At http://computer.howstuffworks.com/internet-infrastructure.htm, howstuffworks explains how the internet is physically connected, and we can relate the internet to some important ideas of graph theory. Say we have a computer in a corporate network (A) that wants to connect to a computer (B) across the country, on some other network. If we start with a simple corporate network of which A is a member, employee computers are connected to a local router. These computers are nodes, the methods of connections are edges, and the router is essentially a gatekeeper since it is the only way for nodes on the company network to talk to each other. This corporate network is connected by, for example, a T1 line to an ISP. This connection from company to ISP is a bridge, since it is the only path between the computers on the company network and the larger internet.

At this point, the ISP will connect to a regional network access point (another bridge, the only path for the computers on the company network to access networks across the country). Other ISPs and large corporations also agree to share information and connectivity at these network access points, which are connected to each other around the country. At this high level, there are no gatekeepers since there are multiple paths between the interconnected NAPs. But when we have reached the NAP which B’s ISP connects to, choices are fewer. From the NAP, we go to B’s ISP, then to B’s local network, then to B (following the reverse process of how A gets to the NAP). We can see that the internet is essentially made up of tons of nodes connected to gatekeepers, connected together by bridges, all essential elements of graph theory. It is fascinating to think that this process takes such a small amount of time, usually much less than a second.

http://www.nytimes.com/2008/02/26/technology/26tube.html?_r=1&ref=business&oref=slogin

http://thelede.blogs.nytimes.com/2008/02/25/pakistan-blamed-for-worldwide-youtube-break/?ex=1204606800&en=1911c530be444f91&ei=5070&emc=eta1

Although I was not personally aware of it at the time, apparently YouTube was made unavailable to most of the world this past Sunday for a period of about 2 hours because of an alleged accident made by the Pakistani state-operated Pakistan Telecommunication Authority (PTA). The two above links will direct you a New York Times article and blog respectively that have further information and insight into this strange happening. The paragraph that follows briefly summarizes the incident.

The Pakistani government recently decided to block its citizens from viewing YouTube, purportedly in response to a specific video on the site that was offensive to Muslims; it was by no means the first country to implement such a block. To cut domestic access, the PTA set up a dummy route for YouTube, which essentially sent the site’s traffic to a “black hole.” What reportedly went wrong was that it next announced this dummy route to its telecommunications partner, the Hong Kong based ISP PCCW, which mistakenly accepted and relayed that route to several dozen other major and minor ISP’s around the world. These ISP’s, now faced with competing routes, automatically accepted the Pakistani one over the original, functioning one because of the Border Gateway Protocol, which favors and selects longer routing addresses over shorter ones. YouTube was thus blocked for hundreds of millions of internet users worldwide until the website’s administrators caught on and alerted the ISP’s to the issues, which thankfully only took a few hours.

This story clearly connects to concepts in graph theory and the structure and function of information networks. While we might tend to think of the physical and virtual structures of the internet as a massive graphs with billions of nodes and edges, such that a popular site/node like YouTube would lie on millions of different paths and cycles and would thus be extremely difficult, if not impossible, to globally block in such an unintentional and uncoordinated way, clearly there are other factors to examine. The block is a good example of the shortcomings of graph theory, in that the structure of a network is not always capable of demonstrating that network’s functioning and behavior, which are in this case perhaps best understood in the context of internet protocol and the network of ISP’s; the former being responsible for the dummy route replacement, while the latter explains the widespread scale and ultimate severity of the block. Ironically, in this instance, the notion of optimal paths, combined with the interconnectedness and automatic processing of ISP’s seemed to prevent proper functioning.

Finally, this story also raises some serious question about internet security and vulnerability; if a country as relatively small and undeveloped as Pakistan could very easily accidentally disrupt YouTube’s traffic and block the site for users worldwide, then what sort of cyber-chaos and confusion could potentially be unleashed by a malevolent country or group? Furthermore, does the blame lie with Pakistan, the ISP’s, internet protocol, or something else entirely? Lastly, what can be done to prevent something like this or worse from happening again, and who is responsible for doing it?

