Week 12 – Facial Recognition Software, Algorithms, and Espionage

In the article “Public faces? A critical exploration of the diffusion of face recognition technologies in online social networks,” Norval and Prasopoulou explore the use of facial recognition technology (FRT) and it’s implications for American society. Facial recognition software is diffusing into everyday life, and is not longer limited to a select few. The idea of privacy in public spaces is complicated. We want to be able to walk down the street freely, but don’t want the government to watch us and know who we are. We show our face to the public, yet your face is an intimate, personal part of you. Today, we don’t just show or face to public spaces, but also to online spaces, meaning that information can be stored.

In class last week, someone mentioned that snapchat has the potential to build a facial recognition database. Millions of users engage with their filters every day, and we consider it a fun and benign thing to do. Once you see your friend use the dog filter, you immediately do the same, without thinking of the wider context or potential security threat.  There are a few articles on the internet discussing the conspiracy theory that snapchat is secretly building a database. Snapchat claims their software is not facial recognition, but object recognition. The features can identify noses and eyes, but it can’t identify your nose and eyes. They say the software definitely does not store any information about your face. They also include in their privacy policy that they don’t share data with any third party. Although I don’t believe snapchat is building a database, the point is that a private corporation has the resources to build one. Features such as the snapchat filter are so normal today, that their users don’t stop and think about it before using it regularly.

It is not just private corporations that are of concern. The FBI has admitted to using drivers license and passport photos to build a database for law enforcement purposes. The Bureau worked with 18 states so they could gain access to the photos, but the citizens were not informed that their photographs were shared. Not only was that unethical, there are some systematic issues. The algorithm the FBI uses is not as advanced as those used in commercial industries, and it is inaccurate 15% of the time. The software misidentifies minorities at a greater rate than whites. The program is also subject to no regulation or oversight by Congress, and the citizens have little to no awareness it exists. The FBI defends the system by saying it is similar to a fingerprint, and simply automates a process the FBI has always done.

There are many significant implications to the use of facial recognition software by the government.  It is difficult, if not impossible, to imagine a state under constant surveillance as a free state. Yet that is where the technology is headed in the United States. Of course, the use of this software increases law enforcements ability to recognize criminals and terrorists, possibly creating a safer world. However, from my point of view, it is not snapchat nor the government that I am afraid of obtaining these databases and algorithms. I believe the biggest threat is an espionage attack.

Any information that is online, compiled and organized, can be broken into. The true threat is if this information falls into the wrong hands. The government and companies who compile this information must ensure they are investing in the proper security for these databases. In addition, privacy policies must be clear because nothing stops companies from selling user data to data brokers or other companies. Thankfully, privacy policies are currently under scrutiny. Yet, we know that we can forget certain events just as fast as we paid attention to them. The conversation must continue because as new technology and practices emerge, the privacy threat increases.

 

 

Week 11 – Infinite Distraction

The book Infinite Distraction by Dominic Pettman takes a critical cultural perspective of social media. While Pettman carefully avoids an overly negative tone, the book does highlight the negative ways social media has impacted society. Today, each user of social media has a different experience than the other. This uniqueness caused what Pettman calls hypermodulation. While one person is upset reading about a recent school-shooting, another person is watching a silly cat video. And then a few hours later, they swap. Thus, no two people have the same experience or emotions at the same time. For each individual who uses a social media account regularly, it is very difficult to use it appropriately. No one logs on, does precisely what they planned to do, and then logs off. The sheer amount of information, and the algorithm that knows what interests specific users prevent purposeful and precise behavior.

mark.jpeg

Even Mark Zuckerberg admits Facebook can be a distraction for many people. In this post about the new Facebook algorithm, Zuckerberg writes, “The research shows that when we use social media to connect with people we care about, it can be good for our well-being … On the other hand, passively reading articles or watching videos — even if they’re entertaining or informative — may not be as good.” Of course, Zuckerberg spins the business side of Facebook in his statement. Facebook is used by businesses to promote their businesses by targeting customers with ads specific to them. However, as Facebook users noticed such an influx of advertisements on Facebook, which keep the company afloat, users have begun to spend less time on the app. Facebook had to respond by decreasing the number of ads the algorithm is displaying to users and increasing the amount of content from their friends. Zuckerberg claims the change was so that Facebook can continue to promote “the heart” of what the app does. However, Facebook is simply trying to maintain more users and to encourage them to spend more time using the app. As Pettman puts it, we are “slaves to the algorithm.”

