Twitter Removing the “@’s” in Tweets

caroline sinders
5 min readMay 24, 2016

Today twitter announced they would be removing @s from their Tweets, so mentioning a person will take up less space. If a user wants to share the tweet with their followers, they simply have to retweet their own tweet. From Twitter’s own blog post: “These changes will allow for richer public conversations that are easier to follow on Twitter, and ensure people can attach extra elements, media, and content to Tweets without sacrificing the characters they have to share their view.” Mainly @ people will autopopulate their name and up to 50 people can be added, from what I have inferred, from people a user is following AND media added to the tweet (URLs, images, gifs,etc) will not count as additionally characters. So tweeting is still just 140 characters, but you can now “do more” with it. Programmatically, things are slightly different: “A hidden prefix region that may contain one or more space-separated @mentions which shall not be rendered as part of the display text, but must instead be rendered as metadata,” (taken from twitter’s blog post).

But I have a bunch of questions, mainly was blocking QAed at all during this? And how do the API changes affect that? Has anyone tested small scale and large scale blocking?

But what was fixed??

Okay additionally, what does that above tweet mean? Anil Dash pointed out this on Twitter today:

The above is super true- what do our multithreaded conversations look like now with new Twitter? Can I turn off being included in threads? What does that do to journalists, activists, or just regular people when character limits kept them from being mentioned?


Blocking is incredibly broken in Twitter. Last year, I did an experiment with a boy I was dating. We had almost hundred mutual followers in common. He was more of a power user than I was, at the time I had 700 followers and he had over 3,000. I asked him and two mutual friends if I could conduct an experiment around blocking. I blocked him for three days and asked everyone to take notes. Friend 1 would tag us in conversations, as would Friend 2. He would be tagged at the beginning, and end, and same with myself over multiple individual tweets. Another mutual friend who did not know of the experiment tagged us in long DM thread right before the experiment started; the thread continued into the blocking. When I blocked the boy I was dating, he was removed from the thread but it said he “left the conversation.” After a few days, I unblocked him, and debriefed with everyone. He could see himself being tagged in conversations by our friends but not me, and could respond. He said he noticed no difference, and that I often appeared in his timeline regardless of where my handle sat in the tweet, and I noticed that as well. Our friends noticed no difference as well. The point of this anecdote is that blocking feels like muting but only amped up slightly. And the infrastructure of Twitter seems to be designed in that same way- meaning infrastructural blocking is not that all different than muting. Users are not removed from each other’s social graphs, or are hard coded/blocked from seeing each other, but rather, the algorithmic ties between users FEEL paused, or muted, but not actually blocked. I was not removed from that other person’s system, but rather there seemed to be a tiny road block between our communication, a very slight barrier to us communication. Myself, “@carolinesinders,” as an entity was not removed, or caught and placed in any sort of list that had actual block parameters. Twitter seems to build algorithms on top of algorithms to solve problems, but these are deep infrastructural problems that need to be addressed. Messy code on top of messy code won’t fix this but a better designed system will.

So how do these new changes to the API affect blocking? I actually have no fucking idea and no one else seems to either.

What would have been more ideal is if Twitter had illustrated this chart and was really straight forward with the changes in their blog post:

Here’s two things I made that could have been included in this post to better explain to users that this is what tweets will now look like:

And even, hey, you can remove the ability to be tagged and this is what our new Privacy page looks like;

Ps that above tweet tagging? Facebook offers that in their privacy settings, but it’s called “Timeline tagging” and you can turn it off or request approval.

But, Twitter could even add this: hey, we thought about blocking and it looks like this:

And then Twitter could say, guess what? We improved the blocking and they can’t tweet at you AT ALL. Currently, if a person you’ve blocked @’s you and another person who you haven’t blocked, they can tweet at you, but you’ll get a notification of the thread if the other person replies. Actually, they can tweet at you no matter what, they just can’t retweet you. If you’ve blocked someone, they can tweet at you all they want and the tweets will still go through, you just don’t get any notifications. So if @’s are improved, did this blocking feature get improved?

Mainly, what I want to know this:

Can I opt out of large scale tags? What if I want to take up space? Can I pause these new threads instead? Or can I demand, as a user, to okay being tagged in things by unfollowers and can you please actually block my handle from being used by people I’ve already blocked?



caroline sinders

Machine Learning Designer and Researcher | Artist | Instigator| online harassment researcher, fellow digital Harvard Kennedy School