Right upper corner or left lower corner? I sit here tapping my foot anxiously, trying to decide which square to choose on Facebook’s campusfood.com Feeding Friendzy. And there it is: that stupid little Zelda elf with the evil laugh. Game Over. I am sure many others have faced the extreme annoyance of trying to win money for campusfood.com on Facebook’s Application. Not too long ago, Facebook created a platform on their site to allow companies to create an “application” which Facebook users can utilize featuring games, music, and other sorts of entertainment. What makes this unique from other traditional ads is that Facebook users can choose whether or not to add this so called “app” and how much information to share with it, whether it is just registering with it for a quick use or putting it on your profile so other people can see it. Many of these applications also allow the user to invite their friends to participate in them. This social networking has generated great publicity for many companies.

In this article by CNet, many emerging companies and websites gathered at a conference for the same purpose: to create an application on Facebook. Companies just starting out, such as the website iLike which features a music discovery tool that allows friends to recommend different songs to each other have had skyrocketing traffic flow to their websites due to the use of the application by many people. While the first thing that comes to mind when we think of Facebook is a social network, Facebook is also an information network. In class we talked about how WebPages are seen as the nodes and links to other WebPages are the edges of an information network on the web. Here Facebook and its applications can be seen as a strongly connected component, where Facebook has an edge to all of its applications’ websites and each website has a link back to its application on Facebook. Each edge is a directed component because each edge only flows in one direction, with a separate edge to get back to the beginning. Example: On the Pandora Radio Facebook application (node), there is a link to pandora.com (edge). If you click on it, it will direct you to pandora.com (node) where you can find an advertisement on the right hand side of the radio to link your pandora.com music with Facebook (edge), directing you back to Facebook (starting node).

Another interesting feature about the Facebook applications as it related to our Networks Class can be seen in the campusfood.com application. This app lets you choose friends to play with (although it is not a competition), and if you win money on the game, they also win the same amount of money. There is also the chance to win a highly valued Wii video game console. Here it is interesting to analyze the payoffs of the game. How should I choose which person to play with? There are two options here: playing with a good friend or a casual acquaintance. If I play with a good friend and win money, then they also win money. This is not only beneficial to you by giving you the payoff of winning money, but also could increase your payoff by having your friend value you higher because you won them money or if you friend decides to be generous and use some of their money to buy you food. Here the chance of your friend buying you food is higher than the chance of just an acquaintance buying you food and the friend will appreciate you winning them money more than an acquaintance who probably doesn’t even remember your name. Also, since you won money for your friend, they will be more inclined to play with you and win you money in return. Since you can only play each friend a week, and play the game only five times a day, to maximize your payoff the best strategy would be best to play your good friends first and then work your way towards playing casual acquaintances. So the next time you are trying to decide which friend to play on Food Friendzy, remember to play your friends first.

While the amount of invitations you receive a week to add different applications on Facebook can seem outrageous and slightly annoying, remember that some of them are worth checking out. Not only could it lead you to a new website that could open you up to a new experience, but you and your friends may come into some money in the process.

Some people are so careful about avoiding sickness while others simply take it as a fact of life. It’s no wonder the flu spreads so easily when things like hand holding, plate sharing, and simply passing pens around can so easily spread germs. I find that many of us also hold a personal bias- when another person is sick, we hope that they carefully release each sneeze into a tissue and follow through with an elaborate and immaculate hand washing ritual - but, when the tables turn, a sneeze is just another sneeze, and hand washing would be a tedious and time consuming task. As this situation persists, the flu virus propagates.

Now, a Center for Disease Control Panel is urging doctors to give a flu vaccine to all children. Though children tend to show lesser symptoms than their older counterparts, sick children have a profound effect on parents. It is interesting to track how networks play into this entire situation. The flu virus can maneuver its was through what is comparable to an undirected graph and, while its effects are probably more powerful in a strong tie than a weak tie, as more tends to be shared within a strong relationship, it is does not discriminate between positive and negative ties. Flu propagates infinitely and, though it is unpleasant to think about, with the easy passage of germs, a person can quickly find himself stuck in an endless network of people he may never knew even exist. Edge are drawn with every turn of a door handle and large “flu networks” easily become one cohesive whole. Finally, this network leaks into another larger, corporate network when parents have to take days off from work to stay home with their sick children. This undoubtedly has an effect on the work environment and may even begin to weaken ties between colleagues as a result of canceled meetings or unreturned phone calls.

