A recent article that I came upon in the New York Times that I thought is relevant for our class was about social networking moving to cell phones leaving computers. I think this is rather prevalent in social networks as they become more and more dynamic. As technology changes networking evolves, networks naturally develop into new forms.

I think that with the oncoming of mobile cell phone networks, and mobile social networking, information cascades will occur more and more through technology through different locations, rather than through word of mouth. In addition as mentioned in the network the anonymity of the network and the diversity of the individuals will lead to more random cascades. In addition the possibility of them ending becomes much larger as more information from different sources becomes available.

Social Networking Leaves Confines of the Computer

According to Heider’s social balance theoryⁱ, individuals (nodes) in a social network tend to form relationships (+ or - edges between nodes) as follows: “for every set of three nodes, if we consider the three edges connecting them, either all three of these edges are labeled + (triadic closure), or else exactly one of them is labeled +.”ⁱⁱ One can prove mathematically that a social network with this property requires it to have “either all nodes are friends, or, more commonly, the nodes are divided into two groups” where the people within a group all like each other and everyone in a group dislikes everyone in the other group. This type of dichotomy in a network can be seen in real life, one famous example being the college karate club observed by W.W. Zachery, where a dispute broke out between the club’s administrator and its teacher, eventually resulting in the club splitting into two clubs. The drawback of social balance theory is that it requires everyone to know and have a relationship (+ or -) with everyone in the network, and that triadic closure is strictly enforced. In a large social network, it is common to find two people who, despite having mutual friends, simply don’t know each other. It is also not uncommon for a person to have two friends who don’t like each other, or for three people to all dislike each other, despite the social dissonance in these situations. An intuition notion that social balance theory does not provide a good model for large networks is that they tend to be composed of many sub-communities, not just one or two.

If it is indeed true that a social network consists of communities, then it is natural to try to identify them. This allows us to have a better understanding of the structure of a network on a global scale. Finding clusters in a collaboration network between physicists, for example, allows us to see the level of research activity of different branches of physics.ⁱⁱⁱ Another use of finding communities is to extract smaller networks from a large one, where the smaller networks might be much easier to analyze and their interactions might actually behave more accordingly to social balance theory. Consider the Zachery’s karate club example: if the researchers were collecting social network data on the entire university instead, then the dynamics of the club would have been lost the in complexity and global dynamics of the entire network. A good clustering algorithm would enable them to isolate the club and study the effects of the dispute in the club. Thus, finding communities allows us to gather data on a large network on a global scale while still be able to look at its sub-clusters and study the local dynamics of individual clusters.

iFritz Heider. Attitudes and cognitive organization. Journal of Psychology, 21:107–112, 1946.

iiStructural Balance. ECON 204 handout. http://www.infosci.cornell.edu/courses/info204/2007sp/balance.pdf

iiiM.E.J. Newman and M. Girvan. Finding and evaluating community structure in networks.

Just a few days ago, Dell announced an initiative to offer Ubuntu Linux on a line of its desktop and notebook computers. This has a broad range of implications for the future of Linux. While driver support is sometimes hit or miss in the Linux world, hardware should be fully supported on these pre-installed machines. Further, Dell’s endorsement of Ubuntu will pressure other companies that offer peripherals and accessories to at least investigate the possibility of supporting their devices on the OS. In the same way that the rise of Windows in its early days caused software companies to write their software for the fledgling OS, this reinforcing mechanism will no doubt find its way into the emergence of Linux.

Munich, Germany, made the dicision to make the switch from Microsoft to Linux back in 2003. Now that Dell is offering Ubuntu, an interesting question in regard to diffusion in networks and information cascades arises. Consider a government employee in Munich who must learn a distribution of Linux for his work. If he has a favorable experience with the OS, would that be enough for him to choose Linux rather than Windows for home use? Clearly, this may be at least part of the reason why Microsoft has been offering aggressive discounts in an attempt to retain its government and corporate customers.

We’ve discussed how situations involving network effects are often not socially optimal, as well as the possibility that such a product may not, in fact, be superior than its competitors. If Ubuntu is indeed a superior product for certain needs, could we be at a crossroads in the future of operating systems? Despite the outpouring of support for Dell in offering Ubuntu, consumers will ultimately vote with their pocketbooks, and only time will tell if Linux distros such as Ubuntu are successful in the mainstream.

