loader-logo

When do you stop verifying?

An important part of any investigation is verifying pieces of evidence that you or your team have managed to collect, be it images, videos, or statements made by witnesses or court documents. In most cases, easier said than done. There are a number of variables that determine the difficulty of verifying, say an image. The context and quality of an image can determine if it would take you 10 min or 10 hours to solve. Luckily, there are plenty of guides to help you geolocate and verify images or videos. I have written one or two guides myself. However, when do you stop? How do you know when its not worth your time?

I often enjoy reading and writing about the creative ways in which people have verified online content. Over the years, people have been able to create new tools and techniques to squeeze every last from of information out of a photo or video. I recall a story about Bellingcat founder, Eliot Higgins, who had used a computer program to make sense of a license plate in a dog kidnapping. The dog-owner had contacted Higgins in desperation. The police had previously determined the CCTV footage of the incident ineligible. However, by using a tool originally developed by a colleague for a murder case in Ukraine, Higgins was able to make sense of the license plate within an hour. The dog and owner were reunited less than a day later.

Another interesting case study was the geolocation of videos of Al-Werfalli execution sites. Al-Werfalli was a commander in the Libyan National Army and in 2017, had an arrest warrant from the ICC for his involvement in mass executions in and around Benghazi. One of the videos was geolocated with the help of crowdsourcing. A twitter user mentioned that the incident most likely took place in the southwest part of Benghazi due to the color of the sand. Demonstrating how useful it can be to have a number of trained eyes and minds helping bringing new ideas.

However, odds are your colleague has not programmed a license plate / CCTV footage decoder or your image is too sensitive to your investigation to reach out to Twitter for help. And despite all the tools and tutorials out there, if you do not have access to the original image/video being 100% certain is often going to be time consuming or impossible (though I often like to say otherwise). Therefore it is good practice to know when to stop and move on. Yet, if you are down the rabbit hole it might not be so easy. So, it is also a good idea to learn from others for the next time you are in doubt. I recommend reading the case study called ‘Not everything is verifiable, but that’s OK’ by Ray Adams Row Farr for Citizen Evidence. The author talks about how the decisions that lead to her eventually moving on as well as the lessons they learned.

Read the full article here.