The flu vaccine can help this pattern. With this growing epidemic, it will be interesting to see how the availability of this high demand vaccine will break up this seemingly endless network of flu contractors and possibly simultaneously expose greater issues such as the limits of health care.

http://www.cnn.com/2008/HEALTH/02/27/flushots.kids.ap/index.html

In December 2007, the Department of Transportation (DOT) announced that it plans to restrict the number of flights permitted to takeoff and land at Kennedy International Airport to around 82 or 83 per hour in the near future.  At peak hours nowadays, more than 100 flights arrive at and depart from Kennedy International Airport.   U.S. Transportation Secretary Mary E. Peters cited the growing congestion and delays when delivering this pronouncement.  From 2003 to 2007, the number of delayed flights per thousand more than tripled from 20.9 to 77.4.  While federal regulators and airliner trade organizations squabble about whether or not this action is constitutional, the real story lies in an even more radical idea. 

Several economists at the U.S. Department of Justice (namely Tom Whalen, Dennis W. Carlton, Ken Heyer, and Oliver Richard) have proposed another mechanism for distributing flight slots at congested airports: auctions.  While this idea is not new (Bell Journal of Economics published an article named A Combinatorial Auction Mechanism for Airport Time Slot Allocation by Rassenti, Smith, and Bulfin in 1982), it certainly deserves to be rekindled.  The price incentives for using scarce airport capacity are not large enough.  Without the necessary consequences, there is nothing to stop each airline company from scheduling too many flights during peak times, cultivating massive delays, and imposing negative externalities upon all passengers and all airlines.  Time slots would have well-defined property rights.  Airlines would be able to resell them to accommodate changing demand and rescheduling.  As reasonable as it sounds, politicians, airline executives, and even some consultants have voiced their vehement opposition.

The airline companies insist that they could collaborate to bring down delays and congestion. However, is an industry-based conspiracy really the solution?  Has the government not been fighting to break down the dominant airline oligopoly in this country?   

                The New York Times believes that auctioning will drive up fairs and reduce competition.  Yet, we have consistently demonstrated that bidding one’s true valuation of a good, independent of others’ values, is the dominate strategy in a second-price auction.  Furthermore, we showed that market-clearing prices always exist in a market and always maximize social welfare.  Airlines that value popular time slots the most will pay top dollar; their costs will be reflected in their ticket prices.  Unlike in the current system, each airline would assume the true cost of their flight schedules.  In the end, auctions will help consumes and airlines assume their costs as monetary valuations rather than as scheduling variability.  Furthermore, I believe these auctions will encourage airlines to start shifting some of their flights to other nearby airports (such as Long Island MacArthur Airport in Islip) where flight times will surely be less expensive.  I believe in free market mechanisms and auctions, and I believe they can solve the congestion problem. 

Additional resources on this topic:

• http://web.mit.edu/airlines/www/board-meetings/meeting-nov-2005/Harsha-Auctions_for_Landing_Slots.pdf
• http://www.nytimes.com/2007/12/11/nyregion/11airport.html?_r=1&scp=1&sq=game+theory+traffic&st=nyt&oref=slogin
• http://www.newsday.com/business/ny-bzjfk1220,0,3175312.story
• http://www.knowledgeproblem.com/archives/002256.html

1. http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9064698&intsrc=news_ts_head

2. http://blog.searchenginewatch.com/blog/050929-072711

The first article examines the innovative method of search ranking system created by Yahoo. The new Yahoo Buzz program will generate a “BuzzScore” to rate news stories, videos, and blog posts on the Web. The new version contains content from nearly 100 publishers and plans to open the site to all publishers. Yahoo Buzz will also allow users to submit stories to social networking sites such as Facebook and Reddit. The second article examines. Yahoo Buzz allows users to determine the ranking of articles and search results by assigning each a “BuzzScore.” The second article examines the new methods of search ranking, explaining that PageRank is only a small part of what goes into determining overall rank. PageRank operates on a network principle consisting of a “fluid that circulates through the network, passed from node to node across edges, and pooling at the nodes that are most important.[i]” It is an algorithm that assigns a weighting to each part of a hyperlinked set of documents with the purpose of “measuring” its relative importance within the set.

