# Matt's Reading Responses (Set 2) ## 5 Reading Responses ### March 16 - Ads Online ads are now a staple of commercial businesses because they can easily track user activity and target ads using a vast dataset of factors including demographics, time of day, locations, interests, online behavior, etc. According to Stokes (2014) in **"Online Advertising,"** social media is an increasingly popular platform for online advertising because an individual's activity on social platforms can provide ad servers lots of information about a user/customer. While some platforms lend themselves better to certain types of ads, social media is also a great tool for advertisers because it can host a range of **display ads** and sponsored content. Some of the main display ad types include: * **Banner adverts,** which are graphics on a webpage that can be interacted with. They vary in size and may expand with interaction. * **Interstitial banners,** which appear on the website between pages. * **Popups and pop-unders,** which open new windows either over or beneath the main web page. * **Floating adverts,** which create a layer over the web page content on the same window. * **Wallpaper adverts** that change the background of the web page. * **Map adverts** which show up as locations on an online map. Many of these ad types carry over to social platforms. Platforms like Facebook and Twitter also offer platform-specific sponsored content such as **promoted posts** and **promoted tweets**. Despite a decline in engagement with display ads, advertisers continute to rely on them as a staple of modern advertising as publishers and servers take advantage of extensive personalized user datasets to target ads. (As a side note, I am slightly confused on the difference between/interaction among ad **publishers** and ad **servers**.) While some users may welcome the concept of targeted ads, others may prefer to retain their privacy while browsing the web. The [Vox](https://www.youtube.com/watch?v=HFyaW50GFOs) video provides disturbing insight into the lengths some online platforms will go to to reach potential customers, even if it means going against their wishes. Ad blockers are meant to prevent third-party ad servers from tracking user's behavior across sites, but Facebook has implemented measures to circumvent the blockers against users' wills. Online advertising is not inherently bad and can be enjoyed or overlooked by many, but those who actively intend to block servers from tracking their online activity should be able to. ### March 26 - Breakup Breaking up can be further complicated by the continual change of social media and other digital communication platforms, but becoming familiar with the current rules of **second-order information** online makes it easier to adapt to these changes. Therefore, while Gershon's analysis of Facebook's involvement in relationships may appear outdated because of the platform's decline in popularity, current **Instagram and Snapchat users** can easily trace certain practices to earlier Facebook norms. Some aspects of breaking up remain unchanged in the age of Instagram: unfollowing each other, deleting old posts, blocking phone numbers. Unlike Facebook relationship statuses and away messages featuring breakup song lyrics, though, these remaining similarities are less directly about public display than about the people breaking up. However, they still hold important information for uninvolved interested members of the public. Rather than seeing a status update about a relationship, Instagram users who aren't in direct contact with the two breaking up can look to see if a former couple has **unfollowed each other** or **deleted old posts** with each other. This involves more clue-hunting and assumptions about second-order information than the more direct Facebook statuses discussed by Gershon. While years ago during the shift from Facebook to Instagram, users might add a **date, name, and/or some emojis to their bio** as a makeshift "relationship status," this practice has largely been left behind. Another difference is that Instagram users can ***retroactively edit* post captions**. Some may choose to do this when their ex is featured in a post that simultaneously marks a momentous event (prom, holiday, etc.). Revising the caption to be about the event rather than the ex can allow someone to keep up a nice prom picture without potential "stalkers" thinking they're still involved with their ex. In terms of signaling the end of a relationship or the start of a new one, Instagram and Snapchat 24-hour **"stories"** can be used similarly to status updates and away messages, though less permanently and more ambiguously. Reading Gershon's interviews with people who used song lyrics in statuses and away messages to convey sentiments about breakups in 2012, I was reminded of some instances in which friends who use Instagram and Snapchat either a) have posted song snippets to their stories for the same reason or b) have read into an ex's story post as being directed towards them. Stories' temporary nature make them useful for sending these types of indirect messages without cluttering one's grid. Gershon also discusses Facebook statuses' involvement in the definition of a relationship between two people as well as their public display of that relationship. Now, someone may start posting pictures of their dates with someone to Instagram or Snapchat stories even **before defining a relationship** with that person. In this way, stories are useful in ambiguously signaling to followers that you're seeing someone new, but are less helpful than Facebook statuses in outright defining the relationship. Because users might be reluctant to make the big move from *temporary* story posts to more *permanent* grid posts about someone they're dating, followers may not explicitly know of a relationship until a "Happy 6 Months!" or similar post down the line. While I've been discussing breakups in the digital age, I wanted to note that what's actually being discussed is the period **post-breakup** when exes are sending second-order messages to each