Archiving posts from the Capitol attack has value for police and researchers

Since the attack on the U.S. Capitol, which was filmed and photographed extensively, there’s been a scramble to find and archive all those images. Law enforcement and researchers are collecting them for clues and also to understand what happened. The research and investigative journalism site Bellingcat collects open-source intelligence and publishes reports on news and global events with a small staff of researchers and digital forensics experts and a big crew of volunteers.

I spoke with Giancarlo Fiorella, an investigator at Bellingcat. He said the site just published a sort of forensics report on the movements of the San Diego protester Ashli Babbitt, who was shot and killed during the riot. The following is an edited transcript of our conversation.

Giancarlo Fiorella: We published an article yesterday outlining “The Journey of Ashli Babbitt,” was the title of the article. So by finding all of these videos that also showed Ashli Babbitt, we were able to, for example, reconstruct somebody’s journey through the Capitol. From the moment that they first entered the building, through the time that they were walking down the corridors to, in the case of Ashli Babbitt, her very last moments when she was shot.

Molly Wood: You know, it’s interesting because you’ve been at this for a long time at Bellingcat, and yet there is this parallel outpouring now of amateur digital archiving and attempts at identification and aggregating data related to, let’s say, this attack. How are you doing it better?

Fiorella: That’s a really good question, and I want to make a distinction here between the archiving and the identification. So imagine you’re on your couch, you’re surfing through YouTube. The algorithm recommends a video that you hadn’t looked for, it was just there, and you notice that it’s a video of the Capitol riot. You might submit that to us, and then we would archive it for you. What we don’t encourage people to do is engage in crowdsourcing identification because that can have really disastrous results. We’ve seen in the past what happens when individuals are misidentified as having perpetrated some sort of crime. For us to say, “Look, we know that the person in this video is — this is their name, because we found their Facebook page,” that’s a very deliberate process. And the chance that somebody will make a mistake is multiplied.

Wood: Even in the process of archiving, are there privacy concerns at all about the, let’s say, a Google Sheet that you’re compiling that anyone can click on?

Fiorella: Well, our view is that because we’re working with open-source information, this is information that is on the internet already. We’re collecting it into a single database. And I should say that our effort is not the only one. In fact, ours is interesting, I think, because it involves public assistance. But there are other groups that have been involved in archiving efforts that can have much more data than we do in terms of hard disk space, like they have terabytes and terabytes of data because those efforts have been conducted [and are] mostly automated. So ours is like hands-on archiving. It’s sort of an old-fashioned way to do it. But again, our view is that, if you were there and you took a video of yourself in the Capitol, and then you put it on YouTube — if you don’t want people to see what you did, not just at the Capitol, but just in life in general, then don’t record it and don’t put it on the internet, because that’s precisely the point of uploading information on the internet, for other people to see it.

Wood: What do you think about the right to be forgotten? As some of these things would be maybe taken down by the people who posted them because they are sorry and/or went to jail or want to be forgiven, but you’ll still have it?

Fiorella: You know, we’re in an interesting position because I am also a social media user, and I’ve shared stuff online that maybe I shouldn’t have. But there’s also, I think, an argument to be made for the historical record. We know from, for example, the Charlottesville Unite the Right rally that the first materials that get deleted are the ones that are really incriminating. So I think there is still a need for people to be out there, making sure that that sort of information is not lost. I totally sympathize with the right to be forgotten. But as a researcher, I’m also, maybe “grateful” is the wrong word, but shocked that so much information has come out about this.

Wood: What is the ultimate goal of your research at Bellingcat?

Fiorella: In relation to this project, as I said, we want to have as much information about what occurred archived and available. Right now, we are looking to the weekend and the inauguration because just as there was ample warning on social media that there was going to be violence at the Capitol on January 6, there’s also ample warning now that these same sorts of groups are going to take part in rallies over the weekend and at the inauguration. So we’re putting together a team of volunteers to live-monitor these protests. And they’re going to be helping us to identify videos that show, for example, activities by far-right groups, videos that show violence, if violence breaks out anywhere. And the idea will be to have these volunteers help us collect these videos and archive them so that we, with time, have a chance to look at them. And if there’s a need to clarify an event or publish an article outlining a particular high-profile event, we’ll have that data available, and we will be able to do that sort of research.

Related links: More insight from Molly Wood

The organization is publishing videos of the Capitol riot that were downloaded from the Parler app as a thread on Twitter and also published a report on Ashli Babbitt. But I wouldn’t watch too much in a row because the truth is it’s traumatic.

The latest platform to discover that it has a QAnon problem, that it was either ignoring or just didn’t think was that big a deal, is Nextdoor. The Verge reported Wednesday that moderators had been asking for a ban on QAnon conspiracy theories since at least the fall — like the idea that Democrats and the media are part of an elite cabal of pedophiles conspiring against President Trump. Nextdoor has misinformation policies that moderators say are highly specific to the presidential election and COVID-19.

And while Nextdoor told The Verge that it considers QAnon a hate group, its online misinformation policies don’t say that, and moderators haven’t been explicitly informed. Moderators say they’re hamstrung because posts containing conspiracy theories are often not explicitly violent, even though the QAnon conspiracy theory actually does culminate in a thing called the Great Storm, where the evil deep state is rounded up en masse and executed. So please consider that, by the way, when you wonder what the goal of some of the people at the Capitol might have been and when you contemplate the hashtag “Storm the Capitol.”

But also, we are again asking businesses and social media platforms to police a cult that is being encouraged and supported and echoed in the halls of Congress and that some of your neighbors and friends might sincerely believe in. And I can kind of sympathize with Nextdoor being like, “I don’t know what you want me to do here” because these are deeply troubled times. Because let me tell you, if I were in charge, I wouldn’t know what to do either.

Stories You Might Like

Let’s talk about accountability

Social media has been radicalizing people for years

Security costs after last week’s Capitol attack

Another fear after Capitol attack: information security

“Disinformation laundering” got us here

Surveillance tech is not accomplishing the things it’s supposed to