When WikiLeaks gave seven news organizations access to 400,000 previously classified military documents pertaining to the Iraq war, one of many challenges that those organizations faced was how, if possible, to try to visually depict some of the information for their readers. This massive amount of new information called for more than photo slideshows, clearly, but each team took a slightly different approach based on its expertise, available staff time, and what content they chose to emphasize.
Al Jazeera English, for instance, made several basic charts, graphs, and maps for readers to click through, and an animated map of IED attacks across Iraq. Der Spiegel created an interactive map of a single day of violence, as did the Times and the Guardian (though they all chose different days). The New York Times’s maps of violence in Baghdad are, like most Times graphics, clean, informative, and straightforward. The Guardian’s Flash-animated and text-narrated map of one day in the fall of 2006, plays like a very understated documentary film, and is much more emotionally affecting.
Simon Rogers, editor of The Guardian’s Datablog and Datastore, who also spoke to CJR this summer after WikiLeaks released thousands of Afghanistan war logs, said that he and his interactive team felt more comfortable with the data this time around. There were more than four times as many documents this time, but the process of trial and error from the last time informed how they chose to filter and present this new information. For instance, before they attempted any visualization project, they found that they could analyze the data just by manipulating rows and columns in a spreadsheet, grouping the reports by type or time period.
They found, for instance, “that murder was the biggest cause of death, which is really indicative of the sectarian violence that was going on in Iraq,” Rogers said. “You can actually learn quite a lot from just doing a bit of basic work within Excel, with that large size of data.”
As Rogers did last time, he also released a small portion of their spreadsheet to the Guardian readers to download for themselves, to see what they could find from sifting through it. Jacob Shapiro, a professor of politics and international affairs at Princeton, quickly responded with his concerns about the limitations of the war logs as a whole, concerns which Rogers then posted and discussed on the Datablog. Rogers said he welcomes this kind of input, and is eager to see what readers do with the data, though he acknowledges that worthwhile analysis will take a sizeable commitment of time and attention.
“I don’t really know what people will do with it, taking it on,” Rogers said. “I suspect some of the real interesting stuff will come from academics, who’ve suddenly got access to this data, and will spend a long time looking at it. I don’t think it’s the sort of thing you can just dash off.”
As for the question of how to best present the logs to Guardian readers, as an experiment, just to see what it would look like, Rogers first used Google Fusion tables and Google Maps, both free tools, to compile all of the deaths reported by the Iraq war logs. That map is on the Guardian website now, and Rogers said it has gotten many more page views than he expected. But he also said that he and his colleagues were trying to be careful not to “sanitize” the information. Each dot on the map represents not only a military event, but a human life gone. Dots on a map aren’t enough.
After sifting through their data, the team at The New York Times assigned to this project (Jacob Harris, Kevin Quealy, Sabrina Tavernise, and Andrew W. Lehren) also realized that they would need a way to focus the experience for their readers. “We didn’t have many preconceptions about what the graphic was going to be,” wrote Quealy, a Times graphics editor, in an e-mail. “We started with the data—a big file with about 17,000 GPS coordinates and dates—and worked from there.”