other or their followers via social media. As [Lenhart et. al](https://www.pewresearch.org/internet/2015/10/01/after-the-relationship-technology-and-breakups/) note, teens are fairly consistent in their belief that **in-person or phone call breakups** are the most acceptable, and social media is not a platform many are using as an initial means of breaking up with each other. Rather, a breakup conversation likely happens between the soon-to-be-ex couple *privately*, and *public* Instagram or Snapchat activity is a means of cutting off that person and directly or indirectly conveying the breakup to followers. This is one more key difference between Facebook and Instagram regarding relationships. While Gershon tells stories of people signaled to a partner that they wanted to breakup by changing their status, Instagram's lack of this feature seems to make the actual **act of breaking up more private** even as the post-breakup continues to be somewhat public. ### March 30 - Shaped The Internet can affect **self-esteem** in various complex ways as demonstrated by the chapter ["Shaped"](https://readingthecomments.mitpress.mit.edu/pub/niu4m6vu/release/2) in *Reading the Comments*, which acknowledges many possible interactions between social media and self-esteem while also making it clear that there is **no one-size-fits-all solution** to Internet-aggravated self esteem issues. The chapter's discussion of **fixed and growth mindsets** as related to self esteem was thought provoking. At first, I too hastily assumed that moving feedback discourses from praising fixed traits to praising effort and improvement would solve issues of self esteem caused by people perceiving aspects of themselves as unchangeable. However, as I read about the increasingly popular practice of **cosmetic surgery**, I wondered how this instance of using one's resources to change something they didn't like about themselves (in this case, their physical appearance) instead of accepting it as fixed differed from other applications of a "growth mindset." While the benefits of encouraging growth-based feedback rather than fixed trait praise/derision seem evident when discussing learned skills, this may not be the best solution when it comes to physical appearance. Emphasizing a "growth mindset" in terms of physical self esteem issues rather than teaching self-love and body positivity may worsen issues of obsessive bodily surveillance. This therefore demonstrates that solutions to self esteem issues are not necessarily universally applicable, and it is important to differentiate when to encourage positive change and when to encourage self-acceptance. Also, despite the fact that I haven't heard of digital comparison services like Klout being widely used today, I know of many social media users with the tendency to mentally **compare themselves to their online friends/followers**. On the opposite end of the spectrum, I have friends who have entirely **deleted Instagram and Snapchat** accounts (sticking to mainly text messaging as a means of contacting a handful of closer contacts) in an effort to be more present in their everyday lives. I find myself trying to cultivate a relationship with social media somewhere in the middle, **turning off all social media notifications** and only engaging when I think to. However, I'm still left with some of the consequences that come from browsing social media at all. For instance, I've had enjoyable moments interrupted by seeing a post of someone doing something apparently more exciting. Although users today are aware and often reminded that social media profiles are inaccurate, curated, polished versions of people's total selves, this knowledge may not matter in practice when scrolling through one's feed. Rather than thinking, "This looks like more fun than what I did...But it's ok because this person is only showing me the best parts of their life and is not having this much fun all the time," my thoughts stop at the ellipis. What I'm considering in the moment isn't if I'm better off than this person overall, but if I've chosen the best use of my time, speaking to anxieties of **FOMO** mentioned in the chapter. This also demonstrates the **paradox of choice** in which too many choices can lead to greater stress, since each activity/event-related post on a feed represents a possible choice that the viewer didn't make for themselves. Overall, the chapter brought up many thought provoking questions and proposed many potential answers, but these questions remain complex and their potential solutions indefinite. ### April 6 - Algorithmic discrimination The readings on algorithm bias led me to reconsider my previous thoughts on algorithmic filter bubbles, particularly those we read about on **Facebook feeds**. Writing about the tendency for Facebook to show users content that aligns with their political beliefs **based on their previous interactions** with their feed, I previously stated that I saw no inherent wrong in an algorithm that in part amplifies a user's own habits and interests. However, this concept does not necessarily apply when discussing algorithms that provide individual users results based on **mass search habits** (such as search engines). As demonstrated by the [Buzzfeed article](https://www.buzzfeed.com/fionarutherford/heres-why-some-people-think-googles-results-are-racist) about Google providing search results reflecting racist biases, these types of collective algorithms often do not amplify the habits or reflect the intentions of the individual user. This may be problematic in the sense that it forces search results on individual users reflecting harmful mass biases that privilege white experiences and erase or denigrate those of nonwhite POC. The Google employee's response to these issues that Google cannot be blamed because "if organisations and media companies ensure what they are depicting is empowering and unbiased from the inside, in turn our algorithm will reflect this on the outside" is untrue. As Buzzfeed explains, searching for images of "unprofessional hairstyles for work" brought up images of natural Black hair