Social Ruminations

Neighborhood Networks

Not long after I moved into my new apartment, I experienced three separate power outages of significant duration within the span of 24 hours.  Since my iPhone was charged and AT&T's network remained operational, I was able to connect with the outside world and inquire about the situation.  Given I now live in an area that is primed for earthquakes at any time, I've thought about how bad the situation could become if a significant event took place.  Having lived through Hurricane Gloria on Long Island, N.Y. many years ago without power or other types of support for two weeks, I have a clear memory of the implications.  In such situations, you are on your own.  And as we have witnessed from the aftermath of Hurricane Katrina, the situation can turn dire in short order.  

As we face threats to our infrastructure networks, both natural and man-made, I am continually surprised by the government's unwillingness to rethink the role of the average citizen in the security equation.  We continue to retain a centralized mindset with federal, state and local government entities responsible for restoring services and order following a major event.  This approach works to a point.  Given we've developed national infrastructure that is highly optimized yet far from fault-tolerant, impacts on critical nodes in the networks can have disastrous effects.  And it is during these catastrophic failures of the infrastructure networks where the centralized model of disaster response fails.  

Consider one of the central weaknesses that is continually exposed during crises: communication.  I find it ironic that in the age of Web 2.0, we still revert to an age-old model of communication during disasters: the radio.  It's one-way nature is a reflection again of the centralized model.  We use a channel that supports mass information distribution but provides no mechanism for two-way or group communication.  

What would it take to fundamentally change this infrastructure?  In sunny California, I'm imagining an ad-hoc peer-to-peer network that is activated by residents following a disaster event.  With solar infrastructure providing power, a community could stand up an emergency network to support information sharing and coordination of activities.  With a viable communication channel that supports group communication, we can then move to thinking about preparing a community to be active participants in establishing stability and security while government entities attempt to surge resources into an afflicted area. (My network thinking is inspired by the OLPC movement.)  

Clearly the technological dimension to this problem is by far the most straightforward. Changing our culture and mindset may be far more difficult.  We need to get to know our neighbors before a crisis takes place.  We need viable trust networks to fall back on when times are tough.  We need a shared sense of responsibility to rise to the occasion and some notion of our role in the endeavor.  The government can choose to support this change or communities can go it alone.  We hear stories of small communities that do exactly that when times are tough.  Everyone knows everyone and looks out for one another.  That is an ethic we should seek to spread and enhance with technology when appropriate.   

Weaving Trust Networks

Several years ago, I had the opportunity to listen to a retired Marine general who, upon leaving the military, continued to pursue his passion in the area of strategic collaboration.  During the course of his military career, at various levels of command, he had seen and wrestled with the challenges of orchestrating the actions of multiple civilian and military organizations in dynamic, uncertain environments.  During his presentation, his objective was to discuss the wicked hard problems in the extreme scenarios of disaster recovery and reconstruction events such as post-Hurricane Katrina operations and Iraq reconstruction.

In his review of the myriad reasons for the failed response following Katrina, one that stood out was an organizational change that took place prior to the event.  From 1979 to 2003, FEMA stood as an independent agency within the U.S. Government.  In 2003, FEMA became part of the new Department of Homeland Security.  During that transition, priorities changed leaving core missions marginalized in the new organization.  This caused a significant fraction of the senior leadership within the agency to leave.

According to the general, the loss of the senior leadership dealt a significant blow to FEMA's ability to respond to Katrina due to a tremendous loss of social capital.  The senior leaders had accumulated years of experience responding to these types of disasters and knew all the key players in the world of disaster recovery and reconstruction.  They had developed these critical trust relationships over the years through past experiences working together on similar hard problems.  All of this was lost in short order.

A question that came up after the presentation was the following.  What options, if any, exist for accelerating the development of trust relationships in such scenarios when you have players from different organizations working together for the first time?  At that time, we envisioned an event similar in concept to speed dating that is mediated by technology.  Several companies have developed devices (such as SpotMe and nTag) for conferences to aid in networking through the use of profiles and location awareness.  In a disaster recovery setting, the goal would be to use technology to mediate introductions between the right individuals that need awareness of one another to support overall mission objectives.

