Setbacks and Adaptations
About the author: Nitish Vaidyanathan ’21 is an FSI Global Policy Intern at the Tahrir Institute for Middle East Policy in Washington, D.C.
It’s been a few more weeks into my time at TIMEP, and in DC. Some exciting stuff has happened in the news, and we’ve run into some problems with our original research model. But like that Bear Grylls meme, we improvised, adapted, and overcame.
As I mentioned last time, I cover Iraq, Iran, and Syria for the Daily Research Brief, and I ended up having to track a crazy story: for the first time in 38 years, Israel had struck military sites in Iraq. You might have read about this in late August, but I had been tracking the story since July 19, when a bunch of videos of a fire at a PMF base in Salaheddin showed up on my Twitter feed. While this was sort of cleanly reported on a month later in the national news, in the hours after the strike happened, nothing really made sense. I remember sending out three versions of events to the team, all of which were supported by various actors on the ground: one claimed that it was just a fire started by an errant worker, another claimed an ISIS drone dropped a grenade on an ammunition depot, and the final one (as we later found out, correctly) stated that it was an Israeli strike. A few hours and half a planet removed from a fire in a military base, the best I could offer the team was multiple choice. But the second strike came, and this time it had a lot clearer markers of a coordinated military strike, and examining satellite footage I was able to venture an educated guess that it was in fact an Israeli strike. Most US news outlets only started to really carry the story after the fourth Israeli airstrike in Iraq. It was really bizarre being that ahead of the news, especially on such an important event. It was an interesting tutorial both in the fractured information space in the news cycle, and in the importance of conducting open source research with an eye to local sourcing. Because of the TIMEP team’s emphasis on using the DRB to follow local stories with a granularity not easily found in the news, we had a full month of lead time on the story.
With regards to the PMF tracking project, we ran into some issues. The principal one was the massive selection bias on our Facebook data. The Facebook data that we were looking at relied on the PMFs own various groups posting their location accurately. What this led to, however, was overrepresentation of groups that had poor operational security, and underrepresentation of the most dangerous groups with good operational security. The most ‘proxy-oriented’ groups like Kata’ib Hezbollah were hardly likely to post locations of ballistic missile depots on Facebook. Moreover, our data wasn’t granular enough to be able to sharply differentiate between groups. So we decided to try something else: we went to the PMF’s official website, and we scraped the data directly from there. With the help of my roommate, I built a Python tool using Beautiful Soup 4 that scraped location data from the PMF website. Using this, we were able to build a new dataset. Next week, I’ll talk about our analysis of this.
Lastly, some stuff about DC: it has become evident to me that Ethiopian food is the best food on the planet. The SIW dorm is located right next to Adams-Morgan, which while being the most stereotypically liberal DC neighborhood, complete with RBG flags on every lamppost, has amazing food. Food in DC is not exactly inexpensive, but more often than not you get what you pay for. The SIW dorm is also right next to the Metro Zoo, and here’s a picture of a tortoise that I took there.