Social media platforms and their algorithm have powerful effects on the users. Instead of a family sharing the television screen at night, the family is dispersed throughout the home each engaging in a different, but most likely online, activity. Yet, the content and the powers that control the content most people see is strikingly similar. Despite these micro-experiences, internet users encounter many of the same shows, memes, tweets, and online videos. Pettman is concerned about the emotions users feel as they navigate from one story to the next. If people are flooded with so much information of varying sentiment all at once, it is unclear how individuals react emotionally. Pettman is concerned that among the videos of cute pets, the horrible stories don’t get the appropriate attention and emotional reaction they deserve.

Facebook controls the content users see when they are distracted. However, many people use Facebook and other social media to read the news. Pettman says media users should be more mindful and to be sure to exercise their ability to think to prevent becoming passive. In today’s media literature, many authors fear the destruction of individual liberties and the increasing power of corporations. Facebook may control our attention online, but I am hesitant to agree with Pettman that they control our entire lives. Of course, I believe users must be mindful of how they allow technology to affect their personal lives. However, I believe that people are more mindful and attentive than books such as this give them credit for.

 

Week 10 – Participatory Culture, Community Play, and Learning From Reddit

In the book “Participatory Culture, Community, and Play: Learning from Reddit” by Adrienne Massanari, the author takes a unique approach to studying emerging media. Massanari used existing literature, informants, and her own contextual analysis of Reddit to further understand the website and participatory culture. Ultimately, the author concludes that while Reddit comprises the best and the worst of the internet, the company needs to do more to moderate the bad aspects of the site. Massanari writes, “It is frustrating that Reddit, like so many other online spaces, creates a discursive environment where bullying, hatred, and bigotry are tacitly accepted in the guise of “free speech.” While most sites want to remain neutral towards user-generated content, there are some areas where sites have to act, such as deepfakes.

Reddit is a place where information spreads easily. A very troubling example is deepfakes. Deepfakes are fake videos created with artificial intelligence. People can replace faces of people in videos, and they can look very realistic. At this point in time, an app does exist, but it is time to consuming to use, and the videos look amateur. Skilled professionals or experienced amateurs must create the video for it to look real enough to fool someone. There are multiple online tutorials on YouTube. Deepfakes are typically associated with celebrity porn videos. Fake porn videos have been made with Taylor Swift, Michelle Obama, Ivanka Trump, and many more. However, deepfakes have been made in other contexts, such as entertainment. On February 7th, 2018, Reddit banned the pornography deepfake subreddit. Other websites, such as Twitter and PornHub have also banned deepfake pornography. However, the sites have not banned other deepfake videos.

It is hopeful that Reddit and other sites took a stand against these videos because of their potential implications. Of course, the videos still exist throughout the internet and will never disappear. Deepfake videos could cause an endless amount of harm. This day in age, we use videos to show that something actually happened. It is a very powerful method of truth-seeking, verifying facts, and destroying confusion. If someone was caught on camera committing murder, as a society we would not question whether or not they did it.

The scariest part is that deepfake technology can only improve. While it takes a certain technical know-how and free time, in the future this may not be the case. It is impossible to know how society will respond to deepfakes in the future. Photoshop has existed since 1987, and while we have not quite figured out how to mediate negative effects of photoshop, we can typically determine if a photo has been altered. Fake news is already an immense problem in American culture, but fake videos? Thankfully, Reddit has taken a stand against them, despite their assertion that they wish to protect the first amendment, even though they are a private company. However, sites could ban deepfakes while remaining fairly content-neutral because the problem is the method of video creation, not the content of the video itself. I believe it is critical that sites and users remain diligent and seek to discourage this type of content. Entire lives, careers, and families could be at stake. If deepfake videos become easier to make and more realistic, it could be very difficult to know the difference between a real video and a deepfake video.

