From Minutemen to Smart Mobs
Address to the IEEE Homeland Security Conference, April 22, 2004
By W. David Stephenson, Stephenson Strategies
Today I will address what I believe is a major gap in homeland security strategy. It is a gap that will increase the burden on first responders if we have another terrorist attack. It is a gap that, if what risk communication scholars tell holds true, will also increase public panic, further complicating the situation.The gap regards the role of the general public in anti-terror preparation and response.There are several disturbing signs that the public is being treated as an after-thought, given limited information — and then only when officials deem it necessary.
For example:
- Because differences between the various threat levels are still unclear, and because they are applied nationally rather than on a geographically-targeted basis, people have begun to joke about them and/or ignore them — witness my earlier remarks
- The introduction of the Ready.gov program was handled so badly a year ago that all that most people remember about it is bad duct tape jokes on late-night TV. Instead of redoubling efforts after that disastrous launch, the DHS appears to have limited its efforts to perfunctory Public Service Announcements.
- The two large-scale terror simulations exercises last year, TopOff II, reportedly hired actors to portray the man in the street, rather than taking that opportunity to train neighborhood leaders who could in turn pass the word along. Even the exercises names, TopOff, were reflective of this elitist approach: they stood for TopOfficers. Of course it was important that mayors and police chiefs participated in these events, but what about the public?
As a result of this failure to inform and involve the general public, it is likely that in another terrorist attack, otherwise healthy people who have psychosomatic symptoms will overwhelm hospitals before those with real injuries even arrive. Meanwhile, experts in risk communication tell us that the combination of alarming people and not giving them a sense of mastery over the situation is a recipe for mass panic.
Fortunately, there is an alternative scenario. I found a model for the kind of emergency communications network involving the general public that we need to deal with terrorism in the history books. Specifically, David Hackett Fischer’s Paul Revere’s Ride. It sets straight a lot of myths about Revere and his ride, and, in so doing, emphasizes exactly how effective a communications system Massachusetts colonists established:
- First, it empowered individuals. Instead of the conscripted British Army, the colonial militia was made up of volunteers.
- Second, it used communications media already in the hands of those volunteers. The media upon which they relied were called horses.
- Third, it linked independent networks that might otherwise have remained isolated. Revere was, in modern terms, the “weak link” (which may sound like a negative description but in fact is a crucial role, because he was personally a member of 5 of the 7 major organizations) who united them.
- Fourth, the structure and information to be disseminated, including the means for response was in place in advance of the crisis — all that the riders had to supply was the alert to trigger the process. Most towns had designated 1/3 of their militia as “Minutemen,” ready to assemble and fight on a moment’s notice. Similarly the famous lantern system of “one if by land, two if by sea” was in place in advance.
- Fifth, the colonists’ approach was flexible and event-driven. When Revere was temporarily captured after his main mission was accomplished, other riders were dispatched to alert surrounding towns. While the British regulars in their bright red uniforms marched in formation, the opportunistic Minutemen wore clothing that blended with their surroundings, and hid behind stone walls.
- Finally, it was redundant: William Dawes also rode from Boston to Lexington, so that the word would still get through if Revere was captured or delayed
Today, we can come full circle: we can — in fact I believe, we must — replicate these same qualities in technology- driven networks that again place power and responsibility in the hands of the public.
These modern networks are called “smart mobs,” a term coined by futurist Howard Rheingold, after observing teens in Scandinavia and Asia, who would come together spontaneously after sending SMS text messages to each other.
According to Rheingold, “Smart mobs consist of people who are able to act in concert even if they don’t know each other. The people who make up smart mobs cooperate in ways that were never possible before because they carry devices that possess both communicating and computing capabilities.”
Why are smart mobs relevant to homeland security?
First, even though funding from the Department of Homeland Security is beginning to flow to communities, there isn’t enough, the government procurement process is time-consuming, and telecommunications lifecycles are shorter and shorter, so it is hard for government to stay on the cutting edge.
By contrast, consumers have an insatiable desire for new personal technologies:
- According to ABI, the number of global smart phone shipments is expected to reach 150 million units (or 24% of all cell phones, by 2008).
- Gartner, Inc. said that Worldwide mobile phone sales last year were 520 million units, a 20.5% increase from 2002 sales, in part due to strong demand for new replacement units by current users.