Flash mobs are a “large group of people who assemble suddenly in a public place, do something unusual for a brief period of time, then quickly disperse.” These groups are also called “inexplicable mobs”, “smart mobs” or “flash groups”, to name a few. Flash mobs carry out carefully organized and orchestrated “missions”, the first (successful) of which was carried out on June 3, 2003 when more than one hundred people went into the rug department of Macy’s in Manhattan and claimed to want to purchase a “Love Rug”. Groups such as “Improv Everywhere” carry out missions on a somewhat regular basis, stating “We’re big believers in ‘organized fun’. In the process we bring excitement to otherwise unexciting locales and give strangers a story they can tell for the rest of their lives.”

Flash mob missions are a means of uniting an essentially random group of people around a common goal.

Those moments of unity are important, say flash mob enthusiasts - especially for isolated urbanites. But because most mobbers still hold fast to privacy, caring little to meet strangers, many come and leave with friends. “Flash mob is a group event,” [Clay] Shirky [professor of interactive telecommunications at New York University] says, “but it’s not necessarily a social event.

These networks are not necessarily based on commonalities between their members, but rather for the sole purpose of carrying out a particular “mission”. They speak to the individualist nature of today’s society while still maintaining some level of community. Although these people gather for a common outing, they are no more linked than if they were everyday people walking in the same direction on the New York City sidewalks. Due to the number of participants and the speed with which the missions are carried out, mobbers most often leave with the same relationships they had when they arrived.

One example lies in the Critical Mass bike riding events that occur on the last Friday of every month in cities around the world. One can infer that their aggregated decision, and thus their “wisdom” is now their mantra: “We Are Traffic”. This speaks volumes about the future of crowd dynamics wherein a random group of virtually disconnected nodes (people) can get together and suddenly alter the dynamic of a given locale or situation. These disconnected nodes become the “many” and therefore gain power despite their lack of specific purpose.

In the end, flash mobs such as Critical Mass and the Macy’s Rug Mission wield great power. They organize and execute missions that can be, and are, interpreted as highly political and thus dangerous to the establishment. By bridging gaps between networks, even if temporarily, these mobs have the power that can be used for good or for evil. They show us that completely random groups of people, connected only by the coincidence of their participation, can have a great affect on their surroundings. These “many” have everything a group needs for an advantage over the existing “many”: size, an environment of acceptance, an individualistic, yet collective nature, and finally, the power of all of their individual knowledge. They are the ultimate Wise Crowd.

Nowak and Krakauer (1999) apply evolutionary game theory (EGT) to the development of a proto-language: “The evolution of language“.

The necessary components of an evolutionary game theory model (selection, variation, and replication) have linguistic analogs:

Selection: the process of acquiring new linguistic constructs, usually based on the tension between accurate, facile communication on the one hand and the minimization of complexity and ambiguity on the other.

Variation: linguistic creativity, reanalysis, or contact with other languages and dialects.

Replication: imitator rather than genetic replicator dynamics - offspring learn language techniques not from random adults but from parents.

Of course, there exists the assumption that successful communication (hearer recovers the intended meaning from speakers’ message) directly influences biological fitness.

The researchers’ investigation reveals three broad steps:

(1) Partners use rudimentary signals (such as a simple vowel-like sound) for each concept communicated. As the number of concepts increases, it becomes difficult for the hearer to correctly identify the signal the speaker is sending.

(2) The number of signals therefore reaches a limit; to achieve a greater lexicon, the signals are joined together into short sequences or words.

(3) Eventually, it becomes impossible to remember enough words to describe all actors, events, places, and more; rather than have a word to represent every event (such as ‘dogbitboy’), compositional syntax develops. Each word describes a part of an event, and each word can be used in combination with others to describe innumerable and complex events (’dog bit boy’).

EGT as applied to language development in this way can be applied to innumerable models. Consider, as a trivial example, that young people are in a perpetual state of self-evolution, constantly struggling to survive as members of one in-group or another. They reinvent themselves constantly, evolving, as it were, from day to day. The ability of a middle school pre-teen to acquire the particular linguistic forms (IM-speak, for example) selected for by the cliques, and to innovate new ‘cool’ wordplays give her a tremendous advantage, a much-improved biological fitness. Mastery of these forms will result in consistent social status stabilized social networks (evolutionary stability).

