String Theories

Twitter's UX and 'bullying'

February 17, 2015 | 6 Minute Read

The longer you've been on Twitter, the more likely it is that you've seen, been part of, or been on the receiving end of what I would politely term a Twitter Clusterfuck.

Someone, somewhere has said something controversial. It might be something mean. It might be something offensive. It might be something stupid or funny or smart. Whatever it is, it draws the attention of a larger proportion of the Twitter network than would normally interact with that single person’s tweet on a daily basis.

The best (or perhaps worst) examples of this are when someone says something blatantly offensive.

And this is where our problem begins.

A tweet that elicits a strong reaction from one person is likely to elicit a similarly strong reaction from many people. As Twitter users, we’re conditioned to see a tweet and quickly hit reply and share our thoughts.

The result of these two factors colliding is a large number of people simultaneously or in quick succession replying back to the original tweet. Often these replies are passionate, especially when the original tweet was particularly offensive.

To the individual senders of the replies, they’re merely highlighting an issue in the original statement. The receiver, however, faces a deluge of tweets, flooding in one after another. Replying back to each individually becomes overwhelming. It feels like an attack.

Lately, the Twitter community has taken to calling these “pile-ons” or even “bullying*”.

That’s a bad-faith characterisation of a problem caused entirely by Twitter’s bad design.

##So, what’s really going on here?

Twitter, the platform, is designed as an intensely personal experience. It’s optimised for encouraging a user to interact with other people. It’s also terribly designed for protecting its users from the extremes of those interactions.

People who have used Twitter for a while have built complex social layers on top of Twitter to make up for the design deficiencies. Like many organically developed social practices, these are not universal in understanding. They rely on learned experience in the community and sufficient daily cognitive capacity.

###How does this relate to “pile-ons”?

When you see a tweet, the easiest actions you can take are to favourite, retweet, or reply. This is what Twitter wants you to do. Twitter adores engagement.

The next easiest thing to do is to open that particular tweet. This will show you some of the replies, but not all of them. It might give you an indication of whether someone’s already raised your concerns. In some Twitter apps/clients, you might be able to click somewhere to see more, but in most you can’t.

The next thing you could do is to open the Twitter feed of the original Tweeter. This will show you if they’ve subsequently corrected themselves, or at least, if they’re replying back to a few people on the topic of the original tweet already.

If you’re lucky, your Twitter client might allow you to view all the original Tweeter’s @mentions. This will give you the best indication of exactly how many replies they’ve received and whether it’s appropriate for you to add your voice to the conversation. However, most Twitter clients don’t show you this, including the official Twitter clients.

The complex social layers people have built involve checking on one or all of these in some order before replying back to the original tweet. However, everyone’s understanding of these practices is different, and even ruthlessly adhering to this doesn’t always work, as multiple other people may have replied in the time taken for you to compose a tweet after checking.

These social layers are also absolutely undiscoverable to new users. Like many social protocols surrounding Twitter, they are learned through observation of the community in action, and often learned the hard way when something goes badly wrong.

In addition to this, human brains are simply not wired to sustain this model of practice on a prolonged basis. It requires continued, concerted effort and vigilance in a setting that we do not typically employ such things.

Social media is a small part of our daily lives. It is something we do in-between other things - while we’re bussing to work, or on a conference call, or eating lunch, or watching our kids, or a million and one other things. We typically dedicate a fraction of our brain’s resources to processing and replying to it. Even someone with the best intentions in the world will slip up, and slip up often.

We’re placing a huge expectation on individuals to strictly adhere to behaviour that is in direct contrast to the behaviour Twitter’s design encourages them to do. I don’t think it’s reasonable to characterise the inevitable mistakes as bullying; it’s Twitter’s incompetence, rather than individual users’ that’s the issue here.

##What should happen?

This is a product design problem and an engineering problem that’s being made into a human problem. Here’s some things Twitter could do.

  1. You type a reply. You hit the ‘send’ button’. You receive an alert: “23 similar replies have already been sent. Are you sure you want to reply?”
  2. An option for ‘Don’t show me replies to this tweet or tweet thread anymore, and don’t allow retweets of this tweet anymore’.

Neither of these are particularly complicated from a design or an engineering perspective to implement. However, both of them contrast starkly with Twitter’s desire to have as much engagement as possible.

Ask yourself ‘Why are we angry at the people who make mistakes using a broken system, rather than the broken system?’

##What’s the hidden problem with this?

The obvious problem with this is the one everyone is talking about: it’s horrible to be on the receiving end of the deluge, and it’s also not all that nice to be accused of bullying when you were actually let down by bad design.

The hidden problem is more subtle: Repeatedly, the groups being accused of ‘bullying’ are marginalised communities. Women. Trans people. The Queer community. Black Twitter. People who have traditionally had no space where their voices have been heard.

Twitter has, for many of these communities, provided a platform where there is space for their voices and where they can discuss issues their communities face. It’s also meant, for the first time in many places, that companies and individuals have been forced to hear what the impact of their words are directly from the people they hurt. And, when those words hurt a lot of people, a lot of people try to tell them that.

Twitter’s bad design is directly leading to attempts to silence the voices of marginalised communities by accusing them of bullying. It’s derailing the conversations from the very important points the communities are raising. That’s not OK.

It’s crap that Twitter’s bad design is letting down everyone this way, but we need to act in good faith a bit more. I’d like to see less accusations of ‘bullying’ being flung about, and more calls for Twitter to help solve a problem that is, once again, disproportionately impacting our already marginalised communities.

*For the purposes of this conversation, the specific misuse of the term bullying refers to multiple members of a community (often inadvertently) flooding a particular tweeter with a response to an offensive tweet. It does not refer to the Wadhwa-definition, where a marginalised person disagreeing constitutes bullying, although this is also utterly ridiculous.