Similar to fake news, deepfakes could threaten democracy. For our government to exist as it does, the truth is critical. Not only is it important for leaders to understand the truth, it is important for all voters to understand the truth. If people are constantly misled and lied to, it would become very difficult to trust any video at all. I know that if I was fooled by a deepfake video one time, it would cause me to be skeptical of all other videos for the rest of my life. Although deepfake videos have mostly been discussed in terms of porn videos, I believe they could get much more serious than that. Deepfake videos are powerful tools, and they highlight the negative outcomes of participatory culture.

Week 9 – Multitasking and Second Screening

In 2016, 46.7 million Americans used a device while watching television, a modern phenomenon scholars call second screening. Second screening is interesting because it does not deal with how multitasking diminishes user experiences, but primarily focuses on its positive effects. The majority of this type of conversation occurs through twitter. Viewers can connect with others who are viewing the show much easier on Twitter than other social networks because they can connect through hashtags. Second screening can benefit companies, however, they cannot profit from second screening conversations on twitter alone. They must employ less direct ways to profit and benefit from users conversations on Twitter.

The Guardian posted the article, “Where’s the money in social tv and second screening?” back in 2013. It discussed ways television companies could profit from second screening. One way is by adding twitter streams of their relevant hashtags to their websites so that they can profit from ad revenue. A second way is to develop apps viewers can download and use while watching the show. For example, the Oscars created an app that included exclusive, behind the scenes content. According to the same article, Twitter said: “40% of its evening tweets are TV-related.” Twitter certainly dominates second screening conversations and I am doubtful companies have been successful at moving the conversation to other apps.

However, I do believe there are ways companies can benefit from second screening. For example, I have seen Shonda Rhimes participate in second screening conversations. Viewers of her shows tweet at her very frequently, and she sends out tweets when they air on Thursday nights. Rhimes has 1.76 million followers, and more viewers engage with her through the hashtags and during the show. She utilizes many different hashtags to conform to in-group behaviors.  It is unusual for the writer of a TV show to be so involved in her viewers, and as someone who produces such political and controversial content, she is in a unique situation to engage with them. However, brands could use these conversations to build relationships with viewers, build credibility, and build a long-term viewer base.shonda tweet 2

The photo above is an example of Shonda Rhimes engaging in second screening conversations during the premiere of one of her episodes. This particular tweet received 311 retweets and 1,727 likes.

Second screening is also a free way for companies to view feedback. It does not necessarily replace other methods, but it does provide feedback throughout the entire duration of each episode. They could employ textual analysis to better understand their viewers and better understand the type of content they like and do not like. The company could then alter the show in response to what is happening on Twitter. For example, viewers responded positively to the episode of How to Get Away with Murder when Analise got ready for bed. I am sure that ABC and Shonda Rhimes took note of Twitters response and have since included more moments that reveal the experiences of black women.

However, there are many limitations to this tactic. It is not wise for a company to make extensive decisions based on the small population of twitter users. The opinions that are widely held by twitter uses and posters may not represent all the viewers of a particular TV show. While this data is interesting, it should not be the sole way companies make decisions about their show content. However, companies that are failing to pay attention to second screening conversations as well as partaking in these conversations are missing an important aspect of modern television viewing.

Week 8 – Updating to Remain the Same

Chun’s book discusses the point where new technology is no longer exciting but becomes part of individuals daily lives. It becomes a habit. She argues that it is important to study technology as it becomes a habit because that is when it silently infiltrates our lives. Her formula, “habit + crisis = update” frames the purpose of the book. Habits are routine behaviors, and crises are unique events that puncture habitual behavior. Crises can shift one’s course and allows individuals to be more adaptable to new technologies.