In class we discussed the difficulty of ranking in Web searches. The problem stems from the fact that keywords do not convey enough information much of the time. There are often multiple interpretations of the keyword searched and no way for a computer to know what exact meaning was intended. Advancements in ability to predict the item for which you were searching has led to drastic changes in the appearance of the Web. Google has built an empire off an original idea that proposed that a search engine that analyzed the relationships between websites would produce better ranking of results than methods which ranked results according to the number of times the search term appeared on a page, which had been the current method of search ranking. Since the early days of the Web, development of the most effective search engine has served a vital role for companies dedicated to this purpose. Yahoo hopes to compete with Google, which now serves 49 % of searches compared to Yahoo’s 24 % market share with this innovative method of ranking.[ii]

[ii] http://searchenginewatch.com/showPage.html?page=2156451

No-Limit Texas Hold’em is a complex game with many different variables, including your cards, the cards on board, the number of chips that you and your opponents have, the number of chips in the pot, and more. With all of these elements changing every hand, it may seem that there is no equilibrium strategy in hold’em simply because there are too many different strategies. In fact, in many cases there isn’t, as a previous post discusses. However, in certain very specific situations, there are Nash Equilibria in which sticking with one strategy is optimal for both players, and deviating from the strategy in any way is detrimental to the player who deviates.

Consider the situation at the end of a poker tournament with two players remaining (known as heads up play), when the blinds (forced bets before the hand starts) are large compared to the amount of chips that the shorter stack (person with less chips) has, and the small blind (player first to act) chooses either to move all-in (bet all of his/her chips) or fold. This situation is in fact very common in many online tournaments. Because it is profitable to play optimally, this situation has been analyzed by many experts and it turns out that a Nash Equilibrium exists. The SAGE system, created by Lee Jones, a poker author/poker room manager and James Kittock, a math professor, defines how each player should act in this situation to be playing optimally (note that in practice, if one player is not playing optimally, then SAGE is merely a winning strategy for the other player; both players must be using the SAGE system for it to be an equilibrium).

The SAGE system works by defining a range of hands that you should move all-in with if you are first to act, or call with if you are second to act. Here is the reference table:

First, you compute R, which is the ratio of the short stack to the big blind. If you are the small blind, you should be moving all-in if you have the “SB cutoff hand” corresponding to the current R value as shown in the table, or if you have a better hand. Otherwise, you should fold. If you are the big blind, you should be calling an all-in if you have the “BB cutoff hand” or a better hand. This is a Nash Equilibrium; if both players are employing this strategy, then if either player changes strategy, they lose expected value (they lose money in the long run).

Employing the SAGE system will guarantee you at least breaking even in the long run in these specific situations. Once you learn the SAGE system, when playing against an opponent who is not playing optimally, you gain an advantage estimated between 5 and 40 percent, which is huge in poker. SAGE illustrates that there are indeed known optimal plays under very specific conditions at a poker table, and those who use SAGE can’t lose money in these situations in the long run. Note that SAGE won’t make you a winning poker player automatically, as you will need to play well in a tournament before the conditions necessary to use SAGE are met. However, it can certainly help a player of any skill level play a better heads up game, and therefore make more money (or lose less) in the long run.

