Social Media Information Manipulation

I probably spend about 15 minutes each morning on Facebook before I even get out of bed. I scroll through my newsfeed, check up on what my friends are up to, follow a few interesting links, etc. I might check it once more before I head out the door. Once I am at work, I usually try to stay away until lunchtime (after all, who hasn’t logged on to Facebook for a quick check-in and then found themselves still scrolling 30 minutes later?) and then might check once more before I leave work around 5:00. If I am staying in that evening, I will probably have Facebook open on my computer until I go to bed. Even if I am watching a movie or reading, I check what’s going on in social media land every 30-45 minutes. Just a quick scroll, unless I find an interesting link to follow.

All in all, I probably spend 2 hours each day actively scrolling on Facebook per day. It is important to me. Since I recently moved across the country, it is how I find out how my friends and family are doing. I also use it find out what events are going on near me—music shows, films at my local theater, parties, campus events at my institution. Honestly, I am more likely to find out what is going on at my institution through Facebook than the newsletters I get emailed multiple times a week.

More importantly, Facebook is where I get most of my news. I follow The Washington Post, The Guardian, Ars Technica, Jezebel, Library Journal, and The Comic Book Legal Defense Fund, among others. Through the links posted by others, I may also get stories from Gawker, The New York Times, Politico and any number of other sites. And I’m not alone in this. A 2015 study found that 63% of Facebook and Twitter users get the majority of their news from social media.

Which makes it troubling that one of the questions submitted to Mark Zuckerberg at a recent employee Q&A was “What responsibility does Facebook have to help prevent President Trump in 2017?”.

nbc-fires-donald-trump-after-he-calls-mexicans-rapists-and-drug-runners

Don’t get me wrong: I don’t like Donald Trump. He’s an easy man to hate (and such an attractive target for every major news corporation!) and would be a terrible president. I don’t think it is far-fetched to say that if Trump is elected president and allowed to take office, America will enter into a severe decline—economically, politically, ethically, take your pick. But it makes me uneasy that Facebook could influence the outcome of any election.

The company has already dabbled in politics. In the 2012 election, it actively encouraged voting by showing users pictures of their friends with the “I Voted!” logo. Furthermore, according to a Mother Jones report, “in the three months prior to Election Day in 2012, Facebook increased the amount of hard news stories at the top of the feeds of 1.9 million users. According to one Facebook data scientist, that change—which users were not alerted to—measurably increased civic engagement and voter turnout.”

On one hand, hurrah for increased civic engagement. On the other, Facebook is manipulating the information received by its users—without telling them.

And this is not the first, or even the most infamous, example of Facebook manipulating its newsfeeds. In early 2012, computer engineers at a Facebook joined forces with a group of Cornell scientists to conduct an experiment with Facebook newsfeeds. Using a randomly selected group of users (who were not informed that they would be part of an experiment), they manipulated newsfeeds so that some of the group saw a feed with fewer positive stories and others saw a feed with fewer negative stories. The experiment went on for a week and sought to test whether social media users can experience “emotional contagion” (the act of transferring emotional states between individuals) online. Their results, published in Psychological and Cognitive Sciences, indicated a parallel relationship between the number of positive or negative posts the users generated and the emotional content of the stories on their newsfeed—thus indicating that their emotional state was affected by the posts they saw in their newsfeed.

An “Editorial Expression of Concern and Correction” added to the results post-publication claimed that since Facebook is a private company the experiment was not held to the standards of Cornell’s Institutional Review Board (IRB). Rather, the researcher’s actions were allowable under Facebook’s Data Use Policy, i.e. the policy most users don’t read before hitting “I Agree”.

Facebook is a private company. As such, it is legally allowed to do what it wants with its product. And it is certainly not the only information source to demonstrate a bias regarding certain issues. Bulwarks of the newspaper industry like the New York Times have a tradition of endorsing certain presidential candidates.* However the NY Times still carries stories about the opposition, and its endorsement is clear and out in the open. If Facebook follows the precedent it set with the 2012 election and the emotional contagion experiment, the company will not let users know that it has chosen to combat Trump’s candidacy.

*In the upcoming election, the NY Times has chosen to endorse Hillary Clinton and John Kasich.

When a platform has this much sway over the news stories its users receive (not to mention their emotional state), it needs to be regulated. Given Facebook’s user base, I think its closest comparison shouldn’t be newspapers but rather television. The FCC already regulates political broadcasting on television; why not on social media?

rtr4c0uz

This is something I want to talk more about this future (particularly regarding what will happen when candidates start trying to use money to influence how Facebook presents their news stories—because, of course, that will happen) but for now, just a note about current rules regarding broadcasting and politics. According to “Political Broadcasting: Questions and Answers on the FCC Rules and Policies for Candidate and Issue Advertising” “[television] stations must provide equal amounts of time for candidates for the same office, and otherwise treat candidates for the same office in the same way” (11) and “if you sell time to one candidate for an office, you must be willing to sell an equal amount of time to the opponent” (14).

To put it simply, Donald Trump is the worst, but we shouldn’t allow our information sources to be polluted. We have the right to make up our own minds by accessing all available information.

 

Agree? Disagree? Think my interpretation of broadcast policies is amateurish? Let me know in the comments!

The Dangers of (Warrantless) Cell Phone Tracking

A federal judge ruled warrantless cell phone location tracking constitutional on Wednesday, April 13th. This ruling makes it more likely that the issue will eventually go to the Supreme Court.