Chun’s book made sense at the beginning, but I struggled to connect the end of the book with earlier chapters. However, I do believe the underlying argument of Chun’s work was interesting, although it was difficult to follow at times. Her background in Systems Design Engineering certainly comes through, especially in the chapters modeled off coding language. The stories about bullying and slut shaming were examples of ways the internet is both public and private. In the case of Amanda Todd, she suffered from internet connectivity because she lost control over the photo. But to express her pain and suffering, she turned to online networks. I think Chun included this chapter to illustrate the complexity of how we relate to new technology and the place it plays in our lives.  It seemed like the book took a large shift and could have been better connected.

Her book addresses a serious concern present in most of the media literature we have read this semester. The way we interact with new technology is complicated — one moment we are grateful for its existence and the next moment we are cursing its power. It is impossible to approach media studies through a binary lens where we attempt to define the internet as either good or bad. It is also important to keep Chun’s work in mind when approaching network analysis. She writes, “Networks are odd entities: they are both technical projections and naturally occurring phenomena” (Chun 46). She is not demeaning of network theory, but she seeks to precisely explain what it means. She argues that networks are models so they cannot be completely the same as the “real life” network. She also argues that networks are just traces of our collective habits. Networks are always changing. The nodes and edges shift, and networks can spawn other networks. Therefore, by the time a researcher analyzes network data and publishes their work, it is “belatedly too early.”

Not only does her book contribute to media theory, but it also caused me to think about my personal media habits. Media is so connected to my daily routine, and it is important to be critical of it. We no longer view laptops as an interesting technology because it is not new, but it is such an integral part of all of our lives. I could not get my work done without it. In fact, we hardly think about our laptops until it breaks. However, it is not clear to me if Chun is suggesting research on habitual media. Should we look at how “older” technologies are incorporated into our daily lives?

Her book ends with a discussion about individuality and collectivity. Her point is important when performing network analysis. The crucial limitation of network analysis is that it gives information about the collective habits of individuals. The nuance between that is critical because the network is made up of individual, unique stories. My takeaway is the importance of triangulating your methods. Network analysis should be accompanied by qualitative studies and individual storytelling. Overall, her book was a unique approach to media studies and makes many interesting points. However, it is difficult to follow at times, and I think her charge at the end could have been more transparent.

Week 7 – Emerging Media Psychology

Media psychology is an important area of mass communication research because all of us are influenced by media in some way. Today, media messages are frequently transmitted through pictures and videos. We use images to invoke emotions in viewers that will motivate them to take an action or buy a product. However, the “third person effect” says that we believe mass media influences others, but that we are smart enough and not affected by media. We believe we are immune. In today’s world, we are bombarded by media messages to the point where it is nearly impossible to avoid them. Images of others living happy lives are everywhere. Every day we see images on print advertisements, social media feeds, and television that create intentional or unintentional emotions within the viewer.

Cultivation theory is another critical theory in media effects. Media scholars have built on cultivation theory since George Gerbner published “Living with television: The violence profile” in 1976. Cultivation theory is important because it is applicable to many different conditions. While it was originally concerned with violence on television, it is highly relevant to social media messages. I think it is even more relevant today because of the quantity of media messages individuals are exposed to.

The study “Picture Perfect: The Direct Effect of Manipulated Instagram Photos on Body Image in Adolescent Girls” by Kleemans et al., was a great example of the power of media. It is interesting that there is so little academic research on Instagram. I believe Instagram has significant power over its users because it is image and video based. Images affect us differently than text and possibly with more intensity.  Many of the participants in the study did not even notice the photos were altered. It is easy to portray a life publicly that is not congruent with your private life. In my own experience, people consistently overestimate the happiness of their friends on social media. We are so connected to each other’s public lives but increasingly more distant from each other’s private lives. Social media’s negative effects on mental health is also a widely studied area, specifically when concerned with anxiety and depression.

I also think the positive effects of media is a very interesting area of study. Many people use media as a form of stress relief. Video games are a very common way to use media for stress relief, but watching TV can also help reduce stress. This semester we have discussed many of the positive effects of media, specifically by creating online communities and building social capital.

