Twitter chief executive Jack Dorsey is on a media tour to fight for his company’s reputation. On the left, activists are angry that the platform hasn’t kicked off Infowars conspiracy theorist Alex Jones; on the right, the president of the United States accuses Twitter of “shadow banning” conservatives. As Dorsey promises to make the company more transparent and works to combat misinformation, it’s worth debunking some of the more common myths about Twitter.
Once, “shadow banning” was a relatively obscure term, referring to one of the oldest moderation techniques on the Internet — hiding a user’s posts from others on a platform without booting them outright. It has since become a politically loaded catchall for censorship, as a conspiracy theory from the fringes of the Internet has crept into the mainstream, claiming that Twitter is sytemically banning conservatives from the platform because of their ideology. This is how President Trump, in a recent tweet, came to claim that Twitter was “‘SHADOW BANNING’ prominent Republicans.” His tweet followed a Vice News article discussing the fact that Republican National Committee Chair Ronna McDaniel and “several conservative Republican congressmen” were not auto-filling in a drop-down menu in Twitter’s search function.
Although Twitter does limit the reach of some accounts, based on a variety of factors as part of its anti-abuse and anti-harassment strategies, Vice’s report didn’t prove that the platform was targeting conservatives for their views. Democratic lawmakers were not similarly affected, but the issue appeared for other accounts on the left, according to at least one journalist who tested the search autofill at the time.
In a statement, Twitter said the accounts were not limited in the search bar because of their beliefs, but instead based on the accounts’ behavior and the behavior of other accounts interacting with them: “Our behavioral ranking doesn’t make judgements based on political views or the substance of tweets.”
The company corrected the search results shortly after they went viral on conservative Twitter. The clearest proof that this theory is spurious may be that many of the voices amplifying it, such as Trump and Rep. Matt Gaetz (R-Fla.), have a powerful platform from which to do so: Twitter.
that brand account.
Whenever a social media account for a major celebrity, brand or (ahem) newspaper tweets something particularly clever — or particularly offensive — you’ll inevitably hear that it was the work of a “social media intern.” In 2017, for instance, the A.V. Club credited an “anonymous social media intern” with having the KFC brand account follow only “11 Herbs and Spice Girls.” (Get it?) When IHOP said it was changing its name to IHOB to promote its burgers, one typical reply to its tweets about the campaign read, “I feel so sorry for you, poor social media intern, but you can’t make this work.” And, of course, then-candidate Trump blamed an intern for a manual retweet in 2016 mocking Iowans.
But in 2018, interns are not generally put in charge of major accounts for influential organizations or people. In 2008, sure — back when social media was a new idea. Today, running a major branded account is more complicated.
“The days of getting interns to run your social media accounts are long gone,” said Kristin Johnson, director of content and communications at Sprout Social, a social media management, advocacy and analytics software company. In 2018, running a social media account means taking on the responsibility of a major, public-facing communications channel for that brand, a job that requires access to “significant and sometimes sensitive information.”
It’s been that way for a while. Between 2010 and 2013, one study using data from LinkedIn found, postings for social media management positions increased on the job-finding platform by more than 1,000 percent.
Twitter’s rules bar hateful conduct and violent threats. So whenever Trump tweets something that seems to be a threat — as when he lashed out at Iran’s president in July — there’s a viral call to ban him from the platform for breaking the rules. GQ has argued that Trump’s account is “very much in violation.” A Change.org petition lobbying for the president’s account to be banned has more than 15,000 signatures.
There’s just one problem: A handful of loopholes protect Trump’s account from running afoul of Twitter’s policies. The most important is an exception for “military and government entities.” As the company explained in a January blog post, “Blocking a world leader from Twitter or removing their controversial Tweets would hide important information people should be able to see and debate.” (Still, Dorsey told BuzzFeed in a recent interview, the loopholes might not apply should Trump use his account to engage in rule-breaking behavior toward private citizens.)
“Account has gained over 5 million followers in less than 3 days. Mostly bots. He’s getting ready for something,” tweeted @th3j35t3r last year. That message accumulated more than 10,000 retweets before it was deleted, spawning an unsubstantiated rumor that spread through Twitter and beyond: Hillary Clinton appeared to cite it at the time to raise questions about what was going on with the president’s account. In an interview with Recode , she referred to “new information about Trump’s Twitter account being populated by millions of bots.”
The Washington Post and others debunked the rumor : Trump did not gain more than 5 million followers in less than three days, as the tweet claimed, and it wasn’t even close. The followers he did gain were not mostly bots. Nevertheless, the myth won’t die. As my colleague Philip Bump noted recently when it resurfaced in another viral tweet, there are many explanations for the faceless new accounts that appear to follow Trump en masse, including the fact that Twitter recommends popular accounts to new users — Trump’s among them. And while Trump may have a substantial number of bot or suspected bot followers, data suggests that it isn’t any worse than for other prominent politicians. When Twitter recently purged suspicious fake or automated accounts from the site, Trump lost 300,000 followers, or 0.58 percent of his total . Former president Barack Obama lost about 2 percent of his followers in the purge, which amounts to more than 2 million accounts.
of the Internet.
Twitter feeds can sometimes feel like a free-for-all, where everyone’s trying to talk at once. Because of that, there’s a tendency to mistake the informational avalanche as representative, which is how you get all those headlines about what “the Internet” is thinking or feeling. “WhatsApp was down for two hours and the internet wasn’t happy ,” reads one Verge headline. “The Internet hates McDonald’s new uniforms ,” declares Food & Wine. Then there’s this one from my own archives: “The Internet won’t let Harambe rest in peace .” In many of these articles, a handful of Twitter users are standing in for “the Internet” overall. And then there are Twitter polls — those unscientific multiple-choice posts that are mostly for fun but sometimes get cited as a reflection of public opinion, sometimes even by figures like Trump .
But Twitter is not representative of public opinion. According to a 2018 Pew Research Center poll, about 24 percent of the American population uses Twitter. Forty-six percent of those users visit the platform daily. And, as is true for many social networks, young people are much more likely to use Twitter than are other age groups surveyed, which is to say its most active members aren’t a representative sample of the Internet’s diverse denizens, let alone humanity as a whole. Spending a lot of time on Twitter may make it feel like the whole world is talking about one topic or another. But step outside the timeline, and you might be surprised.