http://www.latimes.com/news/local/la-scientology062590b,1,279442.story?coll=la-news-comment

For the past several years, the Church of Scientology has received a great deal of media attention. Tom Cruise publicly proclaiming his alliegance to the church and the increase in the membership of high-profile celebrities are some of the examples. According to the article on the L.A. Times, the Church of Scientology sees active recruitment of celebrities as one of their highest priorities. The Church attempts at attracting new celebrities by providing them with first-class treatment and by convincing them through the Church’s influential celebrity members. They are using the theory of information cascade to their advantage in drawing in more interest, which ultimately results in new memberships. The church’s use of information cascade is similar to the example of restaurant A and B given in the lecture. Following the herd to the crowded restaurant B over the empty restaurant A is seen as a rational behavior, whether or not restaurant B is superior to restaurant A.

In the last several years, the church of Scientology was successful in converting celebrities such as Jerry Seinfeld, Beck, Nancy Cartwright (voice of Bart Simpson), Brandy Norwood, Katie Holmes and hundreds of others. We can attempt at explaining this phenomena by applying it to the simple model given in the textbook. Celebrity #1 joins the church after he observes a benefit in the first-class treatment provided by the church. Celebrity #2 observes the benefits of joining himself and he observes the celebrity #1’s action. Since both signals are high, celebrity #2 joins the church. When it’s celebrity #3’s turn to decide, he observes his signal along side the actions taken by the first two celebrities. Even if he does not receive a high signal, celebrity #3 joins the church since celebrity #1 and #2 joined. Celebrity #3’s low signal cancels out with one of the high signals received by the first two, which results in #3 being in the same situation as celebrity #1.

dab97

http://www.vnunet.com/vnunet/news/2212947/microsoft-friends-social-networks

The above article discusses Microsoft’s plans to integrate their Windows Live system into leading online social network services. Windows Live is a set of online features that can be accessed through a browser that include instant messaging, search engines, and web security. Microsoft seeks to open up new ties to networks such as Facebook, Bebo, Hi5, LinkedIn, and Tagged to allow subscribers to easily and securely exchange information among these different networks. Under this new partnership, Windows Live users would be able to connect to peers and friends located in these closed networks without opening up private information to outside networks.

The rise of social networking giants such as Facebook have created multiple bridging opportunities among existing internet services such as Windows Live, and have also created problems in information privacy. Windows Live seeks to bridge massive components online in a way that opens up new paths yet protects privacy and ensures security for its users. It is plausible that this development, if successful, could be the first of several steps toward integrating all social networking sites. While multiple online networks may become integrated, it is also important to note that they may not be necessarily connected.  The networks would remain mostly private, but users would be able to maintain contacts among the different sites within their own respective networks. Microsoft is not necessarily bridging networks and creating new paths, but they are making existing networks more open and powerful.

Throughout this course, we have discussed the internet as an amazing tool for connecting people across the world, which “has made possible entirely new forms of social interaction, activities and organizing, thanks to its basic features such as widespread usability and access” (http://en.wikipedia.org/wiki/Internet). As the most frequently visited websites, sites such as Google, Yahoo, MySpace, YouTube and Facebook have become increasingly useful tools for the spreading of information and ideas. What we have failed to discuss in class, however, is the effect of censorship on such sites, or censorship on the internet as a whole.

Recently, shareholders of the internet search giant Google have proposed a new policy which would allow access to the website to all users regardless of local laws; hosting data in a country where free speech is protected so as not to directly violate domestic laws, but not to participate in proactive censorship. In Turkey earlier this week, Google owned YouTube was banned because of video clips that were deemed disrespectful to Mustafa Kemal Ataturk, the country’s founder. YouTube has also been banned in the past by the governments of countries such as Pakistan and China for hosting anti-government content.

In the case of a single idea or a single video, for example, these restrictions effectively remove all edges to individuals in those countries. The question is raised: do these websites have a responsibility to make this content available to everybody, or should they respect the bans of these governments? The following articles document these stories, as well as Google’s point of view.