References:

http://www.cardplayer.com/magazine/article/15250

http://klopzi.blogspot.com/2007/09/sage-sit-and-go-endgame-system.html

In response to several blog posts earlier on this semester, describing our course’s broad application to biological networks (such as this one, this one, and this one) I highlight that Piers J Ingram, Michael PH Stump and Jaroslav Stark published research in 2006 demonstrating that network effects in biology are not only often poorly understood, but in their example are also surprising in the manner which they fail to elucidate function in their larger systems. The article, available here, describes the simulated dynamics of a bi-fan motif found in the transcription regulation network of S. cerevisiae (baking yeast). Nodes represent genes, and the directed edges between them represent how the protein products of the genes regulate the transcription- the process in which the information in DNA is copied to RNA- of other genes. Figure B, as taken from the article, below, is known as the bi-fan motif, where a motif is defined as a particular node-edge structure that is found with particularly high frequency in any network, relative to a “null hypothesis” with that network constructed with random edges.

By using differential equations to describe the concentrations, over time, of the relevant molecules in this network found in S. cerevisiae, these researchers were able to simulate how the molecules in this network would be produced. What they found was that depending on the particular genes they chose, there was, in fact, no set of consistent regulatory behavior that the bi-fan motif exhibited.  

This is a curious result, for several reasons. First off, although essentially what’s happening here is a set of interactions between genes and their protein products which regulate other genes- a situation that ostensibly lends itself naturally to analysis by networks- this scenario seems to not be easily modeled by networks, or at least by network motifs. Networks are not enough. Rather, a priori biological knowledge about the situation is required to generate an accurate model. Furthermore, and perhaps surprisingly, this implies that the motif structure of these small networks do not define their function. In contrast, as oft repeated as a theme in introductory biology courses, and as is found true in much research, biological structure determines biological function.

Despite what shortcomings this work by Ingram et al. may have in terms of network modeling, it certainly provides a valuable, abstract suggestion to biological reductionists interested in interactions between components of biological systems: perhaps structure does not indeed always determine function. 

—–

 The article discussed in this post can be found as: Barabasi, Albert-Laszlo and Zoltan N. Oltvai. “Network motifs: structure does not determine function.” BMC Genomics 7:108 (2006): 28 February 2008 <http://www.biomedcentral.com/1471-2164/7/108> 

The study of social networks is crucial to understanding economic development in lesser developed countries. Too often, the traditional approach to analyzing demographic behavior - with individuals acting independently with each other - does not fit in with real life trends. Hence, it is necessary to investigate decision making in social networks to fully understand the dynamics of a developing country.

In the paper entitled “Social Networks in Ghana” the connections amongst farmers in Ghana are examined. It is clearly shown that the level of knowledge that a farmer has of new technology is directly correlated to the number of relations that he has to other farmers with this information. We can consider this in the context of networks, where farmers are modeled as nodes, and the ties to other farmers as edges. The edges represent more than simply a friendship between two people. The more ‘edges’ that a farmer has, the more access he has to new resources and skills. Hence, the farmer can learn techniques from other successful farmers and become more profitable himself.

Economic development is also influenced by political power within a social network. In Southern Ghana, access to land is largely determined by social rank within a group. Those who have strong connections with leaders are likely to be able to fallow their land for a longer period of time. Women are more likely to support each other across other villages whereas men share the risk with extended male family members. We can model this network as being a triangular unbalanced network. The three nodes represent the leaders, women, and men of a village. Both women and men have weak ties to each other because they are busy competing and strengthening their weak ties to the heads of the village. Triadic closure is uncommon because strong ties with leadership are rare and exclusive. Hence security to land flows is entirely based upon the strengths of a farmer’s political connections and social authority in his local network.

http://globetrotter.berkeley.edu/macarthur/inequality/papers/UdrySocialNetworksGhana.pdf

Econ 204

Blog Post

            A social network that has recently become a hot topic in both news and amongst college campuses is the website Juicycampus.com. This website has sparked a new wave of opinions on these internet based social networks that are present in out day today. The website is a blog post website which is basically a forum for anyone to post topics or comments on other topics that people have posted before about basically anything. These people can post anonymously or with a name, but with my experiences, people are posting anonymously. This forum has no real controls over what is being posted unlike some blogs, which must be approved by an administrator in order to be seen. Some of the blog topics, especially here at Cornell, are about drugs, sex, drinking, and rumors about sport teams. This website goes toward tarnishing the reputation of people and forming a sense of embarrassment, much of which could and should be avoided.