- Perhaps most significant for homeland security and smart mobs, as I will detail later, are projections by InfoTrends, that more than a quarter of all mobile phones to be sold worldwide in 2004 will include cameras.
As a result, an emergency communications system that capitalizes on individuals’ own technology can be constantly upgraded without significant expense by government. Even if only a relatively small percentage of the population have the newest features, such as GPS or cameras, those trend setters may be able to spontaneously assume leadership roles during an emergency, capitalizing on the power of their handheld devices. Equally important, people routinely carry their cell phones and PDAs with them on a daily basis, so that if a disaster strikes with no advance warning and they are away from car or home radios, TVs or their computer, they can receive — and equally important — transmit emergency messages instantly.
Second, a homeland security strategy that capitalizes on “smart mobs” will help reduce panic on the part of the public by giving them the information they will need to act without waiting for detailed instructions from officials. As David Ropeik of the Harvard Center for Risk Analysis says, “Giving people something to do makes them feel less helpless … the more control we feel, the less we fear.”
Third, an emergency communications network built around “smart mobs” is crucial because, as 9/11 showed, much of the conventional communications infrastructure may be destroyed and/over overwhelmed by the volume of communications. An alternative network, especially using Internet Protocol, packet-based communication such as SMS texting, may still be able to get critical information through. Or, as I’ll discuss later, it[base ‘]s even possible to cobble together an effective emergency network using those cheap walkie-talkies that you bought to keep in touch with your kids at Disney World.
Fourth, planning for the worst, first responders may be overwhelmed by the volume of casualties — and/or, as happened with the WTC, they may be victims themselves. Thus, the general public may have to rely on each other, without instructions.
Finally, because the nature and scope of a terrorist attack is so unpredictable, we may need to change strategies and tactics on a moment’s notice, and get that information to people in a way that allows them to act on the information.
“Smart mobs,” informed and empowered to act as full partners in homeland security can be a reality today because of the convergence of new technologies and new scientific understanding of the nature of human and technological networks.
Let’s first look at the technological components.
A key one that hasn’t received a lot of attention in the US (however, analysts tell us 110 million US cell phones can send and receive text messages, Americans already send more than a billion text messages a month, and use of texting grew from 22 to 45% among 18 to 24 olds last year) is SMS (short-messaging system) texting.
Texting is far more common in Scandinavia and Southeast Asia — in fact it was SMS texting in these areas that led Rheingold to coin the “smart mobs” concept. Teens use it to spontaneously gather at the mall or elsewhere. More profoundly, it has been credited with mobilizing opposition that toppled the Estrada government in the Philippines — and, most recently, with the surprise upset in the recent Spanish elections.
I work with YellowPepper, a company that designs texting marketing campaigns for clients ranging from Fleet Bank to Coke. Its Dartzs two-way texting system is also ideal for emergency communications: in fact, the company will work with the City of Boston to provide text messages for homeland security and other emergency situations.
Because text messages are packet-based, they can route around interruptions in the infra-structure and use little bandwidth. The system is opt-in, an important way of building user confidence, because people who don’t want the messages won’t be inundated by them. Marketers like it because recipients who are interested in an offer can respond instantly — the same is true with an emergency message, so the agency will know the message has been received. Because the messages pop up instantly on a cell phone display, they demand attention — ideal for emergency communication.
The Dartzs system also allows an emergency agency to create a variety of groups so that the messages can be tailored to a given group’s information needs.
Perhaps the most important technology to facilitate “Smart mobs” is the mesh network technology originally developed by Mesh Networks for the military. Perfectly complementing the real-time nature of “smart mobs,” a mesh network is independent of landlines and conventional infrastructure: self-organizing and self healing. As a result, it can survive a disaster that would destroy landlines and re-form instantly, as well as be extended to additional areas on the fly. Unlike conventional networks where each additional user that’s added reduces bandwidth, each device with a mesh card increases throughput, resiliency and capacity, serving as a router “hop” instead of having to rely on stationary, dedicated routers.
Three mesh network applications would form the backbone — a backbone that is formed spontaneously and can heal itself instantaneously if broken — of the “smart mobs” network.
