Please view this site in landscape mode.

by Christopher Radler

After having been rather illiterate in the field of Social Media and the huge potential it offers, and only being able to receive a first glimpse through an online course conducted by the TechChange Institute for Technology and Social Change, I was delighted when the German Center for International Peace Operations, in close cooperation with various other organizations, thereunder the ICT4Peace Foundation offered a Crisis Information Management (CIM) course in Nairobi, Kenya.

Amongst the various CIM tools covered by the course was also the use of Social Media. Hence, I was delighted to be given the opportunity to deepen my knowledge in that particular field. The ICT4Peace, represented by Mr Sanjana Hattotuwa, one of the best teachers I ever experienced by the way, though only revealing the tip of the iceberg, opened a whole new world to me as a Security Risk Manager. Tasks such as information gathering, incident and conflict mapping, travel security, crisis and information management are an everyday challenge and require a variety of sophisticated and very expensive tools. Without further elaborating on the subject: Social Media has the potential to revolutionize this and gives smaller organizations solutions at literally no cost. But I'm wandering from the subject.

The author during a visit to the iHub, as part of the course at IPSTC

Back to Kenya where subsequent to the course the 2013's presidential elections took place and I was foresighted enough to not fly back home directly. To cut a long story short: I participated at another training in Nairobi's iHub, Kenya's version of the Silicon Valley and home to a company which provides some one of the most sophisticated Social Media tools I discovered so far: Ushahidi. One of these tools is Uchaguzi, exclusively developed to crowdsource information on the polls and subject to the training mentioned above.

Up to this point it was just an extension to the lectures I had received from Sanjana and that in itself would have been good enough. But than fate would strike me by letting me cross the path of a former colleague and friend of mine, who was hired to supervise the CIM of a huge company. However, given that a new outbreak of hostilities between the various ethnicities was not unlikely, as it was the case in the aftermath of the presidential elections of 2007/08, I was immediately hijacked to support him. And I gladly let it happen as it would give me an opportunity to apply the newly gained knowledge in real life.

When engaged in information gathering it is of utmost importance to utilize as many sources as possible. You have your personal network, which shares information with you, you have your people on the ground and all kinds of Open Source Intelligence (OSINT) in earlier times mostly mainstream media. While your personal network in a new environment is most likely to be not that extensive and your (professional) sources on the ground are limited for obvious reasons, mainstream media mostly gives you already processed information that on top of it is often delayed. Receiving real-time information from the ground on a large scale was a thing one could only dream of. Here is where crowdsourced information, in the following especially Uchaguzi, kicks in and adds a new dimension to information gathering.

Uchaguzi basically allows everybody with a mobile phone or access to the internet to submit information via SMS, email, Twitter etc. to the Uchaguzi website where it is mapped in real-time. A dedicated team of volunteers edits these reports, i.e. translates, geolocates and verifies it by sending people to the field. Of course, due to the huge amount of information received, not every report can be verified but this is clearly indicated on the platform.

I need to make a preliminary remark: Public unrest is all about group dynamics and expecting it can easily become a self-fulfilling prophecy. Let me put it that way: If you expect someone to attack you, though it might not be his intention, you are likely to strike first if you have your back up against the wall. At the end of the day perceived reality is reality!

So, at the day of the elections we accessed the platform and monitored the incoming reports. Though limited in many ways and often of questionable reliability, they were allowing us to capture the ambience on the ground and by taking into account the above mentioned, thereby foresee trends. For instance, Uchaguzi allows you to filter reports by category. First of all, we could identify that there were no major security incidents (except one in Mombasa), which was further corroborated by Twitter and, of course, our people on the ground. There were a lot of reports in the morning regarding minor security issues, which we on the one hand traced back to the slum areas – easily done through Uchaguzi's mapping function –, where violence is rather common, and on the other hand to a general tension amongst the population. Accordingly, the number of incidents in the non-slum areas decreased rapidly after it became clear that the general mindset was peaceful. This was indicated by the fact that Twitter was drowned with tweets promoting peaceful elections for instance. Hence, we concluded that there most likely will be no major incidents throughout the day. Interestingly, from an ex-post analysis point of view, it would have even been possible to foresee the increase in incidents shortly before the polling stations were officially supposed to close: As it was clear that the polling stations could not handle the enormous voter turnout in due time (there were queues reaching up to a mile!), tensions rose. Once it was clear that every voter will be able to cast his ballot, inter alia twittered (!) by the Independent Electoral and Boundaries Commission, incidents decreased again.

Another issue of major concern was if there might be organized election fraud, which could have easily led to unrest. By filtering Uchaguzi for Voting Issues and Polling Station Logistical Issues, we observed an accumulation in the large polling stations, which we ascribed to organizational issues, and thus were able to rule out organized election fraud.

To sum it up: Social Media did not only serve as a great monitoring and CIM tool for our operation, but it also helped to calm down tension amongst the population. I need to stress, however, that I am only talking about the election day itself and no one knows how the situation will evolve after all the ballots are tallied or a candidate is announced winner, respectively. Because unfortunately, Social Media could also serve as a means of mobilization of unrest. After all, it was a great experience to utilize those tools for CIM and unlike the Titanic, I would be happy striking what is below the tip of the iceberg in the future!