            This blog site, Juciycampus.com is indeed a social network with many related topics to some of the material we have covered already in class and our problem sets. The website information as stated in the article, is passed through social networks. These networks are a bunch of nodes that are connected to each other with edges. The nodes represent people and the edges, if positive, represent a friendship between those people. If one node or “person” notices that one of the people in his social network are mentioned on juicycanmpus.com, then he will pass the information through his network until it reaches the person the post was about. This is directly relates to how information passes through all different types of social networks in our world today. These friends that are passing information to its “neighbors” of “friends” can be considered local gatekeepers, which in this class are known to pass information to its neighbors through edges. I know that I did not know about this website until I heard that I was on it from a group of friends that I would include in my friendship network. This shows how the information went from each of my friends through edges until it reached me. Another topic that is related to this course from the article and the website is that of gatekeepers. Gatekeepers are nodes that connect two otherwise separated networks. Basically, there is no other path that the information can be passed through except for through this node, which makes it a gatekeeping node. This website was passed along and became popular through gatekeepers, in which friends of one network of friends that had a friend in common with another network spoke of the website and information was passed on through this gatekeeper.

            Another topic that is included in the website is that of game theory. When you see a post with your name on it or a friend’s name, you have two options to play. You can either post back attempting to correct the information and rectify you or your friend’s reputation or choose not to play the game and not post at all. If you choose to play the game and post back, you may gain some positive back, but that could change if the poster says another negative thing about you to put your reputation down worse. If you choose not to play, you might lose some embarrassment, but it shows that this post does not affect you and you may end up losing less than you would if u replied.

            All in all for me, I do not agree with the website and I would not reply back. I feel that the people that are posting on this website for means to hurt or harm somebody are just doing it to draw attention to themselves and really are the ones that are insecure. I would rather avoid the whole situation and hope that this website is soon fixed, banned, or more closely watched.

The current scenario in debt and equity markets across the globe brings together many principles of networks and graph theory. I will briefly outline the modern structured finance industry to provide not only a window into what has happened, but also to examine critical gatekeepers – the bond insurers – whose potential insolvencies will shock the financial markets in the next round of defaults.

This article from The Economist explains the current issues in the bond insurance industry; and I will elaborate on the gatekeeper role which has concentrated financial risk and led MBIA, Inc.’s CEO to remark that “we were not designed or structured to be the most important company in the entire financial system.”

http://www.economist.com/finance/displaystory.cfm?story_id=10733072

Initially, bond insurers like MBIA focused on the staid business of insuring payments from municipal bonds. But over the past decade, they began deriving the bulk of their revenues from – and writing over half of their policies on – structured finance.

And thus the story begins with the lending industry, which has evolved from the old days of going down to the local savings bank to take out a loan for, say, a new home. The industry now involves an intricate network of agents, which can be viewed as nodes. This network is the basis of risk-transfer that is at the heart of structured finance. Furthermore, the edges between these nodes are very difficult to identify and discern – hence the recent bloodshed on Wall Street, which stems directly from these credit markets.

Unlike the old-fashioned home loan between you and the local thrift, these new agents – lenders along the lines of 1-800-MORTGAGE and FastMortgage.com – do not end up keeping the loans as assets. Instead, they undergo an arcane process of being pooled together with a vast number of other loans, by a second agent, into what is called an asset-backed security. And it does not stop there. They are then passed on once again to a third agent, who uses pools of asset-backed securities to create a new security, a collateralized debt obligation (“CDO”), whose interest payments are contingent upon the performance of the giant pools of underlying loans. In fact, there can even be an additional agent in this line, who constructs CDO-squareds – CDOs to the second power! – or in other words, CDOs of CDOs.

We then reach the fifth agent, the asset manager, who may make a small investment – yet keeps taking fees in return for managing these complex securities for investors. There is also a vast array of accountants, lawyers, middlemen and other financiers, all taking profits every step of the way.