Reporters tap into social networks and find gold

Social networking websites have certainly gained prominence in the past several years. Less than a decade ago, social networking sites were practically unheard of. Now, MySpace, Facebook, and their lesser-known competitors play a major role in the lives of over 300 million people. With such a large member base, it is no wonder businesses are scrambling to utilize these websites to generate revenue. However, the true power of these social network websites lies in its ability to log interactions between each of its members. For example, by clicking “View Friends” on somebody’s profile on Facebook, you can see the names of all the people that that person is friends with on Facebook. This has incredibly far reaching implications on the world of journalism. Oftentimes, reporters have a hard time tracking down witnesses of a crime or friends of the victim. In the past, reporters had to go door-to-door to find the information they needed to tell the whole story. Today however, reporters are able to use social networking websites to find contacts quickly and efficiently. This proved to be pivotal in the reporting of the death of a cyclist in Chicago last week.

Held once a year, the Tour Da Chicago is an unofficial street race that takes place in stages scheduled at various points throughout the winter. Cyclists participating in the Tour Da Chicago have to compete not only against other cyclists, but also against the cars on the roads. This obviously has inherent dangers, since traffic laws such as stop signs and traffic lights have to be disregarded in order to outpace one’s competitors. This proved to be fatal for Matthew Manger-Lynch who ran a red stoplight and was killed when an SUV struck him. Initial reports spelled his name wrong, and did not provide much information about the participants of the race at all. The police report was also sparse. However, a reporter working for the Chicago Tribune was able to find and fit pieces to the puzzle by scouring social networking websites for information. Looking through MySpace, the reporter discovered a subculture of cyclists that were into hardcore street cycling, called “alleycats.” One of the cyclists contacted her, and provided her with more contacts. Outfitted with this information, the reporter was able to produce a follow-up report that shed light not only on the circumstances of the accident, but also the underground world of street cycling.

Clearly, the ease with which personal networks are accessed through these social networking websites allows reporters to find crucial information more quickly than ever before. As a result of these social networking websites, distances in networking datasets are decreased, degrees of separation are lessened, and social networks are larger than ever before. The chief implication of the shift towards online social networks however, has to do with Network Exchange Theory. In any network, the person with exclusive information and high “betweenness” has more power than the rest of the people. In a sense, social networking websites eliminates this monopolization of power. For instance, if the reporter mentioned above only has one source, then the source has high power in the network of people connected to the victim. If this source does not talk, then the story goes dead. However, in online social networks, everyone that is friends with the victim is in a position of power, and the chances of obtaining information are much higher.

As I write this blog post, I’m engaging some opponents in fierce online gameplay in Settlers of Catan.  As far as I can remember, I’ve played online multiplayer games since I was in 4th grade-starting out games against computer A.I “bots”, and then working my way up to people I’ve never met before online toward 7th and 8th grade.  My game of choice- Diablo2.  You could play the main story line game with friends, or people you don’t know, and attempt to get the best armor, weapons, and other miscellaneous items that would give you more power than the other people in the network.  It was the first time I had to encounter a huge amount of competitors in one arena- with people playing from all countries(or at least most of them).  It was amazing- and highly addictive. 

As most of you gamers out there know, games such as starcraft, Diablo, and Warcraft were huge components of the uprising movement of online gaming.  Soon after the internet began turning into commonplace for most of the socioeconomic classes, it was more accessible to those who wanted it for gaming.  Soon, common websites such as yahoo would be creating games such as hearts, scrabble, and boggle to name a few, that were newly remastered so that people online could play each other.  This wasn’t the only movement to make money on this online gaming, for as the first systems to have online capabilities were the ps2 and xbox.  Today, their empire still stands with their new creations, the xbox 360 and ps3.  These systems allow for a person to create their own profile, and add friends that recognize their username and enjoy playing with each other.  These ties to people that the player may know can form a network, and as a whole the gaming world form a huge network where many people are tied to smaller components. 

