Leveraging the Expanding Social Network

Ever since the first social media app appeared on “smartphones,” the U.S. public safety community has sought to tap into, and effectively use, the real-time feed of localized, first-person inputs to improve their own ability to monitor and respond to incidents, both large and small. Indeed, the concept of operations for “NexGen 911” portrays a future in which calls to Public Safety Answer Points (PSAPs) will result in “live links” between dispatch centers and callers’ devices. Live links would not only enable audio communications, but also provide precise location information, digital images, and even live video – that can then be captured by dispatch centers and used both to aid response efforts and to support the ongoing monitoring of events.

This capability already has been piloted in many jurisdictions across the United States, with some solutions already in service. In addition to the direct lines used to dispatch centers created by 911 calls to PSAPs, the vast Internet-accessible universe of social networking can provide an even richer source of information to emergency managers.

“Occupying” the Internet – A Two-Way Tool 

A good example of the value of this real-time information occurred in January 2012 during the so-called “Occupy” demonstrations in Washington, D.C., and other major cities. After the United States Park Police (USPP) faced criticism for not enforcing its “no camping” rule, National Park Service Director Jonathan Jarvis initially affirmed the protesters’ First Amendment rights to assemble, but ultimately notified occupiers that, as of 30 January 2012, the Park Police would be removing tents found to be in violation of the no-camping rule. In addition, the Occupy participants in Washington’s McPherson Square were notified by USPP officials exactly when the enforcement action would begin.

One result of the advance warning was that Occupy participants carefully planned for comprehensive coverage of the evacuation by using their own handheld devices – connecting to Twitter, Facebook, Multimedia Messaging Service (MMS), Ustream, and other outlets – to thoroughly document the USPP actions. However, although the Park Police had their own monitoring resources on site, including closed-circuit television (CCTV) cameras positioned on buildings surrounding the square, the value of the content provided by the USPP’s own monitors was dwarfed by the constant stream of information coming from dozens of Occupy “protester dispatchers,” who were (unintentionally, in all likelihood) providing incident commanders with rich, real-time information in unprecedented ways.

Emergency managers tasked with carefully observing any disturbances taking place during the evacuation were able to watch the live Ustream videos broadcast from numerous Occupy monitors. The result was a stream of videos, Twitter updates, Facebook images, and other information that provided the USPP incident commanders and other authorities with an unprecedented view of the evacuation process as it unfolded.

In addition to helping in the response efforts themselves, the live updates also enhanced officer safety. In fact, after one officer was struck in the face with a brick, the suspect was quickly apprehended with images and videos provided by the protesters themselves documenting the event. In the end, the event was comprehensively documented in a way that only the Internet can do. The evacuation actions also were electronically preserved for later analysis by anyone, including the general public and the press, to use for their own lessons-learned analyses.

Humans Vs. Machines: The Humans Sometimes Win 

The challenge for any large National Special Security Event (NSSE), such as the January 2013 Presidential Inauguration, is to find a legally and operationally effective way to filter and harness the flow of information available into a digestible set of actionable operational options that can be used by incident commanders. However, although real-time analytical solutions on social media feeds have been piloted, the technology is still in its infancy. Moreover, recognition by the general public that their Twitter and Facebook feeds are being monitored could lead to a mechanism that criminal elements will be able to use to intentionally publish disinformation. By creating plausible distractions for incident commanders and/or causing resources to be deployed or transferred unnecessarily, such disinformation could help create exposed areas where security could be compromised.

Today, the best analyses continue to come from human beings who can discern, associate, and mentally cross-reference information in ways unmatched by automated solutions, particularly visual information such as live video. Because computers still struggle with analyzing visual information, a number of programs – CAPTCHA, for example – have been developed to “protect websites against bots” by creating and analyzing tests “that humans can pass but current computer programs cannot.”

In addition, newly integrated CCTV systems that allow for the creation of “on-the-fly” video walls made up of multiple streams – including CCTV, Ustream, and other outlets – of information are something that only a human can quickly turn into actionable “intelligence.” Fortunately, the ability to quickly republish such intelligence (including a specific set of video feeds) to multiple consumers has also become easier.

Companies such as SitScape (headquartered in Tyson’s Corner, Virginia) provide software solutions that allow users to instantly – via a simple browser – see informationentified by an analyst in a central environment. In such a scenario, an emergency management analyst mayentify a set of images, streaming video, and Twitter feeds that warrant immediate law-enforcement intervention, and then instantly publish the content to a virtually unlimited number of recipients, including field personnel. In short, technological advances are helping to create a “perfect marriage” between the unique skills of human analysts and the technology that supports them. However, they are not one and the same thing – not yet, at least.