How we balance speed with rumor control: With any big story these days, social networks are both an asset and liability. Matt Roller summed it up this way, tweeting this soon after the Boston explosions:
“Twitter does its best work in the first five minutes after a disaster, and its worst in the twelve hours after that.” - @rolldiggity
At Breaking News, our goal is to balance speed with an editorial filter, keeping rumors at bay while incorporating hundreds of sources. With the Boston story, we’re not quite as fast as Twitter (the crowd is always faster), but we provided a lightning-fast stream of coverage that avoided or downplayed nearly all incorrect reports, including a third explosion, a bombing at the JFK Library and a Saudi suspect in custody.
Here’s our technique: moments after the explosions, our editors tracked dozens of Boston news sources — news organizations, officials and eyewitnesses — looking for a new report on the story. Just as on-the-ground news organizations compare sources before reporting new information, our editors compared these new reports with coverage from other news and official sources.
For example, when one news organization reported that five unexploded devices have been discovered, we noticed that no other news organization or law enforcement source was reporting anything similar. In that scenario, we table the story and wait for a second news organization to confirm with its own sources or an official source makes an announcement. There was no confirmation, and the story turned out to be false.
We use this multiple sourcing approach for reports with a higher risk or a history of being wrong, and that’s where our editorial experience comes into play. Early reports from mass casualty stories are often in flux, and here are a few examples we’ve learned:
- Leaked reports of suspect descriptions are nearly always wrong and overly vague — in this case, it was a “dark skinned” or black man with a black backpack — so we avoid reporting any description until officially released by law enforcement.
- Initial reports of a second suspect are nearly always wrong. Most mass shootings in the U.S. are initially reported as a “possible second suspect” but nearly universally are the work of a single person.
- Casualty counts are often inflated in the first hour or two, sometimes dramatically. So we tend to report these numbers carefully. Interestingly in Boston, the opposite was true: early estimates were remarkably low.
These are just a few of the red flags that we’ve learned along the way, and they also apply to eyewitness reports, which we run through a more stringent verification process. We have two rules of thumb: 1) If it’s too good to be true, it probably is 2) As much as we want to be fast, it never hurts to wait a beat.
This is an imperfect science, and we’re not always right. But when we’re wrong, we want to be the first to admit it, applying the same urgency of telling a breaking story with the occasional necessity of correcting it.
We’re working on a new version of Breaking News — coming early this summer — that takes our unique blend of reliable, real-time coverage a step further. Stay tuned…
(Post by @corybe. Photo by Elise Amendola / AP)