that it pulled from posts critiquing racist attitudes towards hairstyles. This means that, unlike the Google employee claims, it is possible for the Google algorithm to take **search results out of their intended context and make errors that perpetuate bias**. Therefore, while Google's algorithm may be meant to reflect the intentions of users, this is not always the case in practice. If these biases were not intentional prior to them being called out by the public, their **maintenance will be intentional** if improvements are not made as soon as possible. Furthermore, algorithms tailored to mass search habits may not be benefitting the individual user, but I have no solution to propose since increasingly personalizing search results to the individual would have the downside of increasing the filter bubble effect of the Internet. Possibly a reevaluation of the way mass and individual habits are weighted by search algorithms is in order. In terms of commerce, [Hannak et. al](https://personalization.ccs.neu.edu/Projects/PriceDiscrimination/) demonstrated in their study that many sites have algorithms intentionally exhibiting **price discrimination**. Some like Cheapticketz and Orbitz clearly label the reason for price differences as a benefit of membership. This seems to be a reasonable tactic to encourage people to sign up for a particular service, but sites that do not make clear how to get the lower price being offered to others or that do not tell users they are being shown a higher price at all are engaging in deceitful price discrimination. As explained in the paper, some travel sites inform their discriminatory practices with users' **click behavior or previous bookings**, which is an unequal practice that unfairly charges certain users more or steers them to more expensive bookings. While some may say this is justified because it is based on the users' behavior, **pricing should not be covertly variable**. Also, just because a user previously booked something more expensive does not mean they currently are interested in or can afford more expensive options. I'm realizing this may contradict my above thoughts that personalized algorithms are fine because they amplify the behavior of the user. Through the same reasoning I used on price discrimination, it can also be said that someone who previously viewed left-leaning sources on Facebook may not currently be interesting in this kind of news. I need to think more about why, but it feels different when the user's money is involved. ### April 9 - Collapsed context **"i wouldn’t tweet anything i didn’t want my mother/ employer/professor to see,"** as stated by one of [Marwick and boyd's](http://www.tiara.org/wp-content/uploads/2018/05/Marwick_boyd_TweetHonestly.pdf) participants, is an effective summary of many concepts covered by their study. It illustrates that even as a Twitter user tries to tailor their tweets to an conceptual **"ideal" reader**, they must also consider the potential **"nightmare" audience** of the relatives/employers/professors. Though far from a perfect rule, I think generally the ideal imagined reader influences what a user *will* say while the nightmare reader limits them, determining what they *won't* say. A person may tweet a joke because they think their ideal intended audience would enjoy it, but censor or alter parts due to the potential of an unintended nightmare audience reading it. In the sense that a user must consider all potential readers aside from the intended target audience, networked online spaces like Twitter **collapse contexts** and force users to conform to a presented self that is palatable for all, including those not in the intended audience. However, users have many means of re-expanding these contexts. While the above holds true for someone's broadest-reaching public personal profile, many social media apps now allow users to log into and easily switch between **multiple accounts**, encouraging the potential presentation of multiple "selves." For instance, one individual may have a public personal account, a private personal account, a fan account, a meme account, an art/photography account, a business account, etc. Marwick and boyd seemed to focus primarily on the first category (public personal), which definitely collapses context most, but does not limit one from presenting other personas in other contexts through accounts that are private to non-followers or anonymous to a degree. Many nichely targetted fan/"stan" accounts and nameless meme pages are run by users who also maintain a broad-reaching personal public account. Therefore, just because contexts are collapsed on one account does not mean they must be on another. In terms of authenticity, Marwick and boyd's respondents had varying perceptions of **authentic** that often included being **"personal"** and **not being "influenced"** in what to tweet. However, as Marwick and boyd point out, authenticity as a concept depends on what viewers will read as "authentic" and therefore a presentation of oneself as authentic can't be separated from social influence. I'd also note that being "authentic" to the Twitter platform or a specific realm of Twitter does not always involve being "authentic" to oneself. For instance, people are viewed as fitting in to parts of the Twitter community when they adopt popular circulating **Twitter-language and tweet formats** in that community. Rather than being how someone would speak free of influence, these presentations of being an "authentic" Twitter user are inherently tied to the influence of the platform. Furthermore, I found it interesting that microcelebs who profit from their social media presence would be judged as inauthentic if they appear to care about the profit/attention they receive. In order to maintain the profit and attention, they must essentially pretend not to notice it. Therefore, these microcelebs are driven to presenting themselves a certain way in order to conceal and maintain their monetary stake in their presentation. They are inauthentically creating falsely "authentic" selves for their audience.
Last changed by  
 