Maybe some clearer insight can be gained from author Tom Wolfe, in a comment he made last year from the floor of the New York Stock Exchange:

“Joseph Schumpeter, the economist, once said that stocks and bonds are evaporated property, meaning you’re pretty soon going to lose sight of the underlying assets. You’re dealing in pieces of paper. Now, these people are selling shares in evaporated evaporation to millions of people. And this is now evaporation cubed. And nobody knows what they’re investing in. It’s E to the third power equals … insanity?”

Here is a graph of a CDO to help visualize the process, components, edges and nodes:

And here is a graph of the cornucopia of mortgage debt, a different dimension of nodes which are being funneled into the CDO:

In this network, risk flows through the graph in the order that I have described, from agent to agent, node to node. There are dozens of large investment banks that form clusters of nodes – with hundreds of minor agent-nodes scattered alongside and around them, each with varying directional and multi-directional edges. Also, most of the big banks have functioned as several of the different agents at the same time. Suffice it to say, if this were such a simple graph to visualize and understand, bankers would have happily avoided their firms’ hundreds of billions in losses and, in many cases, their own pink slips.

But wait! There is an astonishing moral hazard here: every agent is paid upfront – including the agencies who rate the quality of all these securities – except for the ultimate holders of financial risk. Yet at the same time, the underlying loans have minimal visibility to the ultimate holders of risk, who must rely solely on the ratings given to them.

And thus the natural question arises: Who is the “ultimate holder” of risk anyways? Who bears the losses if, say, I decide to stop paying interest on my home loan?

We now return to the bond insurers: the linchpin-like gatekeepers who bear a staggering amount of risk and liability because they must pay claims on every piece of defaulted and evaporated evaporation that they’ve insured, across the entire spectrum of structured finance. MBIA is an example of such a company, and in a rather strange line of business: insuring over $50 billion of subprime CDOs and similar credits using nearly 150% leverage. This is … ummmm … a curious arrangement to say the least.

Sadly, it doesn’t end there: MBIA’s perfect credit rating, given to it by the same agencies that rated the CDOs, is critically important for protecting the value of nearly half-a-trillion dollars of other obligations that it has insured. But if MBIA is swept by a tidal wave of losses from CDOs, these other guarantees will be rendered worthless, leaving their counterparties holding the bag.

As The Economist writes, the extreme concentration of $2.4 trillion worth of securities that are collectively guaranteed by MBIA and its fellow bond insurers presents a massive risk to the financial system, which is an unimaginably more complex and intricate graph.

On Facebook, Scholars Link Up With Data

The New York Times

By Stephanie Rosenbloom

December 17, 2007

A team of researchers at Harvard and UCLA are monitoring the Facebook profiles of a class of students to study how personal tastes and habits affect how social relationships are formed. Research has found that the benefits of Facebook friends may include a positive impact on students’ well-being. Researches have also noted the large presence of “weak ties” with people met at parties or functions that are not very close friends. These weak ties have been found to provide the person with new perspectives and opportunities that might not be present in their circle of close friends. The Harvard-UCLA researchers are investigating the concept of triadic closure present in Facebook. They have noticed that one’s friends are becoming friends with each other. The researchers have found that triadic closure can be very helpful in bringing people together and preventing social isolation and suicidal thoughts.

This Article describes research on Facebook involving concepts of social networks that we have discussed in detail. The research investigates how social relationships are formed. In class we explored how a network is structured and how common traits will create a tie between two people, and how triadic closure can fill in missing ties in a network resulting in the strengthening and connecting of the entire network. Towards the end of the article they mention the research Harvard and UCLA are doing on triadic closure specifically. The researchers have found data that shows people friending friends of their friends, which ultimately results in a stronger network, as is seen in the reduction of social isolation and suicidal thoughts. Along with the concept of triadic closure comes the use of “weak ties” in the article to describe when a person friends someone they met at a party but don’t know too well. The researches have found data that this occurs, and that it is also very positive because it gives people new opportunities and perspectives. Facebook is one of the best ways researchers can investigate the creation and strengthening of networks. It is an easily traceable network of millions of people that are constantly making new ties with other people through triadic closure and other methods of socializing.