By: Rich Kane
Data journalism. Numbers and words.
Oil and water, for many a reporter and editor.
But the U.K.-based Guardian is doing something that makes the cross-pollination of these two orbits a bit less intimidating. Enter Swarmize (swarmize.com), which takes the job of collecting data from users and turns it into information that readers can easily comprehend. Development for Swarmize was paid for out of a $35,000 Knight Foundation grant.
The key is simplicity, according to Matt McAlister, the Guardian’s general manager of new digital business.
“Editors can set up simple polls and embed them onto a Web page in minutes,” he said. “Developers can build more custom interactives like a live TV feedback mobile app. And Swarmize customers can pour in high volumes of data, such as all the tweets from a popular hashtag, as we did during the Scottish referendum in September.”
To use Swarmize, an editor will create a “swarm,” adding questions about whatever they like from a form builder. Then, they either embed the form, or send the swarm to a developer, who can then build a custom interactive from the swarm. The developer can also pour in data from other sources like Twitter hashtags.
A good example of Swarmize can be found at fergusonnext.com, which was put together as a collaborative piece by several news outlets, including the Guardian, the St. Louis Post-Dispatch, and the Riverfront Times. The site is essentially an online community forum set up following the death of Michael Brown, and solicits comment and ideas from the community.
And Swarmize data is designed to last. McAlister said that one of the problems with most data collection is that it seems to die after the initial news report.
“The data behind an article or infographic often exists in a spreadsheet on someone’s computer somewhere,” he said. “It might be on a publicly accessible Google Spreadsheet if we’re lucky. But how would you ever go about finding that spreadsheet again after that news story is over? It’s not easy. Swarmize was designed to make data findable, usable and re-usable. It’s easy to search for data. It’s easy to get data from a previous swarm. And we’ve even added a ‘clone’ feature so editors can benefit from swarms that have happened in the past.”
As an example, McAlister said an editor could take data from one source, such as local government spending, and map it against data provided by users in a swarm such as their opinions of local government services.
“We think Swarmize can help with everything from lightweight surveys to custom interactives to large scale, high volume, fast-paced, data-rich initiatives,” he said. “We can tell all kinds of different stories.