matt-rose77
110

Read more

Home Page

Matt's Home Page Required assignments Wiki tutorial Web search and evaluation Filter and label your email Adblocking Privacy footprint Reading responses

4/19/2021
Matt's Privacy Footprint Practical

While I don't usually pay much mind to my Internet footprint and am not typically too concerned about my privacy, I found segments of the readings eerie, particularly those describing intentionally covert tracking methods that everyday platforms use. In Chapter 11 of Kernigham's D Is For Digital, Kernigham describes methods such as invisible 1-pixel embedded images that create cookies and track users across site and hidden JavaScript in URLs/HTML that disguise third party cookies as first party cookies to circumvent blockers. While I don't really care when I can tell I'm being tracked, I still find these practices disturbing because they are intentionally misleading and even directly go against permissions explicitly set by the user in the case of the disguised cookies. As I've said in previous responses, I don't see an inherent wrong in user tracking myself, but I still think users should be able to disable tracking. Rather than seeing these settings as a challenge to be overcome, ad services should acknowledge the boundary set by users who disable third party cookies. In the Haridy's article about the myth that Facebook listens to our conversations through our devices' microphones, I found it interesting that he explained how Facebook wasn't listening because the cell phones tested didn't use as much data as a virtual assistant like OK Google and then did not elaborate on what these assistants are doing. I've heard the same rumor about virtual assistants that I have about Facebook listening in to our conversations. While I understand that listening for keywords is part of their service, I'd be interested to see what these same researchers have to say about the rumor that these assistants track us in the way that people imagine Facebook has been. My Identity As I'm writing this, I'm realizing the instructions say to conduct a Spokeo search but I earlier clicked a link to Intelius that I can no longer find. After putting in my name and hometown into Intelius, the service was able to link me to my parents and my brother (whose name appeared twice in my relatives list for some reason). However, it displayed by middle initial as "B" when it's actually "R" and I'm curious to know where online my middle initial it listed as "B." I also tried entering my name and town on Spokeo and was not able to find myself in their list of Matthew Roses from the area. Browser