Google and the threat to free speech

http://technology.timesonline.co.uk/tol/news/tech_and_web/article3634123.ece

Google investors seek censorship ban

http://www.mercurynews.com/google/ci_8700662

YouTube Removes Videos Banned in Turkey

http://ap.google.com/article/ALeqM5iKUx9hP8rzGIKGJC5_Ml7OViYraQD8VLS25O6

Wikipedia Entry for Internet censorship

http://en.wikipedia.org/wiki/Internet_censorship

http://www.technologyreview.com/Infotech/20152/

On January 30th, 2008, two undersea cables connecting Europe with Egypt and  the Middle East with India were cut.  These cables were fiber-optic cables providing Internet access to and from the regions.  Immediately large portions of the Middle East and India were without Internet access because these cables were bottle necks for Internet activity.  The article states that undersea cable damage is not rare, more than 50 repairs were made in the Atlantic last year, but Europe and America are connected by many different cables so damage to one does not largely affect Internet activity.

The Internet is a great example of a real network used by billions of people in the world.  The nodes can be computers or servers and the edges are the wires (or signals) that connect the nodes.  This article describes a part of the network where the underwater cables are probably local bridges.  In fact they are very important local bridges, connection huge chunks of the network (in this case countries) together. When the cables were cut, the main connection to the Internet for these countries was lost.  The cables are only local bridges because there are other connections, but they are not as widely used, older, and not as fast.  This is an example of how important it is to have multiple connections in networks in case a connection is lost.  Even in many other applications of computing it is good practice to have repetition in case an important part of the system goes down.

In class we discussed how online search engines make high revenues through keyword based advertising. From the advertisers’ perspective, this form of advertising is an excellent way to ensure that their ads are viewed by users who have expressed a similar interest through their online query (word search). However, it is not always the case that a user is truly interested in viewing those ads nor is it a certainty that a user who has clicked on an ad intended to purchase the advertised product (he might have clicked on it by accident). In both of these cases, this form of advertising fails to meet the advertisers’ needs.

Yesterday, YouTube added a new feature called YouTube Insight which makes it possible for users to view detailed statistics about the videos that they upload to the site. An uploader will be able to find out where the viewers have come from, and exactly when they watched the video.

In the context of advertising as a matching market, the advertisers want to be assigned to slots which will maximize their total valuation. In class, we assumed that advertisers know the clickthrough rates associated with each slot, but we ignored the mechanism of exactly how this is possible. As the New York Times article points out, the tools provided by YouTube Insight will greatly benefit advertisers by assisting them in determining which audience groups they should target. In other words, advertisers who care how many clicks their ad will receive, can accurately match themselves with the appropriate slot.

For example, a band may upload its music performances, determine which states it was popular in via Insight, and then plan its tour around that. Similarly, marketers can upload several versions of a movie trailer, and subsequently determine which version is catching on where. Based on this information, they may choose to run the trailer as a TV ad in a particular geographic location.

Referenced Articles:

http://www.nytimes.com/2008/03/27/technology/27youtube.html?scp=2&sq=youtube&st=nyt

http://www.redherring.com/Home/24035

http://www.youtube.com/blog?entry=IRJjhiDz6RU

http://www.ft.com/cms/s/0/3fc8f6dc-1ff0-11db-9913-0000779e2340.html

http://www.techcrunch.com/2007/03/21/digesting-googles-new-ppa-advertising-product/

The past several weeks we have discussed the internet and the implications of link analysis 

and the ranking of sites. In today’s extremely volatile stock market, the search engine giant 

Google stands at as one of the lone bright spots in an incredibly shaky economy. As discussed

 in Chapter 10 of our text, internet advertising has become a multi-billion dollar industry and 

the primary source of revenue for Google. Net advertising is a topic very closely tied to the 

central themes of this course because it combines elements of both social and information 

networks. As Google rose above the pack of a multitude of startups in the early 2000s, 

it was a strong advocate of “Cost Per Click” advertising or CPC. This (up until about 2006)

 was seen as the most advanced and efficient means of online advertising. The clicks were 

considered an action of intent and therefore communicated a specific interest. However, 

as competition heated up in online retail, companies started using the malicious tactic of 

click fraud to drive up the cost of advertising for other companies. This occurrence is 

outlined in detail in the Financial Times article I have listed. As a result, a more efficient

 means of gauging customer interest and advertisement pricing was needed. Just recently

