Category: Readers

In Mejias’ chapter on “Computers as Socializing Tools” he sites the critique of social computing, a field focused on the modeling of social behavior, that in some way it simplifies human behavior by “dumbing it down” to a level which a computer can understand it. The creation of artificial intelligence or games like the Sims demonstrate attempts to recreate human life in the digital realm. These attempts so far have fallen short in recreating a 100 percent realistic picture. Therefore, the question remains can computerized algorithms, however complex, ever truly model how we behave? The answer depends who you ask. Critiques such as those mentioned by Mejias might challenge the plausibility. However, some physicists may argue otherwise. Theoretically, if a unified theory of physics such as string theory were ever to be proven, any concept of free will would be nullified. This would mean our brains operate no differently than computers, collecting stimuli and working through algorithms to arrive at an appropriate action. If this is truly the case, then there is no reason computers could not accurately model human behavior. The only obstacle to this goal would be understanding the complex algorithms of the human mind, which is certainly a daunting task.

David Easley and Jon Kleinberg’s chapter on graphs provides an instructional background on the mechanics of visual aids. Their explanations on distances between nodes, paths, and connectivity, however, read as overly specific. Developing “basic network properties in a unifying language” (23)  for graphs sounds unnecessary to a common reader because the very point of graphs is to make information accessible without the need for complicated directions. Visuals are so prevalent in our daily life that directions are unnecessary. Even seemingly private information made available to us with visual aids, like our Facebook data, is immediately digestible. Reconsidering how graphs are established, however, allows us to find similarities between visual aids that may be unexpected.

Through their discussion on social media, Easley and Kleinberg hinge upon a link between visual maps and connections with friends online. They introduce a concept named the “small-world phenomenon— the idea that the world looks “small” when you think of how short a path of friends it takes to get from you to almost anyone else” (35). The Guardian’s article on internet privacy opened our eyes to how closely linked humans are- even people who don’t know each other. I would argue that other graphs also work to shrink distant locations, and apply the idea of social network analysis to physical geography. Easley and Kleinberg use airline routes and subway maps to illustrate the graph’s ability to document “direct connections.” Shrinking one-hundred city blocks into twenty orderly dots makes an urban environment look compact. I can travel across an entire city by making a few stops along this lined? A flight from Atlanta to Tokyo, a fifteen hour flight across an entire ocean and continent, can be reduced to a two-page spread in a magazine. That journey looks pretty simple. As previously mentioned, graphs are incredibly accessible and benefit from not needing the mechanical directions provided by Easley and Kleinberg in their chapter. One thing graphs lack, however, is a sense of context. While they may be easy to read, their simplicity may imprint “the small-world phenomenon” on geography as well as people. Accessibility comes with a cost; whether or not such a cost is positive or negative is up for debate.

Through technology, we choose who we associate ourselves with. The association with someone would create an edge between nodes. The edge forms a network that connects two nodes together where the nodes of family, friends, and associates are now connected to both nodes. For example, Facebook is a digital network where two people can become friends on Facebook. Now, the friends of these two individuals are connected through a path that are formed by two individuals, unless the friends already have mutual friends. Ulises Ali Mejias states in Off the Network how “to ‘Friend’ in a social network that establishes a correspondence between two records of data (47). Every day, Facebook users are being connected together as one can find Facebook to consist of a giant component that encompasses every Facebook user being connected through edges. Even if one creates a new Facebook account, then they will soon friend someone else that will tie them into the giant component. Yet, the downfalls behind the giant component of Facebook include disruptive forces like a person impersonating a real-world friend that could disrupt the connection between you and your friends.

Furthermore, through Facebook, we witness various Facebook users discover their friends to have hidden mutual friends that may have not been discovered through technology. David Easley and Jon Kleinberg states this idea, small-world phenomenon, where the world looks “small” when you think of how short a path of friends it takes to get from you to almost anyone else (35). The downside from this phenomenon is it doesn’t mean you’re socially close to them.

David Easley and Jon Kleinberg, “Graphs” from Networks, Crowds, and Markets: Reasoning about a Highly Connected World (2010)

Ulises Ali Mejias, “Computers as Socializing Tools” from Off the Network: Disrupting the Digital World (2013)

In Cartographies of Time, Daniel Rosenberg and Anthony Grafton illustrate how easy visualization of data can facilitate a deeper understanding of its meaning.  They take care to introduce and detail multiple different different data recording systems from the last several centuries, explaining the nuances and distinctions among each of them.  The evolution of these charts helps illustrate the relative importance of chronology as a field of study.  It seems, however, that while these charts are immensely detailed and intricate, today’s charts are considerably less so.  A timeline is no longer a revolutionary new method of visualization but more likely a homework assignment for a second grader.  These charts have grown simpler over time, resulting in the standard line graph, something that was previously innovative but is now mundane.  This lack of detail and intricacy points to an argument that Mark Monmonier makes in his book How to Lie with Maps, that exclusion of data and distortion of reality in a chart is necessary to get the most important points across.