 And now, with the also rising capability of wireless networking, other games systems such as the PSP and Nintendo DS have taken over in all types of locations.  As long as the place people want to play at has wireless capabilities, then people within the vicinity can battle or play together on a common mission.  Overall, in these types of games where many people are involved, players have a multitude of choices to make in gameplay.  The possibilities for this type of gaming are endless, and only time can really tell how the gaming world will try to expand this global gaming network to the max.

New Scientist Tech: Wireless worms will follow influenza’s example

It probably comes as a surprise to many people as to how closely real-life flu is related to its digital counterpart, or rather, the other way round. After all, wireless worms are just mere exploit codes running on devices capable of communicating with other devices. Devices are not humans, and these digital malicious codes are not real life bacteria/virus either. They are two different beasts. Intuition also tells us that minute differences in local conditions and factors could create a ripple of changes that reverberate throughout the bigger networks. How then, should we expect wireless worms to mirror influenza’s spreading pattern?

The answer to that mind boggling question does not lie very far away: right in your pocket. Many cellphones and blackberries have some form of communication capabilities: infrared, bluetooth, wifi, et cetera. Due to their wireless nature, they are not tied to a particular location like their wired counterparts do. Hence these communications protocols are “mobile”, so to speak. They can be moved around, seamlessly creating and destroying connections to other devices, shuttling data around you and through you in the form of electromagnetic waves. With increasing automation, these devices speak to whatever devices they find, communicating their names, serial numbers, and even perform an aptly named handshake protocol.

So how does this relate to our topic at hand? If you observe, our increasing reliance on such devices means we are carrying them along with us all day. They follow us wherever we go. Their pattern of movement mirrors exactly the very mobility of individual nodes moving around in a social network we all engage in. If we consider influenza (or any contagious bacteria/viruses for that matter) for a moment, they also rely on human-to-human contact. When two humans are in proximity, there is a significant chance of the viruses in one host hopping to the other. In the same way, these mobile devices, should some of them got infected, they would spread it around as they come in close proximity with other devices. Long range communication protocols only serve to exacerbate this problem, as infected devices can reach out farther to more devices.

There is a significance to this digital network’s mirroring of their physical counterparts. Firstly, it does not matter what form the network structure takes on. This implies networks need not have a physical basis for its functioning. A network can take on any form so long it encapsulates the notion of nodes and the connections between them. In other words, this is a validation of the tried and true definition of a network. Secondly, networks rely very heavily on proximity of nodes, or rather, links between nodes. Differences in connectivity, locality, temporality can effect changes to the network as a whole. What information or physical entities flowing through a link, dictates how it will propagate down the chains of links, and also how it will affect other nodes in the future. Thirdly, the principles that apply in a network may be readily transferred to the other. This implies networks, no matter of what form, (social network, phone network, outbreaks, solar systems, etc) all have a common modus operandi. Ultimately, we may discover a plethora of a certain kind of local effects that can effectively produce a common global effect on the network. This will serve as network scientists’ box of tools in shaping and manipulating a network, much as how we use hammers and nails in carpentry, or use integrals in differential equations. Network calculus, so to speak.

Why it’s rational to vote

This October 2004 post on a Columbia University blog deals with the seemingly simple question of whether voting in a presidential election is a rational act. They present a utility function U=p*B - C, where p is the proabbility that your vote will change the outcome of the election, B is your benefit from your candidate winning, and C is the cost you incur by going to vote. They begin by throwing out a value of p=1/10,000,000 as a reasonable value for p. This makes it seem as if voting should be irrational (negative utility) because even for B=$1,000,000, the cost of voting would have to be less than 10 cents. They present 2 explanations for why people still go out and vote. First, people feel happy performing a “civic duty”. Second, they claim that you should consider the potential gain per person, instead of just for yourself. Their example numbers are B=$100/person and p=10,000,000, which means voting is rational if your cost C is less than $30. They support this thinking by citing a study in which British citizens said that they vote in order to obtain “benefits for groups that people care about”.