However, from the studies that I have read, they focus on the positive or negative effects of media. How do we know the overall effects of media? Is it ultimately positive or negative? I think there is responsibility on the media consumer. Kleemans et al. suggested Instagram provide a disclaimer informing the user that some images may be manipulated or distorted when opening an account. They call this “visual literacy.” I am not convinced of the effectiveness of requiring such a disclaimer because it is unlikely users will read it or remember it as time goes on, but I did find the term interesting. It would be interesting if researchers could conduct studies to further understand visual literacy. They could look at demographic and psychographic variables to see if they could determine who has higher visual literacy than others and why. In a world where images are so present, and where they can have such harmful effects, it is very important to find effective methods to increase visual literacy. I have seen campaigns regarding photoshopped images in the past, but scholarly research in the area could reveal an unknown causation or interaction. Girls scrolling through Instagram may not realize the powerful effects the images have on them, and increasing visual literacy could help to mediate those effects, even if just by communicating the power of media on our thoughts.

 

Week 6 – Social Networks and Social Capital

When the internet was still new, most online groups consisted of individuals connecting by interest.  A lot of them did not know the people they communicated with online in “real life.” Today, social networks consist of individuals we mostly know in “real life.” But when it comes to politics, people on twitter connect with others based on party affiliation and who they agree with.

User-generated content is one of the defining features of social media. It can be very useful for political campaigns. The article, “Social Media and Citizen Participation in “Official” and “Unofficial” Electoral promotion: A Structural Analysis of the 2016 Bernie Sanders Digital Campaign” by Joel Penney analyzes the role user-generated content played in Sanders’ campaign. Many groups, such as “Bernie Sanders’ Dank Meme Stash” contributed positively. However, some groups, such as the “Bernie Bros” who were sexist and racist on social media. The “Bernie Bros” incident gave the caused negative publicity for the campaign. The official campaign stayed vigilant in watching the online content, but they did not partner or work with them because they feared it would ruin the humor.

During this weeks reading, a quote from the article “Political Rumoring on Twitter During the 2012 US presidential election: Rumor diffusion and correction” stood out to me —  “Debunking is dull compared to spreading an attention-grabbing rumor” (Shin et al. 1227). The focus of the article was revealing if rumor sharing decreased after fact-checking sites debunked the rumor. While there was a slight decrease, it was minimal.

Online political communities are more than participation in democracy or support for a political candidate. They also form online communities that help foster an individual sense of identity and belonging. One political candidate may have various subgroups that support them, but supporters may not feel like they “fit in” each group. Some Bernie Sanders supporters may not want to be part of “Bernie Sanders Dank Meme Stash” because they don’t think it is serious enough. I also found it interesting that rumor sharing was done mostly among strangers, but among people who have similar views. These people are connecting with one another because they believe in some cases was blatantly false, but the other person “gets it.”

Today, online grassroots networks are mostly unbounded and less-structured groups. When campaigns have many of these groups, it is impossible for them to control the image those groups project to the internet. If individuals are seeking something more than politics in these groups, such as group belonging or social capital, then the job of the digital marketer becomes even more complex.

It would be interesting to study individual motivations for joining political groups or online conversations, although it is likely the same reason individuals join any other group or organization. It would be interesting to look at social capital within political organizations as well. Do they share esteem support, informational support, and tangible aid when someone in the group has problems or does the group not bridge into their personal lives, mostly focusing on the candidate? I would hypothesize that it varies by group.

While politics should be about “hard facts” and policy, that is not the reality we are currently experiencing. While I am in no way saying facts are irrelevant or don’t play a role, while user-generated content and communities become a larger part of the picture, campaigns become about more. It becomes about humor and entertainment (BSDMS), affirming your already formed worldview (rumors), and communicating with other people. Whether that is good or bad is hard to say.

Week 5 – Affective Publics

Papacharissi’ argument in Affective Publics was very reasonable and thorough. She moves the conversation beyond the discussion of communities. Affect is crucial when studying online media. What happens online is greater than the tool itself. She recognizes that different movements, contexts, and situations cause the media to interact differently, and therefore have different results. Affect is not just feeling, but a reaction to events. It is a way for the citizens to feel like they are part of something, and it is a way for citizens to connect through that feeling. The internet is not inherently good or bad, strengthening or weakening. Specifically, her work helps to explain the complex political environment that exists on Twitter by looking at Twitter data.