Packethop.com has developed a mesh network system specifically for homeland security. The system works with both 802.11b devices including laptops, Tablet PCs, and PDAs. In a disaster, it allows a wide range of first responders from multiple agencies to share data instantly. The types of data include a wide range of broadband applications, including video, video, situation awareness, web access, and image transfer that can be used to rapidly download maps, interior building layouts, multimedia incident reports, medical information, patient’s video, or images/fingerprints from any mobile site.
Integrating with mobile mesh networks are stationary ones such as MIT’s Roofnet that use mesh networking to bring low or no-cost broadband access to both rural areas and under-served urban neighborhoods. These mesh-based community broadband networks offer many of the same advantages as mobile mesh networks for homeland security — self-organizing, self-healing.
They also offer another important advantage I will detail later: by bringing cheap or free broadband access to neighborhoods that can then be switched on the fly to emergency use, they get residents used to using the system on a daily basis, so that they will know to turn to it in a crisis as well.
The final mesh networking component is remote sensing in case of a biological or chemical terror attack. One local company, Millennial Net, builds ultra-low power transmitters, i-Beans, that are just bigger than a dime — the smallest wireless sensor devices ever. One prototype the company has developed with Ferro Energy runs on “free” energy generated from minute vibrations of pipelines or other equipment, and even the standard model can run several years on a tiny battery. Like the other mesh networks, i-Beans are self-organizing and self-healing.
The next piece of the technology side of “smart mobs” are location-based technologies such as GPS and e911 that, in an emergency, will allow sending people emergency information based on where they are at the time, including in their vehicles. The chart you see here was developed at Carnegie-Mellon University, showing (in real-time) where Wi-fi users are located on the campus. In a crisis, this mapping technique could be used to deploy mesh-network equipment to a critical area where none was deployed before.
The final technologies that a “Smart mobs” network can use illustrate, on one hand, the rapidity of change, and, on the other, the potential of low-tech devices.
Camera phones burst on the scene in the past year. They are already so popular that an estimated quarter of all cell phones to be sold worldwide this year will include cameras. While we[base ‘]ve heard a lot about their negative side — the potential for industrial espionage and locker-room embarrassment, we haven’t heard as much about their potential as an adjunct to the criminal justice system and emergency response. However, last summer, a New Jersey teen abducted by a pedophile had the presence of mind to snap a photo of the perp with his cameraphone. Alerted instantly, the police arrested him.
Just last week, LG announced a new cameraphone, specifically marketed for emergency situations. The Aladdin will automatically send a photo to three other phones in an emergency. Just think if you had photos of the 10 Most Wanted Terrorists on your handheld, recognized one sitting just feet away from you in a restaurant, and could discretely photograph them and send the photo to the FBI without calling attention to yourself?
To complete my review of the technology to enable “smart mobs” for homeland security, I’ve chosen to highlight the least-costly, least-powerful, least-advanced technology at our disposal. In Washington, literary agent Bill Adler organized a word-of-mouth network, the DC Emergency Response Network. It uses the humble $15 Family Radio Service walkie talkie that many people bought to track their kids at Disney World. Every Sunday at 8 PM, participants go outside, tune their FRS radios to Channel 1, and, without any funding or consultants, patch together an ad hoc network that can spread the word in an emergency across the District of Columbia and into the Maryland and Virginia suburbs.
At the beginning of my talk, I said that the potential for “smart mobs” in homeland security was in part due to new technologies such as SMS texting and mesh networks. Equally important, however, is that in the past 10 years researchers have learned that human networks actually follow some fairly well-established patterns. I suspect that while most of you have at least some familiarity with the technology I’ve just reviewed, you may not have been exposed to network theory. I will examine this theory in some detail, because it is critical to understanding both the power of networks and how to facilitate their growth and use.
First, when it comes to fighting terrorism, it’s time to remember something that Paul Revere and the Minuteman exploited so long ago: a highly regimented, bureaucratic organization is at a disadvantage fighting a networked one. Only this time, the roles are reversed. As network scientist Mark Buchanan wrote about the terrorists in Nexus, “The west is battling against a decentralized ‘network of terrorist cells’ that lacks any hierarchical command structure and is distributed throughout the world.” Yes, we’d love to capture bin Laden but does anyone in his right mind think that would cripple Al Quaeda?
So what does the science of networks tell us that can be valuable in homeland security?
First, we need to encourage what are called “Group-Forming Networks.”
