Why I no longer read “news”

I realize that there is a rather tired trope of “fake news”, and maybe if you’ve read other of my posts here, you’ll see a trend where I’ve tried to detox from the monolithized “mainstream” media and use tools like RSS to refine what I see. Today one of them slipped through, so I can show you precisely what my grievance is with a solid example.

One of the paramount reasons we call a story “news” is because we trust that there is a reasonable relation between the article we read and the sources it cites. A good journalist will follow through in ways such as reaching out to the people involved in the article, or ask clarifying questions for the reader to consider. Newspapers also have a section where a writer can indulge in their own opinions on a topic, hence called the “Editorial” or “Opinion” section.

So take a look at this “article”, from The Guardian, an outlet typically held in high esteem:


The author opens with a single source cited (we’ll get to that in a moment), proceeds to draw a conclusion based on the source that car insurers are charging higher premiums for people with Hotmail email addresses than other address types, and proceeds to indulge us with a personal recollection of how things “used to be.” She has a couple paragraphs of nostalgia, followed by a concluding remark that if Hotmail is the “worst” email to have, a personal vanity domain is the best, backed up by a single anecdotal experience she had. In my opinion (ironically), any editor worth their salt would have tossed this in the bin, or at least routed it to the Opinion section, where it belongs, rather than being paraded under “Technology>Email>Shortcuts.”

The verdict gets more damning when you bother to read the source. First, one look at the source website screams tabloid to me. Second, the source article is based on an “experiment” run by the author, hardly scientific. Third, the correlation between the points in the source and the points in the parent article is nonexistent. Let’s look at some examples.

The Guardian article headlines with a quote, “A sign that you’re not keeping up.” Clearly the implication is that the conclusions for this article were justified by something car insurance companies have said. Except that the quote is nowhere to be found in the source. Second, the article headlines with the phrase “car insurers.” Plural. But the source shows a simple experiment with a single insurance company, in the UK. Third, the source contains a brief quote from the insurance company in question, which never made it into the parent article. Therefore, the entirety of The Guardian article is baseless lard, opinions of the author being passed along as fact.

The reason I’m analysing this so deeply is the reason I discovered this excuse for news. There is currently a thread on Hacker News, bearing a headline that actually grabs the first sentence of the article in question, and there is a 200+ comment discussion assuming that the article is fact, and people coming up with theories as to why such widespread discrimination against Hotmail users is happening. For example, the very top comment declares it must be some kind of AI modeling.

So what we have here is the following chain: a barely functional tabloid article about a nebulous-at-best non-scientific experiment on a single car insurance company in the UK, which then turned into an opinion-piece-masqueraded-as-fact declaring an epidemic happening with multiple insurance companies not limited to any location, which then turned into a discussion about why such systemic discrimination is happening, and tapping into clear fears people have about how these insurance companies are clearly doing lots of data mining.

I wish this could be a post about why The Guardian has fallen, but I’ve seen similar pieces from the New York Times and other places. This kind of garbage is precisely why I’ve had to start honing my RSS feed to pay attention to particular journalists who have earned my trust, and disavow the fourth estate as whole.

Social Autopsy

Since there has been a lot of drama and misinformation both about this project and my stance on it, I’ll take some time here to expand in more detail.

Yesterday I became aware of a kickstarter called “Social Autopsy.” As far as I can tell (from its video), the premise is to create a database of information about people who are being annoying harassers on the internet. While this created all kinds of controversy, I think it raises some interesting and important questions.

First, it’s important to note that the creator has specifically said that the database will only be comprised of public data. This is a very, very important point. A lot of people have criticized the project for doing dox dropping, which it does not. I’ll explain why in a moment. Second, I have come out, not in support of the project’s aim (which I disagree with), but in support of the project’s right to exist. Unsurprisingly but amusingly, I have taken a lot of heat for this stance. So I’ll elaborate on that in a moment too.

Regarding the project, as far as I can tell, it is a central gathering place of public information into a format that is more readable and accessible. At first glance, it seems people can contribute to it, although it’s not entirely clear how this process will work. There are a few comments about accepting screenshots and so on, which raises eyebrows. so here is my analysis.

First, if what they are making is basically a┬ásort of filtered and easier to read Google, with information that could be found already online, I have absolutely no problem with it. One could argue that there is public and then there is “public”, which I think is a fantastic discussion to have, and is really what is on trial here. A lot of people post things in a forum they don’t consider public, such as posting something on a twitter account with 20 followers, and it then hits the front page of Gawker. This is actually a problem with unreasonable expectations, and not a privacy violation per say. If, on the other hand, someone tweets on a private twitter account, that is then screenshotted and posted onto Gawker, I would say Gawker is completely in the wrong. When the person twittered, they did have a reasonable expectation of privacy, and Gawker, not them, made the tweet public by publishing the screenshot.

There are some interesting scenarios that will come up. What happens if someone tweets, then deletes the tweet, but it is captured and posted on this database anyways? We’ve already seen this question come up with Politiwhoops, but one could argue that it is different for a private citizen. Also, what if someone tweets from a fake account that looks like you, and it gets added to your profile? And finally, what happens if a tweet is taken out of context and put in the database in a way that unfairly reflects you in a negative light? All these questions need to be considered.

In terms of why I support the projects existence: first, I fundamentally disagree with the premise of it. I’ve spent years doing research into the nature of names, privacy, and identity, and have found absolutely no conclusive evidence that forcing people to say things under specific names has any net benefit without causing collateral damage. Further, I think that putting unrelated items together in a list and suggesting that they are related without any proof is also problematic– this is why slander and libel laws exist. However, I think that if you have two items related to an individual that are both public, putting them next to each other on another public page is completely reasonable.

I did state that people should be more careful about what they post on public places, and I meant it. If you have a twitter account with your full legal name, and a linkedin account with the same legal name, and both are public, it seems unreasonable to assume that nobody would be able to put the two together. In terms of what happens when someone has unpopular opinions that bring consequences in another forum, there are two points to consider: context and proportionality. With the recent LambdaConf drama, for example, there was a speaker who had made some racist comments in a context completely unrelated to his LambdaConf presentation, and yet people merged these contexts together and demanded unproportional retribution. A couple thoughts here: first, was he entitled to curtilage, and second, what good would have come from banning him from the talk? And finally, if both of these personas he had were considered in the public sphere, is it reasonable to suggest there might be more local context-based versions of public?

Since this Social Autopsy project doesn’t even exist yet, I think it’s a great place to begin a serious discussion on these issues, and perhaps get some feedback on how to solve the underlying problems (harassment) without compromising personal privacy. We shall see.