While their argument reaches a nice conclusion (we want people to go out and vote, in theory), their analysis seems far to simple. The most glaring problem with their reasoning is their value of p=10,000,000. This roughly represents a value of 1/n (n = number of voters) that is seen in other articles on this topic. I would propose a different model for determining p. Call x the fraction of voters who prefer your candidate. If x = 0.5, you expect their will be roughly the same number of votes for each candidate. In this case, the probability that your vote will be the deciding vote is found by the binomial distribution n C (n/2) * (0.5)^(n/2) * (1-0.5)^(n/2). Using Stirling’s approximation, this gives p=1/3963 if there are 10,000,000 voters. That’s good news for showing voting is rational, but there is a downside. If you aren’t sure the election is extremely close to a tie, the p values get reduced dramatically. For x=0.495, p is roughly 10^-221 already. This makes it surely an irrational prospect to go vote according to our utility formula. In states where 60% of the vote is expected for 1 candidate, going to vote could surely never change the outcome of the vote.

How else can we rationalize voting? Attacking the problem from a game theoretic perspective, each player has 2 choices, vote or don’t vote. It would appear that there is no pure strategy equilibrium, because if everyone chose “don’t vote”, then a single person could deviate to “vote”, win the election, and get his nice payoff. If everyone else is voting, then a person would chose “don’t vote” because there is essentially no chance of swinging the outcome of the election. There will be some mixed strategy Nash equilibrium where each person goes and votes with some probability p.

I believe the key to rationalizing voting is to realize that the outcome is not some binary feature based on who won the election. The percentages that each candidate receives are important for many reasons. For example, the outcome of one election surely has an impact on the next election 4 years later. States like Massachusetts are considered “blue states” because they have a history of overwhelming democratic majorities, so it would seem that voting is illogical and it will always stay a democratic state. However, republicans consistently go out and vote, they will pick up larger and larger fractions of the vote each year, and eventually the state will be contested again. However, I don’t feel that this reasoning will get far enough to actually make it a financially rational decision to go vote. Let’s hope that people are happy with it being their “civic duty”.

The government’s role in regulating the internet has recently been a controversial issue recently. Many internet service providers are attempting to increase profits by charging content providers like Google and Yahoo! for broadcasting of their content to users. Content providers have supported Congressional proposals for a government-enacted ban on such charges while internet service providers insist it is a way to ensure availability of critical services and increase profits to allow for network improvements. 

With recent internet success by companies like Google, eBay and the Facebook, internet service providers look to reap some of the success by exploiting power they have intrinsically on the network. Below is an example diagram of the structure of the internet. Note that end user’s must receive all of their data via their respective internet service provider. This gives service providers power over the content obtained by end users and, since content providers depend on visits by these end users to earn their massive profits, service providers are now looking to exploit their power in the network to increase profits as would be expected in any economic network.   

Content providers insist that such taxes would be a detriment to the end user experience and suggests that the government enforce the ideal of Network Neutrailty, where no ISP can discriminate in the way traffic is handled based on content provider. Internet Service Providers, however, argue that these charges will allow for the institution of internet “fast lanes” that would allow critical content from heavily used content providers like Google and eBay would be delivered quickly and reliably to end users. Though congress has been reluctant to approve such a mandate as of yet, this will be an important issue to watch in the future.

http://www.news.com/8301-13772_3-9871834-52.html

“Social network invites can be a plague”

Posted by Daniel Terdiman

Web-based social networks is probably one of the fastest growing industries in the contemporary age. It effectively brings friends and families a lot closer through the internet. We are no strangers to Facebook, MySpace, and many other social network sites. And we are definitely familiar with all kinds of email notifications automatically generated by them every day, such as friend requests, event notifications, etc. And the chances are at large that we joined these social networks in the first place were because of friends’ invitations. This article “Social network invites can be a plague” from the CNet News discusses the problems exist among this new kind of social networking.

For some people who constantly have their inbox stuffed with invitations and notifications from all sorts of social networks out there, their first reaction is probably annoyance. In a way, these emails could be considered spam, however, they are much harder to be taken care of then simply click “Delete”, since they are actually sent from a friend or an acquaintance of yours. You are probably unsure whether you can just ignore it if you still want to maintain this friendly edge between you and your friend.