If organization dynamics were driven solely by mission and function, this might provide additional value.  Yet one important dimension is missing from the equation.  The development of strong bonds between people is primarily driven by compatibility in terms of personality and objectives.  The relationship chemistry is not a factor that can be captured in a profile.  Yet the determination of compatibility a priori can have a significant impact on the effectiveness of an organization that one is actively constructing.  Our goal is to weave trust relationships in such a way that our formal and informal networks are better aligned, leading to positive outcomes in terms of the overall mission.

To improve the effectiveness of mobile devices for network weaving, unconscious or honest social signals must be sensed and processed to assess compatibility.  As Sandy Pentland's work in recent years has shown, these signals can be effectively measured and used to predict the outcomes of social interactions in a range of scenarios.  According to Pentland, such signals transcend culture and language which is critical in international development contexts. 

By profiling individuals both in terms of their organizational role and patterns of social signaling, the goal would be to utilize devices to mediate introductions between people that both need and hopefully want to work together.  Taking into account any characterization of social signaling early on could potentially prevent significant breakdowns in communication later on when stakes are high.  We want to build effective organizations rapidly by accelerating the discovery of valuable social connections.  Today that search is far more time consuming and fortuitous.  In reconstruction and recovery events, where time is of the essence, any improvement in efficiency of networking to construct effective organizations is significant.

Honest Signals

Each one of us has had the experience.  We meet someone for the first time and immediately feel a resonance.  We talk about how some people simply radiate, creating a presence that is hard to ignore.  The nonverbal communication channel is powerful and shapes outcomes in social situations in profound ways.  In recent years, Sandy Pentland's research group at the MIT Media Lab has been systematically studying this communication pathway and characterizing more precisely the impact honest signals have.  Such honest signals are described by Pentland as "signals... that are either so costly to make or so difficult to suppress that they are reliable in signaling intention." 

In his recent book entitled Honest Signals: How They Shape Our World, Pentland makes the distinction between conscious and unconscious signals.  Conscious social signals are planned signals such as facial expressions and fashion choices that are intended to signal a particular message and therefore can be misleading in terms of intention.  Pentland claims that one should focus rather on signals that are processed unconsciously, and therefore are uncontrollable, in order to identify signals that are honest in intention. 

Four examples from conversational turn-taking and gesturing referenced by Pentland are

• Influence - Degree to which one person causes another person to match their own pattern of speaking
• Mimicry - Reflexive copying of one person by another during a conversation resulting in the exchange of smiles, head nods and interjections
• Activity - Increased activity levels signal interest and excitement
• Consistency - Consistency of speech and movement is a signal of mental focus, while greater variability may signal greater openness to influence

In recent years, Pentland's group has built wearable devices they call sociometers to unobtrusively measure these honest signals in a wide variety of social settings.  Based on these measurements, they are able to reliably predict social outcomes in a range of scenarios. 

For those developing technologies to aid communication and collaboration, this book provides a number of ideas to ponder.  In future posts, I'll explore some of these ideas in more detail. 

Mobile Phone Entrepreneurship in the Developing World

This morning, I've been reading about an exciting ongoing project called EPROM (Entrepreneurial Programming and Research on Mobiles) that is focused on fostering mobile-phone related research and entrepreneurship in the developing world.  The EPROM project is lead by Nathan Eagle, whose seminal Ph.D. research at the MIT Media Laboratory focused on understanding complex human dynamics from social interaction data derived from mobile devices. 

To fully appreciate the motivation for EPROM, it is important to consider the trends within Africa.  The following statistics are from the EPROM site.

2006 Statistics

• Worldwide, there are more than 2.4 billion cellphone users, with more than 1,000 new customers added every minute.
• 59 percent of these 2.4 billion people live in developing countries, making cellphones the first telecommunications technology in history to have more users there than in the developed world.
• Mobile phone shipments grew 19 percent to 810 million units in 2005 and are expected to rise by 15 percent to 930 million units in 2006
• Cellphone usage in Africa is growing almost twice as fast as any other region and jumped from 63 million users two years ago to 152 million today.
• Kenya is currently the fastest growing mobile phone market in the world. In June of 1999, Kenya had 15,000 mobile phone subscribers. By the end of 2004 the country had 3.4 million subscribers, and in the last 18 months this number has grown to over 5.6 million, despite the fact that only 200,000 Kenyan households have electricity.
• It has been shown that adding an additional ten mobile phones per 100 people boosts a typical developing country’s GDP growth by 0.6 percent.
• Access to market information through mobile phones provides rural communities with invaluable information about centers of business; many African fishermen check the local fish market prices on their phones to determine where to bring the day’s catch. The Kenya Agricultural Commodity Exchange (Kace), now provides crop growers with up-to-date commodity information via text message (sms). This allows farmers to access daily fruit and vegetable prices from a dozen markets, and many have quadrupled their earnings because they have access to information about potential buyers and prices before making the often arduous journey into urban centers to sell their produce.