4/2/2021
Matt's Ad Blocking Practical

Before UBlock Origin: After UBlock Origin: I definitely noticed that sites loaded faster with UBlock on, and I found I was able to more efficiently read web pages without the insistent distraction of attention-hungry banner ads, pop ups, and floating ads. While this no-ad system makes it easier to interact with the content I intend to, if every individual blocked all ads it would create a problem in which ad-supported free content would not be sustainable. Free sites may have to implement pay walls or, if no users are willing to pay for previously free content, shut down entirely. As I argued in a prior reading response, ads are not inherently bad and, while they may be annoying, they allow us to have easy, unpaid access to lots of online content. However, I still believe that users should be able to deny permission to be tracked by ad servers. Rather than moving towards ad blockers, users who are bothered by tracking should be encouraged to activate tracking protection as discussed by Marti in "Targeted Advertising Considered Harmful." In this way, users would be able to browse the Web without being tracked across sites by third-party ad servers. Users would still be shown ads, and therefore tracking protection would not be a detriment to advertisers as ad blockers would. However, these ads would be relevant to the content on the page rather retargetted based on a user's previous online behavior. Marti explains that this makes ads more meaningful to viewers and increases their signaling power as long as users know they are not targeted, which the user of tracking protection would know. Ad "blockers" like AdBlock Plus that allow some ads through their filter, as long as they are deemed acceptable and as long as advertisers pay the blocker, are after their own profit and do not offer the best solution to users. While ads may be more in line with the page's content, this is the same effect that tracking protection would have. Furthermore, the model is susceptible to corruption in which unrelated ads eventually get through the filter for high prices, creating a similar environment to the pre-blocker Web except with another middle-man seeking to profit. Therefore, the best solution for those who are bothered by ad tracking but enjoy free ad-supported content is to implement tracking protection. Unfortunately this does not provide a solution for users like myself who are not really bothered by tracking but would simply prefer to view web pages for free and without ads, as UBlock allows. I suppose we are instead forced to accept that we have to allow ads on our pages if we want to continue to enjoy the benefits of ad-supported content. Although as ads lose their impact and users develop "banner blindness," a term coined by Betsy Haibel and quoted by Marti to describe users gradually tuning out constant online ads, even this system people like myself now take for granted may need alterations.

3/17/2021
Matt's Reading Responses (Set 1)

5 Reading Responses January 29 How the Web Works offered a simplistic metaphorical explanation of the relationship between clients and servers that was difficult to grasp and felt like it was hiding something when I would have preferred a more in depth explanation. While I was able to mostly follow their metaphor that compared the parts involved in accessing a webpage to parts involved in a person getting an item at a store, I have trouble truly understanding the process even at a surface level without a literal depiction of what is happening. I had similar issues understanding Hartley Brody's article explaining HTTPS. This second article did offer better explanations of the parts in the system it described, and the color-mixing graphic was useful in illustrating the mathematical concepts. However, I still think I need to see it functionally explained with pictures or examples of the true process in action in addition to merely reading about it. In both articles, there are many unfamiliar terms talked about in relation to each other and, even with their description and metaphorical comparison, I have trouble understanding how the system operates without understanding what each part literally or physically is. For instance, I do not know what is meant by describing Transmission Control Protocols as "communication protocols," and I do not know what is meant by describing HTTP as "an application protocol that defines a language." I would like to better understand what each part literally is so that I can understand how they interact as a system. Similarly, I have trouble understanding how the parts physically interact. How does a one wireless device communicate with a server and download information from it on a physical level without being connected to it? February 8 Make It Stick offers an idealistic method of learning that, while mostly convincing of its superiority, is an unrealistic expectation for the majority of learning. The text sets up "rereading" as the most common method of studying and argues that empirical evidence has shown no signficant learning benefit to repetetive rereading. It proposes instead that learners adopt more effortful, challenging, engaging methods of study such as retrieval practice, in which learners are required to recall and/or apply concepts to answer quiz questions, and interleaving, in which learners switch among studying various topics to provide space between practicing each subject. While these suggestions are possible, they make the assumption that learners always have an internal motivation to learn the material they are required to learn when in my experience this is often not the case. I found myself frustrated reading the book's suggestions as it condemned the practice of cramming for a test as only bringing short term benefits rather than longterm mastery; most people don't care about mastering every subject, but we are still expected to get good grades on every test. The text includes multiple praises of flashcards over rereading notes, arguing that rereading is more time consuming than it is worth (10). However, notes are often taken during classtime and can easily be reread, while the flashcards method proposed as an alternative requires the additional effort of the learner to create flashcards on their own time and plan to use them often and spaced out over time. This is often unlikely for students who juggle about 8 subjects in high school and maybe 16-19 credits in college with tests and assignments occurring in too rapid a succession to be able to study far ahead of time for all of them, especially on top of extracurriculars, jobs, volunteer work, etc. that are also expected of students. For those who know they can grasp the material well enough in the short term for tomorrow's test by rereading their notes the night before, mastery may not be a concern. Similarly, the concept of reflection is emphasized as a form of retrieval important to learning (66), when common education formats often provide no motivation or opportunity for such reflection after the test has occurred since its grade is already final.

3/9/2021
Read more from matt-rose77
Published on HackMD