On the other hand, for people with much more social and outgoing professions, they perceive indirectly receiving tons of emails from close or distant friends a sign of their success in social networking, which could be very closely tie to the success of their career. For example, the article mentioned a vice president of an Entertainment Media Venture, due to the nature of her job, she holds a much more welcoming attitude toward all these automated social network emails than a visiting professor of art at Carnegie Mellon University.

Several solutions to this problem are also mentioned in the article. One is to simply terminate your membership at that particular social network like how Bill Gates did, since he was receiving more than 8,000 friend request per day. However, looking at the bigger picture, the social network industry should probably look more into creating more specific, and functional sites that target toward different age groups, different professions, and different interests; so that these social networks can serve a better purpose for all kinds of people that seek easy breezy ways to simply connect with their friends without having their daily life disturbed.

I was interested in the ultimatum game discussed in class, so I did a literature search through “psych info” to find out more about the psychology behind this game. I game across a 2005 article by Heinz-Deiter Meyer in the journal of economic psychology entitled “Norms and self-interest in ultimatum bargaining: The prince’s prudence”(link below). The article discussed individual differences that were connected to either accepting unfair deals (to them) or rejecting them.

One way people differ is on how “Machiavellian” they are. People who are high for Machiavellianism (“high-machs”) are better (and more willing) at lying and believe that the ends justify the means. “Low-machs” on the other hand tend to act morally, believe that honesty is the best policy, and that people are basically good. This difference is measured using the Mach IV scale.

The article found that for one-shot ultimatum games, high-machs were much more likely to accept unfair deals than low-machs. The reason for this finding (according to the researcher) is that while low-machs are highly concerned with fairness, high-machs are more concerned with self-interest. Consequently, high-machs accept more unfair deals in the ultimatum game than low-machs because they are more interested in themselves than in fairness. In a second experiment, the game was changed from a single-shot game to a multishot game. What was interesting about the multishot game is that while both moderately-high and moderately-low machs were continually exploited, the very-high and very-low machs were effective at countering exploitative deals. For the low-machs, the explanation is the same: they were motivated by goals of equality, and as such rejected bad offers. For high-machs, it is in their own self-interest to reject bad deals, so that they can signal that they will only accept fair offers.

This is interesting because in class we discussed how individuals engage in these exchange games as if everyone behaved in the same way. This study shows that different people, motivated by differing goals, will play the game differently. It is especially interesting to consider the role of high-machs in bargaining games. It has been shown in earlier articles that for games in which there are three people, all connected by an edge, high-machs on average come away with more than 50% of the earnings. The models we use assume that everyone is a high-mach (and as such should be motivated purely by self-interest), however, while some people are motivated by self-interest many are driven by goals of fairness and equality (the low-machs). This individual difference variable may help us account for some discrepancies we encounter when modeling network exchange theory.

Article: http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V8H-45D0NT5-1C&_user=492137&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000022719&_version=1&_urlVersion=0&_userid=492137&md5=9e806561bfe467ff41e5ead766bbf204

Recently, we have witnessed a decline in the value of US dollar in the global markets. The US dollar is at a multi year low against a lot of currencies across the world. People are led to believe that a drop in the value of dollar will only affect the US markets and other countries will keep on prospering and their economies will grow. However, such an assumption does not hold true in today’s world and exactly the opposite happens because all the countries are extensively interdependent on each other and a large part of a country’s revenue is through exports. The analogy that can be adopted here is a stone dropping in a pond. The affect of the splash is not isolated to one region of the pond but spreads to all the corners until it dies out.