Such statistics clearly indicate the power provided by communications technologies that support efficient information exchange and market formation.  One particular example of this within the EPROM project is txteagle, a mobile crowdsourcing application that allows individuals to make small amounts of money by completing simple text-based tasks via text messaging. 

EPROM has a strong empowerment agenda as well that is educating future African entrepreneurs on mobile phone application development technologies through the Mobile Phone Programming Curriculum and the SMS Boot Camp.  It will be fascinating to watch how these empowerment activities fuel the development of new mobile social software, further supporting economic development. 

Empowering Citizens

As Hurricane Gustav bears down on Louisiana, I can not help but think about the challenges of disaster response and how I believe we as a nation can do better.  I believe communication technology, including social media, has an important role in this equation.  Yet the most fundamental shift that is required is one of mindset.

Following 9/11, the push for homeland security has focused significantly on deploying technology to enhance the capabilities of security personnel and first responders to prevent or mitigate the impact of threats to the nation, whether natural or man-made.  Within this equation, one hardly hears discussion about the role of the individual citizen.  Therefore the model appears to be one that views the citizen more as the victim rather than an enabler.

Consider as an example what transpired between the 1993 and 2001 attacks on the World Trade Center.  Significant investments were made to enhance the physical security around the towers.  Yet, as described in Amanda Ripley's The Unthinkable, there was significant disarray during the 2001 evacuation due to simple lack of knowledge and preparedness.  No full-scale evacuation drills had been conducted in the towers since the 1993 attack.  Many had never taken the stairs in the towers and had no idea where the staircases were.  The organizational strategy that had been discussed for such events had not been seriously rehearsed.  It assumed that centralized communication with first responders would be in place to coordinate and guide the effort.  When this model was put to the test, it broke down and resulted in loss of life. 

Empowering individuals with information so they can act, either individually or collectively, to survive is key.  It is a mindset that we must emphasize and support through social and technological means.

Dr. Stephen Flynn, Commander, USCG (Retired) is a leading expert on homeland security at the Council of Foreign Relations and a staunch advocate of this mindset.  The example he points to about the impact of empowering individuals is Flight 93. In this scenario, the passengers on that aircraft obtained information about the nature of the situation, deliberated their options and took action.  They undoubtedly saved additional lives through their actions and mitigated the threat before government officials were faced with far more difficult choices. 

Empowering individuals with information can have significant effect in larger scale disasters as well.  If we consider our citizens as potential actors in the equation, actors that may be the first responders when the professionals are unavailable, then we need to think differently about our priorities and needs.  Communication infrastructure is key.  We must be able to rapidly deploy infrastructure that can reestablish a communication network within the affected region.  We also need to think about how individuals will access that network when power may be limited to nonexistent. 

Once the network is operational, there needs to be communication capabilities in place that support emergent collaboration.  Current social media tools provide capabilities that can help and hint at elements that are needed.  One important capability that I believe is absolutely essential is for citizens to have some mechanism to easily integrate their collective observations into an overall picture of the state of the environment.  Without the input of the citizens, it may be difficult to impossible to produce equivalent situational awareness through other means.  (Consider the aggregation of information on the San Diego fires via Google Maps as a recent example.)  The citizens form a distributed sensor network of mass proportion, provided they can communicate and share state with others.  With a view of the global state, the citizens themselves can then start to make better decisions and assess the impact of their actions on the global state. 

Moving beyond simple aggregation of state, the next step may be to provide mechanisms for individuals to coordinate locally and report their actions and intent.  In such dynamic and fluid events, having the tools to enable emergent collaboration and action could be a significant force-multiplier in time of need.  Here the simple act of reporting actions could help better apply the resources available through the social network that is the community.  I imagine significant questions loom here about how and what to share and how that ultimately supports emergent collaboration.  All interesting dimensions for future research.

I hope that in time we will shift our mindset on the role of the citizen and begin evolving our technology to support the paradigm of emergent collaboration.  It must be a socio-technical evolution.  The past is rife with examples illustrating that technology alone is not the answer.  It can often be the easiest part of the equation to handle.      

