This post originally appeared on Medium and is reprinted here with permission.
When Reported.ly’s European team went online Wednesday morning, our reporting operation was less than 48 hours old. We had launched Monday at a measured, almost leisurely pace, our team of six still getting to know each other and learning how to coordinate our work across 10 different time zones. We planned to ease into our coverage over the first week, taking care not to stretch our small team too thin.
Then just before 6am ET, our European staff got word of some kind of shooting incident in Paris.
We didn’t know it at the time, but this was the beginning of a horrifying story — a terrorist assault on press freedom itself that left 12 people dead at the headquarters of French satire magazine Charlie Hebdo.
At 5:56am ET, the team retweeted a post they found on one of more than 200 Twitter lists we’ve created to monitor events around the world, including a list featuring AFP reporter Marc Burleigh.
Within five minutes, another team member found a tweet that was even more ominous:
Over the next 10 minutes, the team tracked down a handful of sources referencing a probable attack at the office of Charlie Hebdo in Paris. We let our Twitter followers know that we were now focusing on the story:
It didn’t take long to find people reporting from the scene. At 6:21am, just 25 minutes after hearing the initial reports, my colleague Malachy Browne used geolocation searches to find footage from the scene of the attack:
By the time I went online from my home outside Washington DC two hours later, the Reported.ly team in Europe — Malachy, Marina Petrillo and Asteris Masouras — had curated dozens of tweets from Paris and around the world, including eyewitness reports, reactions by French political leaders, and the now famous solidarity sign, Je Suis Charlie:
For the next 16 hours, we divided up our work based on specific subtopics and social platforms. Malachy and Asteris created a Storify collection with the latest updates from Paris, while I began collecting protest art from around the world. Marina, meanwhile, anchored our coverage on Facebook and coordinated our Twitter feed.
As straightforward as this sounds, it took a bit of orchestration on everyone’s part. All of us have access to our Twitter, Storify and Facebook accounts, and the urge to post our own updates without double checking with the team can become a bit impulsive at times. On a few occasions, we came very close to double-posting the exact same retweet or nugget of information to the same account; not a huge mistake but still one worth avoiding.
Fortunately, we’ve incorporated Google Hangouts into our workflow. While we’re on the job, we join a Hangout so we can see and hear what we’re all doing. Most of the time we keep ourselves on mute, just so we don’t have to hear each others’ pets barking and meowing. But when we post something through our official accounts, we unmute ourselves and call it out to the group, almost like a barista announcing a coffee order, just in case one of us needs to jump in and say something first. That way, if someone is about to accidentally double-post something, we can alert them to the problem.
As our west coast US team came online, we continued to update the story with a series of Storify collections. While names of potential suspects began to circulate through unofficial channels, we began preparing a dossier on them. As it turned out, one of the suspects had been tried in a French court in 2008 for trying to sneak into Syria and become a militant in Iraq. We managed to find a number of English and French news articles that covered the trial, as well as academic books digitized by Google that investigated the case.
We also managed to find a local news posting that one of the suspects had passed a science exam in high school last summer. That particular nugget got our attention, because the suspect, who’s 18 years old, is nearly 20 years younger than the two other suspects. Perhaps most interestingly, a number of French high school students in the northeast of the country began insisting on his innocence, saying he was at school that same morning. My colleague Kim Bui and I began piecing together their social network, connecting the dots between around two dozen students, all of whom appeared to be either friends or classmates of the suspect. This resulted in another Storify about the suspect:
Before the night was through, a hashtag campaign in solidarity with the suspect was trending on Twitter across France. He voluntarily turned himself in to authorities, and local news reports suggest that police are following up on his alibi, interviewing some of those friends we had discovered earlier on Twitter.
As part of our verification efforts, we put a lot of emphasis on geolocation — ie, using the GPS metadata contained in some photos and videos to confirm their precise location. This had helped us find footage from the scene of the attack, as well as locations where police tactical teams were converging as part of the manhunt. At one point, someone circulated a photo of police near a Halal butcher shop, allegedly in Reims, France. It didn’t have any geolocation data associated with it, but after a bit of sleuthing we were able to identify the exact location using Google Street Map.
On other occasions, our work didn’t involve any bells and whistles — instead it focused on how to handle unconfirmed reports from other news outlets. A couple times during the day, there were reports circulating widely that hadn’t been confirmed yet. At one point, several news outlets reported that at least two suspects were under arrest, one of them even suggesting that the third had been killed. For whatever reason, it reminded me of CNN’s erroneous reporting of suspect arrests following the Boston Marathon bombing. We didn’t want to suggest the reports were true, but we didn’t want to ignore them either. So we did our best to put the reports in an appropriate context — that it was out there, but far from confirmed. Eventually, the news organizations walked back the story and acknowledged that no one had been arrested or killed at that point.
Around the same time, a handful of individual Twitter users were quoting three specific names of suspects, but nothing had been confirmed yet. A few news outlets decided to run with the names early, but once again I thought back to the Boston bombing and how several people had their reputations tarnished because they were erroneously connected to the attack. Still, it was clear that the three names were circulating more and more widely, so we concluded it was important to acknowledge that names were floating around, but that it was still important to be extra cautious in situations such as this.
Eventually, the Associated Press went public with the three names that we had held back. For us, that was as good a confirmation as we were going to get.
By the time we signed off at midnight ET, we had been reporting on the Charlie Hebdo attack for exactly 18 hours. It was a baptism by fire for the team, and all in all I think we did our jobs fairly well. Having said that, I think there are a number of areas that could still stand some improvement.
- Facebook. While I’m happy with the posts we made on Facebook, it lacked the engagement we experienced on Twitter. Granted, the Facebook page still has a fairly small audience and is growing at a slower pace than Twitter, but I think we could have pushed a little harder to converse with the people there, perhaps even involving them in our sourcing.
- reddit. We posted updates to our reddit page, but it would’ve been great to have established a reddit live blog, which has become a great tool for real-time reporting. To put it simply, we were stretched too thin. We need to put some thought into whether this was due to a matter of juggling too many platforms and angles to report with a limited amount of people. My sense is that reddit live blogging is something we could excel at, but we’ll need to become more efficient with our workflow first.
- More frequent recaps. When we launched Reported.ly, we promised we’d put an emphasis on regularly telling people what we know, what we don’t know and why. While our tweets did tackle this over the course of the day, they weren’t particularly explicit about it. Since most people aren’t glued to Twitter for 18 hours straight, it’s important that we make a habit out of reminding people where things stand, just in case they’re tuning in for the first time.
- Choosing the right Twitter accounts. Sometimes I felt like we should have focused on posting via our personal accounts and then retweeting via the @reportedly account. Other times I felt like it should be the reverse. In both cases, I didn’t feel like I had a solid grounding on how we should make that call. We still got through the day just fine, but I think we should put more thought into the pro’s and con’s of doing it one way or the other.
Overall, though, I think we managed to get through Day 3 of our existence without any major mistakes, while at the same time pulling off some decent original reporting and storytelling. Now it’s just a matter of learning the right lessons from that day and applying it to the many days ahead of us when we’ll need to perform at our absolute best.
So how do you think we did? Were there things you liked or disliked, or felt we should have done differently? Were there any opportunities we might have missed? Please feel free to let us know what you think on Medium, or @ reply our Twitter account, @reportedly.