You’ve probably heard of Metcalfe’s Law, which states that the value of a network grows by the square of its size. David Reed, former chief scientist at Lotus, realized that some networks are even more powerful — their value grows exponentially, because those networks enable and support formation of groups, clubs, and meetings that combine the power of technology with the power of friendship. Perhaps the most notable use of this power was the last Korean presidential election, where the outcome apparently was altered in the last few hours as young people. Instead of calling lists of random voters as campaigns usually do on election day for “get out the vote” efforts, activists used their cell phones to send SMS text messages to their friends urging them to vote. Evidently, the same tactics turned March’s election in Spain. Similarly, in a terror attack it’s likely that people would be most likely to respond to those who they trust the most, their friends and neighbors.
Group-forming networks can also transcend gaps between online and physical worlds. For example, the Dean campaign used the Meetup system to help groups throughout the country come together face-to-face to plan and carry out localized activities. That could be crucial in terrorism response, since it would need to be localized.
Next, as the DCERN walkie-talkie network illustrates, a “Band-Aid” solution may be best for homeland security, rather than striving for absolute perfection. As Malcolm Gladwell pointed out in his book on networks, The Tipping Point, “It involves solving a problem with the minimum amount of effort and time and cost,” and parallels the way humans communicate and process information, which is “messy and opaque, not straightforward and transparent.” I’m sure that idea will be a difficult one for many of you to deal with; it is antithetical to your high standards of perfection. Yet, as the saying goes, “the ideal is the enemy of the possible.” In this case, where it may be more important to do something quickly, then refine processes in later iterations, I think it may be realistic.
Similarly, as we found in the wake of 9/11, it is better to have a network modeled on the web, not centrally governed, messy and redundant, rather than a highly centralized one. As we saw in the technical side of my talk, the very multiplicity of available technologies– right down to walkie-talkies — will help a wireless network recover more rapidly than a highly-centralized one. That is the great virtue of mesh technology, because the network constantly reorganizes in real time. As Duncan Watts wrote in Six Degrees, his book on networking theory, in discussing 9/11, if you’ve decentralized, “You can blow a hole right in the middle of an organization, but still pump information around the damage.”
Next, one of the peculiar things about networks is that it isn’t your strong, direct connections who are most valuable in linking with others, but what sociologist Mark Granovetter termed “weak links,” those who are friends of friends. This is the idea between the wildly popular Friendster online system right now, and the LinkedIn one that is becoming increasingly helpful to introduce yourself to possible business contacts. Appropriately, LinkedIn has been invaluable to me in putting together this presentation. In a terrorist attack it will be precisely the weak links between groups, as Paul Revere was between the various patriot groups before the Revolution, who will be able to spread the word. That’s why a variety of public health outreach programs try to inform and motivate barbers and hairdressers, because these men and women touch so many lives.
Finally, a “smart mobs” homeland security strategy network would capitalize on what scientists call “emergent behavior.” This is the innate tendency in ants to form ant hills, urbanites to form neighborhoods, and people in general, as Steven Johnson writes in his book Emergence, to start change not from the top down, but bottom up, so that “people start producing behavior that’s a scale above them … the movement from low-level rules to higher-level sophistication.”
According to Johnson, there are 5 principles of emergent behavior that can be derived from observing ants — principles that apply directly to the challenge of homeland security. I suspect they are so alien to the way that most of us address complex emergency response issues that these principles will probably be hard to understand — and perhaps threatening. Please bear with me!
- More is different: you must have a critical mass of individuals before this higher level behavior is possible. Interestingly enough, individuals going about their daily routines may be unaware of how their actions contribute to a higher level pattern.
- Second, ignorance is useful. As with the quote earlier from Malcolm Gladwell about how Band-Aid solutions may be better, it may actually be advantageous to keep things simple on the individual level. As Johnson says, “Better to build a densely interconnected system with simple elements, and let the more sophisticated behavior trickle up.”
- Third, encourage random encounters. Encourage serendipity, because these random encounters are likely to lead to new understanding by the whole group.
- Fourth, look for patterns in the signs. This can be critical in homeland security: you want to look for patterns that emerge — particularly patterns of behavior by people that seem out of the ordinary or very strange. They may be tip-offs about terrorist activity in the neighborhood.