UPDATE: Here are some interesting links related to the post.

InSTEDD - "An innovation lab for technologies designed to improve community resilience and save lives through early disease detection and rapid disaster response."

• Contacts Nearby - "Allows users to discover geographic locations of other people in their extended social network."

Strong Angel - A serious of demonstrations exploring ways for improved civilian-military cooperation in disaster scenarios.

Strong Angel III Interim Working Report - Covering lessons learned from the most recent report.

• Second Life was used "to demonstrate a custom-built virtual training and learning environment for the Strong Angel community."
• "Dan Gillmor (formerly technology editor for the San Jose Mercury News—the newspaper for Silicon Valley) and David Thorpe (formerly of Random House and Disney, now with Young and Rubicam) developed systems and tools designed to increase the information flow between a stricken population and the central authority responsible for their care. It is a carefully bi-directional system, striving to reduce rumor and increase effective response, actively engaging the population in the process of their own disaster relief."
• "Dr. Ashraf Ghani and Clare Lockhart J.D. observed how the Ghani-Lockhart Framework can be used not only as a diagnostic and planning tool for leadership, but also as a communications and social networking tool. In doing so, Ghani and Lockhart were able to partner with the Center for Citizen Media and Internews to form an ad hoc communications working group focused on Objective #18 “Inform Everyone of Everything Important.” Their on-site discussions of social networking enabled them to develop general ideas and methods for community involvement and collaboration between various stakeholders during initial relief assessments."

It's All About Me

One idea in Chris' entry about Social Transparency and Latent Group Discovery led me to wonder about Facebook's broadcast of my activities.

The propagation of protests against the FARC in Columbia is an example of: "This is important to me, and I want you to know about it so that it is important to you too."

I am intrigued in a different mode, in which I post (or Facebook reveals what I have just done) something that may be only about myself, and I have no intention of influencing my network. (It's all about me.)

For a lot of what I say (or do) on Facebook, whether I want to admit it or not, I want my network to think positively of my actions. I want to be liked. (Who doesn't? It's all about me, right?) I am doing something not because I think my actions and words lead to the betterment of my group/network, not because I think the world would become a better place. I do many things (on Facebook) to feel good about myself.

Yes, telling the world about my up-to-the-minute activities is a way to be connected to my friends and families. It's easier for me to post about my day than for me to call my parents and siblings to chat.

When I post a status, "Jaime just saw the greatest bestest slasher movie," I want my network to think that I am the coolest guy. I am also recommending this movie. But, do I care that my network goes and see the movie, and love the movie so much that they too create a status update so that their own networks will all go see the movie? Naw.

On the other hand, a status, "Jaime is in pain again from eating a piece of bread" has an obvious secondary message. I am telling you about my pain so that you would ask me how a piece of bread could cause pain, so that I could tell you about gluten sensitivity.

Facebook makes it easy for me to feel good about myself. Oh, by the way, the same mechanism helps me stay connected to my friends and family, without having to write email (too much typing) or call (too much commitment). And guess what, if I feel passionately about something that I want to share with the world, I don't have to do anything differently. I don't have to do "more." All that I have to do is to be selfish, write about myself, and the world (well, first, just my friends) may discover that it is passionate about the same thing too.

Social Transparency and Latent Group Discovery

In March 2008, I participated in the AAAI Spring Symposium on Social Information Processing organized by Kristina Lerman (USC Information Sciences Institute), David Gutelius (SocialKinetics), Bernardo Huberman (HP Labs) and Srujana Merugu (Yahoo).  The event brought together a very interesting mix of researchers investigating social media and the dynamic, emergent communities that are enabled by these technologies.  Among the host of intriguing presentations, one in particular that caught my attention was given by Ed Chi from PARC. 

Ed leads the Augmented Social Cognition Research Group which is focused on systems that "enhance the ability of a group to remember, think and reason; the system-supported construction of knowledge structures by a group."  Ed's talk focused on insights derived from their investigation of the Wikipedia community and on a specific software component they've developed called WikiDashboard to enhance social transparency with the community.

The notion of increasing social transparency in the Wikipedia community amounts to providing greater visibility of community member actions in creating and editing the collective knowledge store.  By allowing anyone to easily view the actions taken by others in the community, the assertion is that this will result in an increase in credibility and trust in both the content and among the members of the community.