New York Times article about new 2008 presidential primary schedule: http://www.nytimes.com/2007/05/04/us/politics/04florida.html?_r=1&ref=us&oref=slogin

Before election season even begins, why do prospective presidential nominees spend so much time in Iowa and New Hampshire? What is so important about Iowa and New Hampshire that causes politicians to flock there months (now, years) before an election is even on the horizon? As any savvy politician knows, these days, it is all about the early Iowa caucus and slightly later New Hampshire primary. Conventional wisdom holds that without a victory, or at least a strong showing, in these two early races, a candidacy might as well pull out of the race. With a strong victory in Iowa and New Hampshire, it is thought a candidate can ride on his momentum straight through to the nomination. The idea of momentum in presidential primary elections is remarkably close to the idea of an information cascade. Are the outcomes of presidential primary elections simply an information cascade set off by the Iowa caucus and New Hampshire primary? Indeed, whoever wins the Iowa caucus and New Hampshire primary, the first two party contests in the country, has the momentum necessary to start an information cascade. This appears to have happened in the 2004 Democratic presidential primary season. Could the idea of an information cascade be used to generally predict the pattern of outcomes in the 2008 primary season? The only way that primaries could function as information cascades is if they occur sequentially. In the upcoming 2008 primary season, the traditional schedule of sequential primaries has changed. Over twenty states have moved up their primary dates to the first week of February, almost immediately after the Iowa caucus and New Hampshire primary. States are vying to be first to start the information cascade. However, an unintended consequence is that other states are moving up their primaries as well to the point that most states will now be having their primaries on the same day. Super Tuesday this year will be the biggest and earliest it has ever been, with more delegate heavy states getting in early to try to have a greater effect on the outcome. Thus, candidates will not be able to build as much momentum into a cascade because voters in many different states will be voting simultaneously and thus cannot take cues from each other. I would expect, then, that people will vote more in accordance with their personal calculations and values as opposed to simply jumping on the bandwagon, as appears to have been the case in the 2004 presidential primary.

http://plato.stanford.edu/entries/game-evolutionary/

In this article, found at the Stanford Encyclopedia of Philosophy, the topics of evolutionary game theory are discussed. In addition to the Hawk-Dove game that we learned about in class, they use a model with the prisoner’s dilemma. This game is pat of a dynamic model of how the population can evolve over time. Depending on the values of the payoff matrix, the only stable equilibrium can be either for everyone to defect or, on the other hand, for everyone to cooperate. There are also combinations that result in chaotic or undulating systems which never reach one particular equilibrium.

What is especially interesting about this entry of the encyclopedia is how it approaches the notion of evolutionary game theory and social sciences. In particular, the meaning of adaptation in culture is questioned. One major problem arises from the fact that game theory was designed to apply to economic situations with in which rationality can be easily defined as economic utility. Evolutionary game theory, however, attempts to apply this same framework to a qualitatively different situation, namely the selection of irrational actors whose success is based on some loosely defined payoff.

Although some results reached by the study evolutionary game theory are compelling, this does not mean that they are necessarily true. For instance, recall the Schelling model of segregation, in which only a slight preference for homphily was required to lead to clusters of individuals of different characteristics. In this case, we learned that a posteriori reasoning can sometimes be misleading; that sometimes slight individual preferences can lead to unexpected, unintended, or even undesirable macroscopic effects. Another thing that should be taken into account when thinking about evolutionary game theory when applied to human culture can be distilled from the notion of information cascades. An information cascade implies that we, as humans, are influenced by the decisions that we see others make, at least to some extent; network externalities, specifically the impact on one by one’s friends, are also important here. To define a notion of advantageous in this environment would be very difficult, without first determining a primacy to either autonomy or group behavior. This disjunction between evolutionary game theory as a way of modeling the expansion, growth, and death of species, and its applications to human culture are a product of broken assumptions. Only a drastic generalization of evolutionary game theory can overcome this theoretical hurdle, if this theory is to ever explain, or merely to describe, culture.

In his paper ” On the convergence of informational cascades” (http://www.sciencedirect.com/science/article/B6WJ3-45P12S4-N/2/8a374f4e23f6ea6d04deb1891875e7c1) , In Ho Lee discusses the formation of informational cascades when the action space is continuous.

In the basic model of Bikhchandani, Hirshleifer and Welch, each person makes a binary “Accept/Reject” action after considering his own signal and the actions of his predecessors. It turns out that the binary action space is a necessary ingredient for an informational cascade to take place. This is because each person’s action may not be a reflection of his private signal as he may have acted based on the actions of those before him. As a result, information, in the form of valuable private signals, is lost each time a person chooses to act against his own signal.

This loss of meaningful information can be avoided if we allow a continuous action space. Each individual observes the actions of those before him and chooses an action that has a value between that and his own. In other words, he modifies the predecessors’ actions slightly according to his own judgement. His action thus encapsulates some information about his private signal. Thus, the next person can make a better decision, preventing a cascade.