- Finally, as a corollary to the last rule: Pay attention to your neighbors. Again according to Johnson, “Adding ants to the overall system will generate more interactions between neighbors and will consequently enable the colony itself to solve problems and regulate itself more effectively. Without neighboring ants stumbling across one another, colonies would be just a senseless assemblage of individual organisms — a swarm without logic.” The more interactions between neighbors, the more likely we are to be able to respond effectively in a crisis.
However, last time I checked, we weren’t ants.
So what can we humans do, with limited resources and a lot of other responsibilities to juggle, to turn the combination of wireless technologies at our command and the new knowledge about the science of network into a practical, affordable emergency network?
First, given limited resources, it seems to me that it’s crucial that we not set up another, parallel communication system that would be on the shelf, to be activated only in the case of a terror attack. We need a system that operates now, proven well before it is tested in a crisis. A dual-use network that facilitates community organizations, alerts neighbors when there’s a possible kidnapping, helps with weather related emergencies, and extends to online education and delivery of municipal services, will help build community support and involvement now. When a crisis occurs, the network can be converted in an instant to providing emergency information.
Russet Communications of Lawrence, a firm I work with, has created just such a system in Southeastern Mass, the Regional Electronic and Computer Crime Task Force (REACCT). The broadband wireless network will allow 9 communities to link their police and fir departments in one database, and add digital video imaging that can send crime scene video directly to a laptop in a cruiser.
Specifically, we need to extend the MIT Roofnet concept over urban and rural areas as part of the community broadband movement to bring low cost broadband access to these neighborhoods today. Similarly, mesh-based networks for police, fire and EMTs can improve data sharing and collaboration on a daily basis now, then shift instantly to emergency status in a disaster.
We also must create networks of “weak links” and “connectors” now. In a crisis they can serve as modern-day Paul Reveres to link various community groups in response.
We need to create manuals and policies for citizen involvement, seek out potential participants, and train now. Incidentally, one area that I think requires immediate attention is how to handle camera-phone photos from concerned citizens being sent to the police and FBI. On one hand, we need to put in place procedures, including stiff penalties for petty harassment and deception, both to protect the innocent and to avoid authorities being inundated with meaningless photos. On the other hand, we need to educate people about what truly constitutes truly suspicious activity and what kind of photo evidence of it would be helpful. We need to establish procedures to make certain that photos, SMS messages, or other information that concerns life-or-death situations does get to authorities on a real-time basis so they can take preventive action.
Finally, to fully capitalize on the power of personal technology, especially PDAs and smart phones, we need to get essential information deployed in easily-accessible form to those devices now so that it can be retrieved and acted upon immediately in time of crisis. That way, communication can be kept to a minimum, alerting people with only the specifics they need so they can act based on information already in their hands. If I may be permitted one commercial message, after urging such a strategy in a Boston Globe op-ed three years ago and in speeches and receiving a lot of applause but no action, a year ago this month I created the “Terrorism Survival” bundle. It consists of 4 databases summarizing the best available information from around the world in a format that can be accessed in only three clicks, which could be essential in a disaster.
Having deployed the mesh networks, having trained community leaders and “weak links” in advance, like the Minutemen, we[base ‘]ll be able to act quickly and calmly in a disaster. I believe that this strategy would significantly reduce public panic because people would have the information they needed to act without having to wait passively for instructions that may never come. Because of a mesh network’s flexibility and self-healing nature, we would be confident that the communications system would still function. Not only could authorities get essential information to the public, but also those with substantive information that could help in response would be able to contact authorities as well.
When all is said and done, I don’t think we have any alternative but to embrace the smart mob concept in the face of a networked challenge such as terrorism.
The risk of a disenfranchised, uninformed public panicking in a crisis is too great. The challenge of preparing for and/or responding to a terror attack is too expensive, the time is too short, and resources are to limited to ignore the very real advantages that we have now because of the billions of dollars of technology that already are in daily use by average citizens that weren’t available in past challenges.
I was incredibly proud when I read Fischer’s Paul Revere’s Ride to understand for the first time the depth of planning, the confidence in the ability of the common person and the flexibility of response when the battle finally came that marked the Minutemen. Now we run the risk of being the obstinate Regulars marching in lockstep formation while the opportunistic, networked rebels — the terrorists — pick us off.
I hope, for all our sakes, that we will learn from history.
Thank you.