While listening to Ed's talk, I thought about the broader implications of social transparency and how social media technologies provide opportunities that lead to significant outcomes.  More and more we are witnessing examples of collective action in the real world that emerge from movements that develop initially online.  Take as a notable example the recent worldwide protests against the FARC in Columbia.  It all started from a small group of young people expressing their outrage on Facebook over using kidnapping as a weapon.  Little did they know their message would lead to a movement culminating in synchronized protests in 185 cities worldwide.

By increasing social transparency, we aid the process of latent group discovery, the discovery of hidden groups that share one or more attributes in common.  Facebook groups provide one mechanism for latent group discovery.  They allow individuals to form a group construct on Facebook and easily broadcast that construct to their social networks.  Whether or not it is successful in exposing a larger latent group is dependent mainly on how the group idea resonates and therefore propagates.  Many questions linger here about how various types of social media support latent group discovery which we will examine in the future.

What appeals to me about the pursuit of social transparency is its potential role in combating extremist ideologies and enhancing freedom.  Without a doubt, social media is being used already along these lines.  The Iranian blogosphere, for example, is providing a significant outlet for political expression and organization that is otherwise infeasible.  Yet I believe we have only seen the beginning. 

Latent group discovery is far from systematic in my opinion.  Technologies will emerge that aid latent group discovery and coordination in more dynamic settings such as disaster relief.  Such technologies will also enable greater political freedom I believe by allowing groups of individuals to quickly discover their shared outlooks and channel their collective energies toward positive change.  Consider the potential for such technologies in the Middle East.  Imagine the impact of an emergent group expressing their outrage swiftly online against extremist media messages from groups such as Al Qaeda.  Empowerment begins with group awareness.  And awareness can quickly lead to action.      

 

Crowdsourcing Cyberwar

On April 27, 2007, the Estonian government moved a controversial bronze statue of a Soviet soldier from its long-time resting place to a military graveyard outside the capital city of Tallinn.  The statue, which serves as a memorial to the fallen Soviet soldiers that battled the Nazis in World War II, has long served as a point of tension between the majority Estonian and minority Russian populations within Estonia.  For the Russians, the statue is a symbol of sacrifice and struggle.  For the Estonians, the statue represents occupation and oppression.  After 15 years of independence from the former Soviet Union, the government finally decided to take action to place this troubling symbol in a more remote location.  The reaction from the Russian minority was swift, with rioting and looting of downtown Tallinn that lasted for days.  Along with chaos in the streets came the beginning of a protracted cyberriot known now as the first Internet war.

For Estonia, the cyber-campaign brought negative consequences on a national scale.  Having rapidly accelerated into the online age, Estonians leverage the Internet in every aspect of life.  Therefore widespread denial-of-service attacks had significant implications.  The nature of the attacks at first appeared familiar but quickly morphed into a new phenomenon.  In contrast to cyberattacks of old, consisting of waves of botnets, this conflict exploited the anger of the Russian minority to exact revenge online at scale. 

At the start of the conflict, posts appeared in Russian-language websites, discussion groups and blogs about the recent events and ways to respond.  With a combination of hate speech and instructions on how to launch a denial-of-service attack, average citizens were called to the battle and provided cyberweapons to participate.  As the campaign continued, fresh posts in the blogosphere provided updates allowing the masses to adapt as the defenses attempted to respond to the attacks. 

The Estonian Internet war provides the first example of a new type of warfare that leverages social media to exploit societal tensions and dramatically reduce the barriers to participation in conflict online.  With the assistance and guidance of likely a small number of skilled information technology specialists, cyberweapons were crafted and distributed online to anyone that wanted to participate.  They were crafted such that no special skills or resources were required to operate them.  With access to the Internet, anyone could join in the fight and express their outrage in a significant way with limited risk.  It is a masterful use of knowledge of the human terrain and the technological landscape in concert to exert significant effect.

Without a doubt, this will certainly not be the last time we witness this type of conflict.  With Russia's recent incursion into Georgia, a cyber-campaign was launched in concert with ground operations.  Although the Russian government once again denies involvement in the cyber-campaign, similar activities are being witnessed online.  Online forums are once again distributing scripts for executing denial-of-service attacks that have targeting lists embedded within.  Once again, a few with the necessary technical literacy are significantly lowering the barrier to participation so that the masses can express themselves through their actions online.