 Google launched a new product to counter this click fraud. This new attempt at online 

advertisement is outlined in the Tech Crunch article posted above. This new style is called 

Pay Per Action which charges an advertising fee each time a customer actually places a

 product in a shopping cart. This system is a good way of ridding sites of click fraud and

 better regulating advertisement pricing. However, as the Financial Times article points 

out the company that develops the most modern and advertising strategy would gain a 

serious advantage in the market. This forced me to tink back to my first blog post 

where I was examining the “gatekeeping” properties that Google exhibited. Just as

 before, because of their level of market share, Google sets the tone for all internet

 advertising. Net advertising is one of the most complex services to valuate. This 

makes me question the extremely high rate at which Google’s stock is valued. With 

these constantly changing modes of advertising and complex pricing valuations it would

 not surprise me if Google’s bubble soon bursts. With its constant innovation Google 

will either continue to dominate the web or suffer the consequences of over valuation. 

Either way these articles point out how much of a role the ideas we discussed in class 

play in the modern business world.

A couple of years ago, I remember reading a blog that reported that BMW’s german site had been banned and its PageRank reset to 0.  Google was punishing the company for exploiting its search algorithm. It was found that BMW was using a “doorway page” for many of its sites to artifically increase its ranking. A doorway page is usually a single screen filled with keywords that a crawler, such as Google’s Googlebot, can read. Often times, they also include javascript code that immediately redirects viewers to the “actual” page, but is something crawlers ignore. Google’s Matt Cutts shows what BMW was doing on his blog. BMW responded a few days later by removing the doorway pages and issuing an apology.

You can see similar exploitative techniques on youtube, where search results are driven by tag words. Often times if you look at the most viewed section of videos, you’ll find videos tagged with dozens of completely unrelated tag words. Some even go as far as editing in random screenshots into the video so that they can use it as their snapshot of the video when displayed to viewers.

The internet is the primary source of information and news to millions of people, so it should be of great concern how companies and wesbites are exploiting the page ranking algorithms of search engines. I don’t know about every else, but rarely do I ever click past the first page of results when searching through Google or any other search engine.  Am I really making my own decisions when deciding which web sites to visit or has that already been decided for me by “optimized” web sites and their marketing consultants. Exploiting search engines really takes the social aspect out of the web network. I would assume the first hit I get on a search is ranked #1 because it is popular with other people, when in reality it is just the most optimized to the search engine’s ranking conditions.

http://opus.zbw-kiel.de/volltexte/2003/974/pdf/dp445.pdf

After searching for a long time for an interesting article relating to information cascades I found this. This article discusses the underlying reasoning behind mass emigration from one area to another. It takes an approach similar to the one we took in class going through the basic reasoning and the calculations in an effort to prove that these information cascades are actually rational behavior even though they can, at times seem like blind followings. The article talks about how if people from one area start emigrating somewhere, the other locals will think that the emigrants had some knowledge that they themselves don’t have about why it would be beneficial to emigrate even though their family might have lived in the same village for centuries. Thus their own signal is telling them to stay and live where their ancestors before them had, but oftentimes if enough locals have emigrated to some country they will believe that their is enough reason to leave everything behind. As you can tell this mentality will create a sort of landslide or, emigration of the masses. It usually lasts until This idea was very prevalent in the mass emigrations seen throughout history, such as those from Europe to the United States throughout the 19th and early 20th centuries. Also, this can be related to the recent illegal immigration problems in the United States, where immigrants from Mexico, based on the decisions of past immigrants, believing that those immigrants were more informed and thought they could have a better life, and better job in the United States, decide to immigrate illegally themselves.

There is one point in this article that I would like to disagree with, though. The decision of emigrants to move to the same country is a strong indication of an Information Cascade, but the fact that they tend to live in the same town as fellow emigrants in their new country can be explained better by their personal preferences and not because of the information cascade. Oftentimes they do not speak the language of the new country well or the country has very different customs. Therefore, they would feel much more comfortable living around others that speak the same language or have the same customs. This leads me to believe that this is not a result of the information cascade that got them to emigrate to the country in the first place. This assumption may hold for herd behavior, but not of the specific definition of an information cascade. The reason to live near fellow immigrants is almost purely a result of personal payoffs.