Papacharissi’ work is the best explanation compared to the rest of the literature we have reviewed this semester because it connects the digital to the real world implications, including human emotion. Much of our political environment is mediated. She recognizes that this can either be positive or negative, depending on who is controlling the narrative.

The use of social media in authoritative environments is important and widely studied. It gives citizens an outlet to speak to one another when they previously could not do that. However, social media, specifically Twitter, has altered democratic political processes as well.

The 2016 election in the United States relied heavily on Twitter and social media. Hillary Clinton announced her candidacy with a short, 3-minute video, rather than in front of a large crowd. The Clinton campaign posted the video on their website, but it was not until they posted the video to Twitter when the internet began to really react. In this article, the New York Times presented twitter data, such as how many times Clinton was mentioned in a 2 hour period that afternoon. They stated that “On Twitter, Clinton is much more popular than the other two candidates who have officially entered the race to date, Republican Sens. Rand Paul (Ky.) and Ted Cruz (Texas).”

Donald Trump utilized Twitter very effectively during his campaign. In my opinion, Donald Trump’s campaign worked much like a marketing campaign selling a product. Although this time, he was selling the idea of American exceptionalism, and that he was the only person that understood his base and could fix their problems. Donald Trump was able to curate a lot of affect. His base (that used Twitter) were able to connect effectively through social media and the platform. They often used the hashtag #maga and shared memes making fun of Democrats and Hillary Clinton. Most of his ideas produced a lot of emotion and were easily translatable into simple memes.

One meme that circulated during Halloween showed two people in a costume, one dressed up as Trump, and one as Clinton. Clinton was dressed as a prisoner, and Trump was arresting her. It read “best Halloween costume yet.” Another showed a picture of Clinton and said “I am heavily guarded by armed men with guns. But the American people should only be armed with Pez dispensers and silly strings.” These memes cultivated an important sense of community among supporters who may or may not know one another, but believe in Trump’s message and culture. Memes like this exist on both sides, but I would hypothesize that they were more powerful on the side of Trump. The memes (for both candidates) avoid complex political discussions and zero in on humor and mutual understanding.

It would be interesting to study those memes. For example, who creates this type of content, and what roles do they play within the community? Which memes were the most popular duing which periods of the campaign? Was the opposite side part of the conversation? Twitter was important in the 2016 election, and Papacharissi’ provides greater insight into why.

 

Week 4 – Digital Inequalities

In the past, most researchers looked at internet access. Who had access to the internet and who did not? Why? However, recent research has delved deeper by asking what internet users are doing once they are online. In the article “The digital production gap: The digital divide and Web 2.0 collide”, Jen Schradie found a class-based gap where internet users of high economic classes create internet content more often. The article by Hoffmann et al. was particularly interesting because it showed the influence of cognitive factors, such as self-efficacy, also influence if internet users post content online. While we like to believe the internet is a democratizing force, scholarly research has shown otherwise.

While digital inequality is a problem for many reasons, I don’t find the results of the scholarly research surprising. The internet is a place for major companies and (smaller businesses) to make money. A lot of companies create content for economic purposes, and not to contribute to the “marketplace of ideas.” While libertarian theories regarding internet use sound tempting, it is not how our society operates. The United States values a free-market economy, and the internet is the same system. Consumer attention is viewed as a product, and content creation is often a great way to capture it. Larger companies, news organizations, or influencers have much higher readerships and viewerships than say, me. I can write all my thoughts and opinions pretty easily on this blog, but not many people are going to read it.

Sometimes we believe that gatekeeping theory is no longer relevant due to the development of social media. While it is certainly not the same, I would argue that gatekeeping theory is still relevant. Major news organizations, such as the New York Times, The Washington Post, CNN, and Fox News, ect., still choose what news to discuss and what news to omit. These organizations still have immense power and influence over what type of news and what content internet users see and read.