Police and other organizations track cell phones by tracking where they connect to different cell tower locations. Since most of us have our cell phones with us at all times, this can give a pretty good indication of where a bad guy goes and who he/she sees. So tracking a known drug dealer’s cell phone can show not only where different deals are occurring but also other drug dealers.

That sounds good, right? And it would be awesome– if it were done with a warrant. The ACLU argues in their amicus brief that cell phone tracking shows more than what business the target may be conducting; it also shows  deeply personal information like where they pray and who they are romantically involved with. This is information irrelevant to the police that could nevertheless hurt the cell phone owner and innocent bystanders.

Cell phone tracking without a warrant is a violation of privacy. Police need to find a way to conduct their inquiries while also respecting the privacy of their communities.

Private Speech: Why WhatsApp’s Encryption Matters

To an outsider it would seem like I don’t have a lot to hide. I read a lot of books. Sometimes I go out with my friends and lose a game of pool over a few beers. My one brush with the law came when as a teenager I got a parking ticket for forgetting to move my car for a parade that I was actually marching in. That’s about it.

I still don’t want anyone reading my private conversations. I believe that a privacy violation still causes harm even when it pertains to the most boring personal information available.

Case in point: If I am talking to my friend Sarah over text about what shoes I should wear to a party we are attending that weekend I might share the following information:

  1. I recently went shoe shopping at a local shop where I bought a cute pair of patent leather black flats
  2. I had to wait to go shoe shopping until I got my paycheck
  3. While shopping at DSW, I saw an adorable pair of yellow peep toe heels that I didn’t buy because of my weird toenail fungus (which, for the record, does NOT exist)

For the sake of simplicity, we’ll skip over what Sarah shared with me in return– for the moment.

Now let’s pretend that my shoe conversation is caught up in a sweep of all cell phone conversations and locations on my network. Of course, my particular cell phone conversation is insignificant. It’s just about shoes, for heaven’s sake.

This still bothers me.

First of all, I take issue with the idea that because the information given in my cell phone conversation is not “important”, it is not worth keeping private. According to this view, the value of my privacy is dependent on the value of the information in question. On the face of it, this might seem fine. If my debit card information is stolen, I could lose all my money; whereas if someone listened in on my last phone conversation with my mom, they would just learn more than they ever wanted to know about my sister’s upcoming wedding.

But what measure are you using to determine my information’s value, and why? My debit card information is more valuable on a financial scale, but if your measure is how much any piece of information would affect my family, then a recording of a phone conversation with my mom is much more valuable. What if one of us said something that we want hidden from my sister? The potential for hurt feelings is huge if some third party breeches our privacy in this case. And consider my shoe conversation from earlier. Financially valuable information? Not really. Potential for hurting my standing at my job? Non-existent. But this information does indicate how I relate fashion to my paycheck, which depending on what kind of circles I run in could harm me socially.

Which leads me to my second argument: seemingly innocuous information can contain more important and/or harmful personal information than it initially seems. Here is another interpretation of my shoe conversation:

  1. I was at this given location (local shoe shop) at a particular time and day
  2. My financial situation is such that I cannot buy things at will; rather, I have to time purchases with my paycheck
  3. I have a gross toe-related medical problem

This is all information that could be used against me in the following ways:

  1. If I went to the shoe shop during the workday, my employer could terminate me for engaging in non-work related activities. If I have a stalker, he/she knows that I have visited a certain place. A corporation could make note of my shoe-buying tendencies and start sending me annoying junk mail about their products (which could also hurt the local business I patronized if that junk mail persuaded me to abandon it for a chain).
  2. I could start receiving emails about payday loan services; if I were a vulnerable person, I might take out a payday loan as a result and get trapped in a high interest loan. If a prospective boyfriend found out about my limited financial resources, he may decide that I am not a good option for him.
  3. Everyone at work/school could find out about my weird toe fungus and make fun of me. That prospective boyfriend (who is clearly a weenie) from #2 could lose interest.

I am not exaggerating the possible consequences. A woman in California was fired after deleting a required app on her work phone that tracked where she was at all times. Likewise, by tracking purchases typically made by customers the company knew to be pregnant (unscented lotion, cotton balls, etc.), Target was able to start sending targeted advertisements about baby products to female customers who were buying similar products—something which backfired when they sent the baby product advertisements to a teenage girl whose father was very upset about the implications (the girl was later revealed to be pregnant). And stalkers have hacked into cell phones enough in the past that at least one women’s shelter requires visitors to dismantle their phones when they arrive.

WhatsApp

The point is that information is valuable, no matter what form it takes, and we should have the right to protect our personal information, even if it is just about shoes. That is why WhatsApp encryption is so important. I don’t have to be a freedom fighter to want my information to be private.

Since WhatsApp has embraced end-to-end encryption, a billion people (most of whom are boring, unique individuals like me) can now have truly private conversations. They are even safe from WhatsApp itself—the app administrators cannot see your conversations either. Here is the white paper from WhatsApp explaining the technical side of their encryption. It will be very interesting to see whether this move towards encryption spreads to other widely used companies.

Do you have thoughts of your own that you want to share? Comment away!

Welcome!

This is a blog about all matters related to intellectual freedom: privacy, censorship, freedom of speech, etc. I want to especially focus on how these ideas relate to what is currently going on in the world today. So most of my posts will focus on a recent news story.

About me: I am an academic librarian currently based in Idaho. I read. I write. I bake. I enjoy a good bourbon or craft beer, and I ride my bike. That’s about all you need to know about me. Seriously, those are the important things in my life. Especially the baking (and the eating of the baking).

I welcome any and all comments that follow a few simple guidelines (see the About page), so always feel free to comment on my posts! We learn by talking to one another.