There is one other thing I would like to note about information cascades that does not relate to this article. I was originally looking for an article relating the rapid growth of many religions in the past, such as catholicism to information cascades, but could find nothing. It seems that many religions began, and still expand through information cascades. People make decisions to join the religion based on the belief that others that are already members of the religion know something that they do not, or are more informed. This information is taken to be knowledge of a higher being(s). This is how most religions began, through religious leaders, such as Jesus, that professed a knowledge of things that others did not know about, creating an information cascade. This can be linked to the the rapid growths in history of certain religions. Information cascades are a very interesting phenomenon and it is interesting to see all the research relating to them that closely follows what we have learned in class, dealing with all of the math and probability behind it.

http://www.usatoday.com/money/advertising/2005-06-22-viral-usat_x.htm

This article talks about a popular way of advertising called viral marketing. This technique has become increasingly more popular in recent years. Viral marketing is a way of advertising that uses already existing social networks to increase brand awareness. It is spread in a viral process, similar to pathological and computer viruses, usually by some form of communication . It could be spread by word of mouth, the Internet, text messages, and so forth. With the advent of the internet, viral marketing has been able to spread across social networks very quickly. Viral marketing relies on the consumers to pass on the advertisements to friends, so it is important that companies make their ads catchy and creative so that others will pass it along.

Most recently, movies such as Batman Returns and Cloverfield have taken advantage of viral marketing. Their trailers flooded the internet, and has been spreading to more users very rapidly. One good example of viral marketing is Hotmail. Hotmail only spent $50,000 on traditional advertising and they were able to gain 18 million users in 12 months. Hotmail’s early success can be attributed to users telling their family and friends about their service. They also used viral advertising by having advertisements embedded into each email.

http://www-personal.umich.edu/~ladamic/papers/viral/viral-market-short.pdf

This article goes into more detail about how viral advertising based on recommendations relates to information cascades. Here in this model, one person makes the decision whether or not to buy an object. Subsequent buyers can read the recommendations of previous buyers on whether or not to buy the object. Here we see how people’s decisions to buy a product will be affected by what other’s say about the product. This recommendation system is a slightly different information cascade since only those that have recommended the product have purchased it. Sites like Amazon offer interesting statistics like how many people that viewed the product actually purchased it. So we know the decisions of what other people made, but there is no clear signal like in the model presented in class, where people received a signal of High or Low. The paper also notes that as a person gets more and more recommendations for a product, the probability of purchasing that product increases; however, there is a saturation point, where subsequent recommendations beyond the saturation point have no longer increases the chance of purchasing. This shows how there are many variables in recommendations that can affect the probability of purchasing something.

In the March 4th, 2008 issue of the Cornell Daily Sun, columnist Ben Birnbaum wrote an article entitled The Audacity of Hype. In the article, he discusses the trend, especially of the younger generation, of supporters of Senator Barack Obama for the Democratic nomination to lack any substantial reason to support his nomination. The author even quotes an interview with a Texas senator. When asked why the senator supported Barack Obama, he could not give a valid reason. So why do so many people support a candidate if they have not done the research to find out if he is, in fact, the right person for the job?

To what this all boils down is an information cascade. When you look at the long list of Obama supporters, which include John Kerry, Chris Dodd, as well as Oprah Winfrey, it is hard not to think that they know something about him that you do not. So, like many young voters, you disregard the small amount of information you’ve researched and believe that since so many have supported Obama, they cannot all be wrong. This happens because the individuals who support a candidate see the choices of supporters who have made decisions before them. As an example, say Oprah Winfrey and John Kerry are the first two Barack Obama supporters. If Oprah supports Obama, one can deduce that she must have received a high signal. If Kerry supports Obama, he must also have received a high signal. Then, the information cascade starts, and there is a higher probability that Obama is a good candidate regardless of what signal the third person receives. To be fair, this occurs with any candidate, but the article speaks specifically about Barack Obama.

The Cornell Daily Sun article can be found here: http://cornellsun.com/node/28481