Increased Corporate Ownership of Data

As a result of increased corporate ownership of data, we are living in an era of Internet/social media addiction. We see this clearly demonstrated by the behaviors of the characters in The Circle, but is best explained by Mae’s ex-boyfriend, Mercer. “The tools you guys create actually manufacture unnaturally extreme social needs. No one needs the level of contact you’re purveying. … It’s like snack food. … They scientifically determine precisely how much salt and fat they need to include to keep you eating. You’re not hungry, you don’t need the food, it does nothing for you, but you keep eating these empty calories. This is what you’re pushing. Endless empty calories, but the digital-social equivalent.“ (134) In other words, what Circle creates isn’t something the cyber citizen necessarily needs, but something addictive. Although The Circle describes a fictitious and futuristic society, it is frightening that the characters’ behaviors resemble our actions now. Nowadays, we rarely separate ourselves from our phones, using them and our media outlets as a way to post and share our every move with our Facebook and Instagram networks. Whenever a significant event happens, like the Presidential Election, everyone suddenly becomes an expert on the subject, sharing their knowledge and insights on Twitter. Sharing Economy is the modern version of neo-panopticon, since everyone is watching (or at least capable or watching) everyone else under the rating metrics in this highly connected and autonomous marketplace. In a nutshell, the emergence of new ideas enabled by new technology has provided us with seamless simplicity and convenience, but has also deeply shaped our social behaviors and online culture. The development of technology and sharing-economy is so rapid that it’s impossible to imagine what will unfold in the next decade.

When the data self overtakes the embodied self, we lose the control of how we live our life. Unfortunately, corporations own the data and from that, they want us to behave the way they desire, mostly the “unnatural” way.

How Connected Are You?

For our observation this week, we asked all of our classmates to send us the following lists:

  1. 15 of your friends at Davidson
  2. 15 of your classmates this semester at Davidson
  3. Every professor you have had at Davidson

Our goal was to create a network graph that shows the connections we have through the people in our classes.

We have 4 visualizations to accompany these lists. There is one network graph for friendship:

friends

One network graph for classmates:

classmates

One network graph for professors we’ve had at Davidson:

professor-1

Note that in the overall graph below that the edges indicate either friends, classmates, or professors (or all of the above?):

overall

Here is the link to the Google drive folder that contains our .dot files and pdfs so you may zoom in closer to examine the graphs.

**** Updated ****

Visit this link for a better graph layout: http://truthexistseverywhere.com/network_dig210/index.html

professorsoverallfriendsclassmates

 

A Universal Map

The creation of maps, just like the creation of other data visualizations, involves distorting reality and telling “white lies”. Mark Monmonier, in his book How to Lie with Maps, discusses the misuse of maps, the appropriate use of maps and the nature of maps. Our assigned reading includes Chapter 3, which explains the various techniques used and choices made when cartographers are creating a map, particularly because these become the “white lies” the maps tell.

 

The first essential part of creating a map is the selection of features for the map. However, equally as important are the features that the cartographers chooses not to include on the map. In the area you want to map, there could be multiple points and areas that could be included. Due to spatial limitations, only one of these areas may be included. After this, there are four important generalization operations – simplification, displacement, smoothing and enhancement. These are the steps taken by the cartographer to simplify the map and make it a usable visualization, depending on the purpose of the map.

 

So what is the “white lie”? “White lie” is a result of generalization, a combination of what’s included and what’s not. According to the reading, “a good map with ‘white lies’ suppresses truth to help the user see what needs to be seen”. One most common usage of map is for directions and geographic relations from place to place. On a road trip, the driver cares most about the overview to the destination; the overview of the cities to pass by thus forms the “white lie”, which tells the driver not to focus on the detailed road map of any city on the way. The ability to adjust the scale of brevity and succinctness according to user’s need is what makes maps such fit visualizations of geographic data. However, maps can be used to generate “white lies” on more than just geography, but also potentially any data based on geography. In figure 3.10 and 3.11 from the reading, Monmonier demonstrates how the same data on the same area can yield different-looking choropleth maps with different sets of class breaks. In the pictures, color choices and the area remain the same, but the class breaks for each color are different; while all four maps provide sufficient information as the legends are clearly labeled, it is very misleading to the audience at first glimpse because the color distributions look different. After all, the four maps tell the same information but our eyes tend to convince us not. The manipulation using color to deceive our eyes is considered a “white lie”, which in this case is for political propaganda. Impressions change with perspectives, which is a common trick of manipulation in data visualization.

 

We found 3 different cartographic representations of Davidson College, all of which display the correct layout of buildings, but also include the “white lies” discussed in the Monmonier reading.

Screen Shot 2016-10-10 at 11.09.11 AM.pngThis first map is a map taken from Google Maps. It shows the main roads and only one building on the Davidson College campus. This map chooses to use different widths to demonstrate the business of the roads–the streets on campus are much smaller because there is less traffic. This map would be helpful for someone looking to approach the campus, but not for someone who wants to walk around campus or someone looking for a particular building on campus.

screen-shot-2016-10-10-at-11-08-07-am

This map gives a much more detailed representation of campus. It makes the choice to include a key rather than label the buildings individually. In fact, this is one of the materials included in the packet given out to the Freshmen class in the beginning of the school year to learn the campus. This would not be a good map for anyone coming from off campus, since the roads outside of campus are not labelled well or included.screen-shot-2016-10-10-at-11-10-13-am

This map is not drawn to scale and does not include any sidewalks or street names. This would be good for someone who is trying to get a general sense of the layout of buildings on campus. It is visually appealing to get an overview of the important buildings but it is not an accurate visualization.
All three maps above show the various mapping techniques in place. While they are all valid maps, some may be more appealing for different purposes than others. Does a universally good map exist? Are there some selection techniques that would be best in this case? How would you illustrate and map Davidson College?