A real-life example of a continuous action space is found in the carnival games in which participants are asked to guess the number of marbles contained in a large glass jar, which is anywhere from 0 to 200. Participants make their guesses sequentially and write them down on a white board, or something to that effect. Each person makes a private guess and then updates his guess by finding the mean of the sum of all the previous guesses and his own. As a result, he fine-tunes the most recent guess according to his own estimate. Assuming most people make educated guesses, the guesses will tend to the actual number when the number of guesses is large. This works because, intuitively, for every one guess that is less than the actual number, there is another one more than the actual number by the same amount. Finding the mean just cancels them out in the same way we eliminate random error in experiments. An informational cascade never forms in this senario because the continuous reporting preserves information about each person’s private signal.

In the Apr 17, 2007 issue of Fink Again former Editor-in-Chief of The Cornell Daily Sun Erica Fink analyzes the various ways Facebook has been used outside of its definition as an online medium for social interaction. The piece culminates with a look into how Facebook as a social network has turned into a unifying force for college students. In terms of the network studies presented in class, the analysis provided in this opinion piece suggests Facebook as a catalyst of creating random links. 

The conceptual background presented in class revolves around the concept of ”six degrees of freedom.” Six degrees of freedom highlights the idea of a small world by postulating that every person in the world is connected through a network of mutual acquaintance and that on average the number of edges required to link two people is six. The Watts-Strogatz model for this concept constructs a random graph for which every node has r^2+k outlinks, r^2 representing links to neighboring nodes, and k random links to nodes of arbitrary length away. The ability for two nodes to be connected through such a small number of edges relies on random links being able to reach out arbitrary distances, and overall the number of links required to connect any two nodes as a function of the network size runs on a logarithmic relationship.

Traditionally the concept of connectedness is viewed more in terms of direct links, the r^2 number from the previous example. However, from the context provided by six degrees of freedom, a new definition of connectedness can be formed through an analysis of how random links are distributed in a network. In this sense, connectedness is a measure of how well connected two distant nodes (of which “distant” is a qualitative measure) are. When the number of random links hits a certain threshold, they can no longer be viewed as random and the definition of connectedness returns to its original definition. However, for a network of considerable size, say college students residing in the United States of America, the likelihood of this happening is negligible. So for a set number (that is lower than the threshold) of random links with a given distribution, connectedness of a network can be measured.

The opinion piece demonstrates that Facebook acts as a catalyst for creating random links and thus increasing the connectedness of a network. The example of the Virgina Tech shootings is obviously a little hyphenated. Even without Facebook, such a catastrophic event broadcasted over all major news networks would evoke a strong response (connection) to those who were affected. However, the prescence of Facebook, particularly the creation of support groups, presents an example of how random links are created. Facebook users who may never have heard of VT may have formed a significant link to those who were affected. Those at VT, the victims displayed on TV, or the faceless others who survived but were scarred nonetheless, become more than just a statistic, but an emotional commitment that in terms of network theory, constitutes a random (though temporary) link. In this manner, Facebook may indeed be responsible for making the world a smaller place.

As previously mentioned by jmholloway, the spread of the HDDVD decryption key on Digg forced the site to a crucial decision.

To briefly explain, the key to decrypt the AACS encryption on HDDVDs spread on a myriad of websites, including Digg. The hexadecimal code was posted in news articles and comments throughout Digg, and each posting was “dugg” up, ensuring visibility based on Digg’s popular-voting system. Following cease-and-desist orders from the AACS Licensing Administrator, Digg removed the posts as best it could. The community responded with deluging the site with even more posts and demanding an end to the censorship. Site founder Kevin Rose then made the risky decision to side with his user base against a vastly powerful media industry.

As Ryan Singel of Wired writes, Digg is now the center for the fight over this hexadecimal number – and possibly all of DRM. By siding with his users and not censoring the number on Digg, Rose has invited the wrath of the media industry.

This whole case is an interesting example of user power in a social network and the power of information cascades. The speed with which the code spread both within and outside of Digg is a testament to how fast a cascade can work. This power surely leads to certain network dynamics that can be unfavorable. Although Kevin Rose still maintains the monitoring rights on Digg, it is clear that the users call the shots. The community seized a great amount of control over Digg this week, a move that puts the site in peril but surely adds worries for those trying to stop the flow of sensitive information.

It will be interesting to see if charges against Digg of trafficking in circumvention technology materialize, as they could cost Digg between $200 and $2000 per instance. One thing is for sure: Digg is at a crossroad, either it will die from its user’s posts or grow stronger from its bold stand. Like the Spartans in 300, Digg has taken a stand - could this be its last?