A message that both books present is that simplicity is an important aspect of data visualization.  Clarity is key.  Unfortunately, the act of simplifying allows for the possibility of manipulation, which Monmonier details in his accounts of political propaganda maps. Monmonier alludes to the point that simplification of maps is dangerous because people trust them.  Maps are often awarded more authority than they deserve.  Take, for example, the following map:

This map means a lot of different things to a lot of different people, and is a frequent source of international tensions.  The 9 dashed line (China’s claimed territory in the South China Sea, in red) was originally published in 1947 by the Koumintang, not the Chinese Communist Party.  In 1949 when the Koumintang fled to Taiwan and the CCP founded the People’s Republic of China, the CCP conveniently adopted this map as their own despite it having been created by the Koumintang, who also maintain claims to the area.  The CCP continues to insist upon their right to claim the entire South China Sea to this day, despite the fact that this violates mutually agreed upon international law.  Some might find this map to be completely legitimate while others find it outrageous.  It stands as an example of how lines drawn on a picture of the earth can represent something so incredibly significant as national sovereignty, and how oversimplification of maps can have potentially dangerous ramifications, especially when those maps carry authority.

A common method that coaches tell their players in basketball is to visualize their shot going into the hoop before the action takes place. This is a trick that can help a shooter’s confidence. But are their other ways that sports data can be “visualized” for human use?

This week’s reading by Lev Manovich analyzed the usefulness of data visualization. Manovich provides a rough definition of infovis “as a mapping between discrete data and a visual representation.” The question that arises out of this definition is how accurately does a visual embody the raw data? The answer varies. Manovich mentions data reduction as well as the use of special variables as two main characteristics of infovis. These methods effectively smooth the data in order to represent key aspects to the viewers.

Although useful, I think that these two characteristics of visual analysis can have a tendency to digress from a dataset’s overall meaning. They do offer the viewer an opportunity to make sense of the datapoints, but by omitting aspects, the individualized data could lead to different results. For example, a graph can show a player’s field goal percentage in several games, but the analysis will leave out variables such as whether the shot was contested, minutes played, and the location of the shot.

A new system of NBA tracking has been implemented that tracks the movement of players 25 times per second. This new form of data allows for far more advanced statistics concerning touches, rebound opportunities, drives, and catch and shoots. The system also allows viewers to see video as well as movement animations to form a more complete level of analysis. Manovich would characterize this type of data as “direct visualization.” In this fashion, the NBA tracking system uses actual images and video to make a new depiction of the data. Will this new way of visualization change the way players and coaches think about stats? No way to tell now, but I encourage you to check out the site.

The digital humanities face a variety of challenges as Big Data becomes more essential to the field. One of these difficulties is the broad skill set that a digital humanist must acquire to organize and analyze these large data sets. Lev Manovich states in his chapter Trending: The Promises and the Challenges of Big Social Data, that most social and humanities researchers do not have the skills in computer science, statistics and data mining needed to take advantage of Big Data. There is a divide between those interested in the social implications of Big Data and those who can properly analyze it.

Typically when one thinks of a humanities or anthropology student, they do not associate computer science or statistics classes within their course schedule. A math or computer science student is typically not required to ask humanities questions in their studies. Although I know little about the humanities program at Davidson, I’ve found that there are two course sequences offered, The Western Tradition, and Cultures and Civilizations. These courses “encourage and reward clear thinking, speaking and writing.” Perhaps they could also benefit from computer science and data analysis skills that will be important for future social science discoveries.

From our conversations with Micki Kaufman, we learned she took a very non-traditional route to gain the skills see needed to become a digital humanist. Many of the skills she acquired over time through practice and experience with computer programs came not from her History studies at Columbia, but from the necessity to learn these skills so she could pursue her interests. The changing experiences of humanities with the advance of Big Data may require a rethinking of how students interested in these fields are educated.

Image URL: http://www.bc.edu/libraries/newsletter/2012summer/humanities/starter.jpg

One point that Lev Manovich makes in his chapter titled Trending: The Promises and Challenges of Big Social Data that I found particularly interesting was his second objection to the “collapse of the deep data/surface-data divide” regarding the authenticity of interactions in social networks and the image presented by people online. He states that what people post and their participation online may not be an accurate or exact portrayal of their true personality and behavior. That may seem fairly obvious on the surface; however, when choosing to analyze and explore the vast amounts of data that can be accumulated through tweets, Facebook, Flickr, and Instagram photos, social network posts, etc., Manovich suggests that we bear in mind the data should be considered an “interface people present to the world,” not a clear view into people’s ideas, thoughts, and actions, and analyze the data accordingly.

The extent to which people curate their public persona online varies depending on the site, from a likely large extent on dating sites to a smaller extent on Twitter. But modifying our public personas online to such a large extent in some cases, I believe, mitigates one main purpose of social media, which is, according to Alice Marwick’s article, saving and exploring data about ourselves and others. Presenting fictional or exaggerated images or posts does not create the same level of discovery and connection between yourself and others on social media as your real images and posts do. Also, we may still have control over what we post, like, tweet, and so on, but companies and the government can still glean useful data from our cultivated online personas.

PhotoURL: http://www.someecards.com/usercards/viewcard/MjAxMS1mOWU1ZDk5OWU1MWQ3MDFj