Twitter™ shadow bans

Social networks were supposed to be great. They provided an easy platform to surf the Web, chat with friends and family, join groups, keep us informed. Somehow, they became less fun. In this post, I will share my experience with a Twitter™ shadowban, how I got it, and how I tried to understand it.

For a couple of years, I have been weary of Facebook™, LinkedIn™, and other social networks. In a way, they felt like an unnecessary layer between the open possibilities of the World Wide Web and the user. Still, in the 2000s and 2010s, unless you want to be perceived as an outcast, you must maintain some type of online presence. Twitter™ had always been my network of choice. It was the less intrusive one in terms of privacy. Also, being talkative, it allowed me to post a lot of tweets without flooding my subscribers. It was the nature of the medium after all.

In the fall of 2017, the Liberal government in Quebec cut funds to the national library and archives (BAnQ). I was especially outraged by the cuts to the digital team. As an art historian, their digital archive is a major source for my research. As an art historian and a coder, I simply could not understand how this was a good policy in this day and age. So, I made my voice heard.

Up until that point, my account @MrcGthr had been enjoying an organic growth. It had gained a few hundred followers, mostly from the #DigitalHumanities, #cultureQc and #mnbaq hashtags. I had posted tweets on computer safety, Canadian and Québec paintings, digital humanities, and things in that arena. Nothing political, just cultural and coding stuff. I had one unusal habit, though: I deleted all my tweets after one week. It's not that I had anything to hide, it's simply that I wanted my Twitter™ feed to reflect my current state of mind. I also had the habit of opening and closing accounts after one year, to keep things fresh.

The @MrcGthr account was opened for about six months when the budget cuts to BAnQ happened. So, I decided to show every single way the national archives were useful to me. I began with #MerciBAnQ where I linked to all the resources available online. Then, I saw another user who felt outraged. He was using #JeSoutiensBAnQ which loosely translates to "I support BAnQ". I let go of my hashtag and used that one instead to make our voices stronger. Then, shit happened.

I began tagging politicians in my tweets. It worked. The #JeSoutiensBAnQ movement gained steam. Members of the opposition began using the hashtag to show their support for the institution. Newspaper articles began to be written about the noise on social media. A protest was held, and television stations dispatched crews to report on it. In the middle of all this, Twitter™ shadowbanned me.

That was in 2017. Twitter™ still denies it shadowbans users. However, at one point, I noticed that my visibility stats were down. At the same time, my notifications fell. I was at a loss. Were my tweets less engaging? Had the campaign moved to another hashtag? It took me a while before I understood. What was happening was this: Twitter™ was letting me see my tweets, but it wasn't showing them to anyone else. In effect, I had the feeling of still being in the conversation, yet they had muted me. I was a shadow, I was banned.


What I was seeing


What others were seing

My first reaction was one of shock. Wait, what? Twitter™ does this, without warning? They didn't tell me that I was doing too much, that I was too active. They didn't tell me that I needed to cool down. They didn't tell me that, maybe, this could be a nuisance to other people's experience. No, they just shut me out of a political campaign right at the moment when it was taking of.

My second reaction was instrospective. Had I been wrong? Was I too talkative? Was I pushy? So, I asked. I tweeted in private to people using the hashtag. I asked them what they thought of my tweets. I did not get one negative return. All of them said that I contributed to the campaign. Actually, since they no longer were seing me, they thought that I had dropped out. That I was a quitter.

My third reaction was anger. Wait again, what? Some algorithm, somewhere, had judged my content, had pulled me out of a conversation, and I looked bad because of this? That's not fair. I was a bit relieved when I noticed that Dilbert's creator Scott Adams had also been shadowbanned. I guess I wasn't outlandish.

Finally, I was curious. So, I began experimenting with the network. I opened and closed my account. I opened another one. I created apps on the developer site. I deleted my tweets. In the end, I discovered this about my shadow bans. First, they last about 24 hours. After that, my account becomes visible again to the Twitter™ community. Second, the behavior that seems to trigger the most rapid shadowban is the use of the same hashtag in a short time span. Tagging the same user in many ways (location, handle, hashtag) also seems to do the trick. Finally, opening a new account doesn't really help because the "organic growth" algorithm seems very restrictive at the beginning.


shadowban.eu provides feedback on your account

This is my experience with Twitter™. To show this probelm, I'm using a red cross next to my name and the hashtag #qfd (Quality Filter Discrimination). It's not much, but if it helps anybody, then this has not been written in vain.

last update: November 2018
français
home