Reddit Crowdsourcing
The most widely talked-about technological development was the attempt to crowdsource identification of the perpetrators.
The concept behind this is that by collecting images from sources including mobiles and security cameras, sharing them on social media and getting a large number of people looking at them and conversing, it would create a kind of organic big data search. Add in publicly released police information, for example about what the bombs may have looked like or leads on the suspects, and in theory this should be a powerful tool.
Unfortunately, it wasn’t. The attempt to find the bombers was a disaster which caused innocent people to be named as suspects, and which has been described as an “online witch hunt” due to the highly charged emotional states of the contributors, the public nature of the investigation and the evidence which seemed to say more about the mentality of the amateur sleuths than reality. In fact, it seems that the online investigation was doing so much damage that it prompted police to release images of the eventual suspects earlier than they would have otherwise.
This should come as no surprise to anyone familiar with the painstaking methodology of professional intelligence work. Intelligence analysts are selected based upon their analytical abilities and trained to be able to override their human thought processes and instincts, and to remain dispassionate when dealing with such deadly scenarios. In contrast, the contributors on Reddit had a methodology derived from creative internet group work.
This is the first time such an ambitious project has been undertaken by an online community and it certainly won’t be the last. As technologies develop further we will have access to more information and it will be of better quality; imagine for example if there had been several hours of HD video from hundreds or thousands of Google Glasses available. We are also starting to see technologies like Briefcam which can summarise video and which was used by police to sift through the available content.
As such technologies mature we will be able to increase the efficiency of analysts, helping them reach conclusions faster. Perhaps the most effective use of public crowdsourcing will be to have police coordinate these efforts, perhaps to add some structure to what people are looking for and of course to decide what gets followed up.
What can we learn from this? We have an insight into how people want to work. The instinctive reaction was to collaborate. Collaboration occurred on a large scale, and was driven by a sense of shared purpose and a feeling of empowerment. Existing social networking platforms were repurposed; and combined in order to make the best use of each tool. We need to learn from this: how hard can it be to give people tools they feel are effective, and a sense of purpose? Yet research suggests corporate collaboration tools fall short.
Phone networks down?
Almost immediately after the bombs detonated, many people in the area found they were unable to use their mobile phones, because the cells overloaded. Other pieces of mobile technology were quickly drafted in to help find people. These included the Athlete Tracker, which was hacked to display the locations of its devices, Google’s People Finder, and calls to open private wi-fi.
Mobile networks are managed through a complex layering of cells with capacity at each layer to take up the expected volume of calls. When an incident causes a sudden spike in demand, some phone calls will get through but not others. SMS messages use a different part of the phone connection (the “control” rather than “traffic” channel) to voice calls, so in an emergency it is often easier to get a text message through than a phone call.
What can we learn from this? All the ways today’s technology provides to stay in touch in emergencies use either phone signal or the internet. This has ramifications for how we are increasingly trying to share information over these systems; in routine, everyday operation they work fine, but will overload when something interesting happens, and this is when we get the most value from sharing. Therefore, it is becoming increasingly important that we adopt smart technologies which are not limited to particular platforms, giving us freedom and flexibility to interact how we need, when we need to.
Should we even be doing this?
The bombings have also raised concerns over the role of Twitter and Facebook in spreading rumours, their relationship with mainstream media, and technology websites publishing on the story. This is more complicated than the “mainstream news vs. amateurs” line that some news organisations implied.
The problems are the same as we’ve always encountered with news, in that in a situation where the picture is continually evolving, there will be well-intentioned misinformation. News reporting is a balance between speed and accuracy, and even before the rise of Twitter and Facebook the news media had been driven to be “first with the news” (as opposed to getting the truth). This dates back at least to dedicated TV news channels, with gaps between new information being uncovered being filled with unsubstantiated opinion.
The other part of the news discussion revolves around the threat traditional news organisations feel from the internet. For example, it has been suggested that tech sites shouldn’t be posting news that isn’t about technology. The tech sites see themselves as news sites, albeit technology-focused news sites, which can report with greater authority on technology issues (a point brought home in the wake of the claim of using Android phones to control airliners, when the FAA’s press statement got some significant facts a bit confused).
What can we learn from this? The pressure to be first with the news and provide live updates means that for all their complaints about Twitter and Facebook, that’s increasingly where the news media will be getting its information. So when important news breaks, whether in our companies or the wider world, we need informed, aware people trained to sift through the noise of social (and other) media and report reliably on what is happening. Of course, we used to have such groups. They were called “journalists”.
By David Akka, Managing Director at Magic Software Enterprises UK.
