currency of print-based newspapers has been the snapshot: what happened and when, in the past. To survive and thrive online, “newspapers” need to become trend spotters, trend explainers and thus storytellers of transformation which, when read or seen in small and disparate chunks as newspapers currently vend them, is often impossible to detect or interpret by ordinary readers. The currency of narration online is snapshots over time: animation.

The future of newspapers online, especially for great ones like the New York Times, is to become masters of not maps, facts, and drill-downs, but of the timeline: animate or die!

3 thoughts on “A map is worth a thousand pictures…when animated

  1. gizmometer: “greater data misrepresentation”

    Yes, great danger. However, snapshots are just as prone to misinterpretation as snapshots-over-time. Piling facts upon facts is easy, parsing and analyzing them to arrive at trends is far more difficult to pull off. The hardest is to distill the findings to tell a story in one minute against a timeline. But that’s exactly what makes it so much more valuable. Nurses and lab technicians can give us a lot of facts about our bodies, but it takes a doctor to put it all together, make a diagnosis and paint for us a picture we can understand.

  2. Animation, though highlights time, does not show causation, though it can easily mislead a viewer to assume a cause. I agree with you that the medium of the web page enables a richer visualization of supporting graphics to news stories; however, it presents an equally great hazard of greater data misrepresentation. In particular, trend-spotting is a serious problem. A short-term fluctuation can be misinterpreted as an indication of a long-term change. The problem actually comes down to one of representing scale. The scale of distance has been the cornerstone problem of all of cartography for centuries, but in animated visualizations, it becomes the scale of time. Just like a map that is not drawn in 1:1 scale, an animated visualization not drawn to real-time can quickly become misleading. I strongly agree with your point that the newspapers can use the medium of the web to tell a new kind of story; I caution them at the same time to watch out that they don’t accidentally tell us lies.

  3. Visualisation is powerful. Visualisation of dynamic events even more so. I have some experience from the aerospace industry, where you rendered dynamic 3-D scenarios of events and objects, instead of just looking at telemetry data. It’s a big moment when you see it for the first time. You can describe it in words, but when you see it, it’s a whole other experience. Lately I have been working with visualisation techniques for statistical process control (XmR charts etc.) – i.e. are we on the right track in this software development project or not? The jury is still out in this case, and I can’t say if you get all the benefits you want, but at least you have to try. There is a first step in everything. SPC has nothing to do with innovation, though. Innovation is a whole different subject.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s