According to the above mentioned claim that markets across the world are extensively linked to one another, a decrease in the value of dollar will produce effects all around the world. Other large economies such as Japan and China would restrain their investment in US economy because that would potentially lead to a loss in their investment. Such a strategy adopted by big economies would eventually worsen the conditions of that economy as a decrease in investment would lead to a stagnation rather than flow of money. The following are the recent updates in the world market taken from cnn.com :

Asia Pacific & Australia
		Index	Change	%Change	Level	Last Update *
	Australia	ASX 100	-69.80	-1.53%	4,493.60	2/29 4:47pm
	Australia	ASX All Ords	-71.40	-1.24%	5,674.70	2/29 4:47pm
	Australia	ASX Mid-cap 50	-48.60	-0.82%	5,901.80	2/29 4:47pm
	Hong Kong	Hang Seng	-260.02	-1.06%	24,331.67	2/29 5:59pm
	Hong Kong	HSCC Red Chip	-34.27	-0.62%	5,484.34	2/29 4:25pm
	Japan	Nikkei 225	-322.49	-2.32%	13,603.02	2/29 12:00am
Europe
		Index	Change	%Change	Level	Last Update *
	Belgium	Bel 20	-5.56	-0.15%	3,757.12	2/29 6:07pm
	Europe	DJ Stoxx	-50.76	-1.57%	3,188.82	2/29 7:00pm
	Europe	Euronext 100	-11.19	-1.29%	852.95	2/29 6:07pm
	Europe	Euronext 150	-20.00	-1.25%	1,583.17	2/29 6:07pm
	France	CAC	-74.57	-1.53%	4,790.66	2/29 6:10pm
	France	SBF 80	-74.75	-1.34%	5,523.24	2/29 6:21pm
	France	SBF 120	-53.13	-1.51%	3,473.43	2/29 6:21pm
	Germany	DAX	-114.39	-1.67%	6,748.13	2/29 6:31pm
	Germany	MDAX	-128.78	-1.40%	9,093.54	2/29 6:31pm
	Germany	TECDAX	-18.14	-2.26%	784.03	2/29 6:31pm
	Netherlands	AEX	-6.38	-1.41%	446.53	2/29 6:07pm
	Norway	BRIX	+3.09	+0.08%	3,913.92	2/29 3:56pm
	Norway	OSE Industry	-7.29	-0.43%	314.07	2/29 4:28pm
	Sweden	OMX	-5.79	-0.60%	965.29	2/29 5:44pm
	Sweden	OMSX All Share	-4.94	-1.51%	321.35	2/28 12:00am
	UK	FTSE 100	-81.40	-1.36%	5,884.30	2/29 4:35pm
	UK	FTSE All Shares	-42.08	-1.38%	3,013.02	2/29 4:36pm
	UK	FTSE Eurotop	-38.99	-1.39%	2,757.24	2/29 4:45pm
	UK	FTSE Techmark	-24.12	-1.53%	1,547.85	2/29 4:36pm
Americas
		Index	Change	%Change	Level	Last Update *
	Canada	TSE 300	-291.20	-2.10%	13,582.69	2/29 5:05pm
	Canada	CDNX	-27.43	-0.98%	2,782.07	2/29 5:05pm
	Canada	S&P/TSX 60	-18.79	-2.31%	795.23	2/29 5:05pm

One observation to be noted in the above table is that relatively small economies such as Norway etc are not affected by a change in the US economy. Similarly, we can say that if the Norwegian market crashes, it would not produce the same effect around the world as in case of the US economy. This leads to the conclusion that economies that are at the center of networks and are heavily linked to other nodes (economies) contribute the maximum in the network. A change in these “power nodes”, as I call them, would affect the rest of the network but it would take a large effect in many smaller nodes to produce a change in a power node.

The article (link below) on the decreasing value of US dollar cites that the Federal Reserve Chairman Ben Bernanke is hinting at cutting interest rates, taking into account the decreasing value of dollar and the worsening economy. This strategy will worsen the condition of the economy further. According to the above discussion, the decrease in the interest rate would lead to a decrease in investments by other countries in the US economy hence a decrease in the imports from other countries to US. Thus, it can be concluded that ideally, the optimal Nash Equilibrium will exist in the global markets if each economy is extensively interlinked to all other economies, contributing the maximum possible to the entire network to sustain the flow of money.

References:

http://money.cnn.com/2008/02/29/markets/dollar/index.htm

http://money.cnn.com/data/world_markets/