Responders: October 4th, 2016

Data organization has been a common theme in our readings this week, and we see how such an organization can be especially powerful in real-life examples like the supermarkets discussed in Reader’s post on Monday. The layout of a grocery store is intentionally designed to manipulate customers and encourage them to shop around longer.

Lev Manovich discusses the transition from database, “a catalog of objects that does not have a default sort order”, to data stream, a form of organization prompted by the introduction of social media outlets like Facebook & Twitter. While database is static, data stream is dynamic. A key distinction between database and data stream is that: database is passive and requires human interaction to be effective and valuable, whereas data stream is simple, straightforward but active and initiates human interaction. In other words, data stream turns boring data into interesting information and brings it to our eye, effortlessly. Before the rise of the Internet, we read newspapers, a form of database; now, we go to Facebook or Twitter, because data stream provides the most up-to-date news automatically (in a way, it has somewhat made us “lazier” and “lazier” to explore information–tied up to another previous reading: <Is Google Making Us Stupid? What the Internet is doing to our brains>). In an interview, Mark Zuckerberg revealed that in the early stage of Facebook, he discovered that online users were most interested in knowing what other people were doing; from this client behavior, Facebook went on to implement the “News Feed” feature which is essentially a form of data stream that gives updates of people around you. In fact, social media almost always offers the latest events and updates, either in our friend group or globally, because we as consumers are attracted to the “new things”, which leads to more sales of advertisement.

Group 1 raised a question: whether we as consumers are subject to and influenced by marketing tactics from supermarkets; A similar question can be asked on social media. Just as supermarkets are designed to make consumers shop around longer, the layout of social media websites based on the philosophy of data stream encourages users to stay on site longer by having them scroll through a continuous stream of information and updates. As the writers in this Live Science article observe, (http://www.livescience.com/34649-why-internet-is-addictive.html) Facebook and social media sites are addicting because most of the content is junk with an occasional post that warrants a response. Without any clear stopping point in the stream of information, it is quite easy to stay online for hours on end. So is it possible that our online behaviors are not only monitored, but also manipulated by the “panopticon”? While it’s hard and unwilling to confirm, it’s also impossible to deny. From shopping at a supermarket to browsing latest news on social media, our behaviors have shifted more and more toward what the “bosses” desire them to be, without our awareness.

 

 

To Sleep Or Not To Sleep?

As the observers this week, Brandon, Connor & I sent out a daily survey to collect data on the sleeping patterns of the students (& Owens) in our class. The visualizations below show us the daily times that the students were sleeping:

 

Monday.jpgTuesday.jpgWednesday.jpgThursday.jpg

 

Here is a screenshot of our raw data:

Screen Shot 2016-09-29 at 7.15.22 PM.png

 

And a link to our data in Google spreadsheets here.

Do We Have Access to Too Much Information?

Lauren Klein, in her discussion of archival silence, provides an account of Thomas Jefferson’s letter composition process and the efforts Jefferson took to preserve these letters. More specifically, Klein discusses the scholarly impact that digitizing records has had.

The “epic transformation”, as it was referred to by Ed Folsom, of records from print to digital form has increased both access to records and the links between records. Computer programs have also allowed for more extensive analysis of the content in these archives. Databases are typically equipped with a search query tool, allowing scholars to quickly peruse thousands of records for information about a certain subject. Further, network analysis is made much easier. Klein includes a visualization of Jefferson’s interactions concerning James Hemings, which has an arc for each correspondence with a size relative to the frequency of these interactions. This visualization was made using Protovis, a tool that quickly creates network visualizations about a dataset.

Although archiving records can simplify the analysis of the records and increase the scope of questions that are able to be answered, there are larger implications to digitizing all records. Klein emphasizes that context is often lost. Specific to James Heming’s life, we can study the facts and figures that we gather from carefully analyzing the records, but aside from the data it is hard to know much about this individual (Klein 671). Although databases and archives allow us to quickly search and study information, do they give us too much information to the point that we lose sight of what these articles do not tell us? Can we be given access to too much information?

If digitally archiving records has increased the scope of the data recorded, it has also made it more difficult to conceive the data that is not recorded. While we can analyze this data closely, we tend to ignore the other perspectives of the story- the parts of the narrative that are missing or not recorded. We tend to grow our bias toward the stories that are existent. Nowadays, the media often faces criticism; it has a large amount of digitized contextual data but it’s up to the media to decide which of the data to reveal to the public. Sometimes it blurs the truth and sometimes it can be misleading. For example, in this article “Wikileaks Revealed Massive Political Corruption: Where’s The Coverage?” (http://www.huffingtonpost.com/david-seaman/wikileaks-revealed-massiv_b_11753560.html) David Seaman argues that Wikileaks had sufficient “evidence” on the corruption of the Clinton campaign, especially Clinton Foundation, but there is rare coverage on this subject by the media. Here is one response which counteracts the argument and the “evidence”:

screen-shot-2016-09-11-at-9-21-29-pm(Courtesy of @Calven McVetty)