For example, in the case of health communication, Dr. Oz is probably the most well-known media figure who exclusively covers health issues. Dr. Oz is incredibly wealthy. I found different numbers, but according to a quick google search, his net worth is between $7 – $14 million. Other online influencers who share health information are often paid to promote specific products within their content. Individuals who produce health content on their own blog because they enjoy it do not earn a profit. Therefore, the content that is not profit driven may be the best source of information. However, when it comes to health information, credibility is a critical and very valid concern. Anybody can start posting content online that contains information that is factually incorrect, spreading misinformation.

One limitation of the current scholarly research on digital inequalities is they do not study the different type of content that users are posting. I also think it is important to study who is reading what online. Someone may not be highly educated and may not post online, but reads the New York Times every morning. For this person, this “digital inequality” may not affect them on an individual level. However, certain problems may affect people on an individual level. For example, health communication scholars know that lower income people do not read health information as frequently as higher income people. This affects people on an individual level and could cause negative health outcomes.

While it is important to study who is producing content and who is not, major companies still have much greater power because they have the attention of consumers, which means they make a lot more money than smaller companies or individuals in terms of ad revenue. Well-off people may post on the internet more, but that does not mean anyone is reading it. I think it is critical to study the receiving end of the content so that we can address further social issues, and further understand who is doing what online.

Week 3 – Emerging Media Relationships

Political Engagement

Personal Connections in the Digital Age explores how humans have reacted to the growth of digital media and how it has impacted relationships, communities, and societies. Baym argues that early predictions of an emotionless, socially isolating digital environment are false. Humans have found ways to make the internet a more personal and humanized environment by creating specific language cues and communities. Yet, she still questions “how can we be present, yet also absent” (3). One specific example I found particularly interesting was the relationship between social media and political engagement. Critics of social media believe that it will decrease civic engagement or at least cause slactivism.

However, real-life examples and academic work discredit that notion. The book provides two examples, the Arab spring and occupy wall street, as political periods that heavily relied on social media to spread their message and coordination of their activities.

Social media is certainly useful for uprisings, protests, and emergencies to quickly disseminate information. However, social media was also a critical factor in the 2016 U.S. election. Social media will only become more important to politics. It is a useful tool for any democracy, and I would argue that it enhances it.

Political Engagement at LSU

In my 7001 Research Methods course last semester, my research group conducted a survey looking at LSU students’ use of social media and political engagement. College students are an important population segment their voting behaviors significantly impact elections. The study aimed to address rising concerns about college student’s disengagement in politics due to increased time spent on social media. However, we took the opposite stance because we hypothesized that it could be a useful tool in connecting with college students. 91.4% (n = 158) of respondents reported using social media for news. A regression analysis suggested that increased social media use correlated with increased political engagement.

Baym also points out that we want to say social media is either good or bad, but she suggests that binary perspective limits our understanding of how people are actually using social media. Approaching social media studies without such an assumption could benefit scholarly research as a whole. She accurately names this “the myth of cyberspace.”

Further research on political engagement 

Further research is certainly necessary. Surveys are useful, but it is often difficult to measure actual political engagement. Reliance on reported political engagement may result in a higher level than in reality as respondents tend to overreport activities they perceive as virtuous.

The online atmosphere has changed dramatically, and humans have developed an online way of doing things that allow for greater interpersonal connections than ever imagined at its inception. Academia will have to adapt quickly to keep up with future changes in online media use.

Early critics also believed that the internet would be a disconnected space from the users’ reality. However, Baym argues the opposite, that one cannot understand the internet without assuming it is somehow related to their real lives. Of course, people do use false identities, or catfish others online, but that is an exception (176). For the majority of people, their online behavior and relationships often have real-life consequences. Therefore, future media scholarship must understand this relationship. Interviews and qualitative analyses are highly useful, such as the ethnographic research done by danah boyd. The internet may be a different space, but it is still part of reality, and therefore cannot be studied independently of it.