Why a Twitter ‘edit button’ is an absolutely terrible idea
One of the most frequently asked-for features by Twitter users is the ability to edit tweets, i.e. an edit button.
It’s an understandable frustration: it’s annoying to write a long tweet and then find a typo in it, forcing you to re-upload the entire tweet. When you want to update a tweet with new information but realise you can only reply to your tweet, which many people won’t see.
However: for all possible reasons, an edit button is a terrible, terrible idea. To understand why, let us take some hypothetical examples.
Here is a tweet. Let us imagine I edit it for a moment.
You might think, “well, we can just click to see the edit history, so what’s the problem?” There are several problems:
Most people won’t actually check the edit history
First of all, unlike other websites where edit buttons are enabled, Twitter moves very fast. People spend far less time looking at a tweet than they do, say, on a DeviantArt comment or even a Facebook post (as the latter are often longer). Facebook lets you edit text in posts, though not images or videos — and it shouldn’t let you edit the text at all. It is one of the reasons that Facebook is so infamous for the spreading of disinformation.
Because of how little time a user spends looking at a post on average, the vast majority of the time users will not check if a post’s edit history is malicious or not. Eg: it could just be a typo fix, it could be a totally different post — but many users, especially those who trust the sender, will not check. You can easily imagine how this could be used to malicious effect in politics, marketing and other areas (send a tweet everybody likes, then change it to a political message that those who retweeted would never endorse once it hits a high enough engagement count). Attempting to enforce a rule of not doing this would be impossible; minor changes in a tweet text that result in major context changes would be easy to do, for example, and it would become a huge political row for Twitter to try and determine what counts as a ‘big enough edit to be misleading’.
Engagement on any social media post is usually a tiny fraction of the total number of people who see that post (since people are scrolling on phones, may not suspect any foul play, etc etc). Clicking on anything in a post is even smaller — as such, we shouldn’t be expecting users to check edit histories, especially as many tweets would end up having them.
It’s also quite easy to deceive people into thinking the edit is minor — for example, change a tweet entirely, then write “fixed, damn phone autocorrected me lol” in a reply below it. A whole bunch of your users will assume you made a typo and never check the tweet edit history.
An edited tweet retains its engagement statistics
The whole point of editing tweets is to not have to re-upload them, which means it won’t reset the retweet and like counters. As such, you can easily exploit this: post a tweet, let people share it all over the place, then change the message to something completely different.
Simply by using this tactic, you’ve spread a message to potentially tens of millions of people that would never have gotten that far otherwise. Eg:
Tweet: “I love garlic! RT if you agree!”
*two days later*
Tweet edited: “<fake news article saying that drinking milk summons Satan from his eternal slumber>”
An edit time limit doesn’t fix anything
An often-suggested fix to the above issues is to only let tweets be edited in a certain timeframe after posting. This doesn’t solve anything for two reasons — let’s assume the limit to edit a tweet is 5 minutes after posting.
Firstly, by the time you edit the tweet at e.g 4:55 after posting, your tweet can have easily reached millions if not tens of millions of people on a large account. The damage is largely going to be done, especially if you wait for a little while and then delete the tweet — ensuring nobody can now view the edit history. While of course some people will see through this, and others will publicly debunk it, many users will remain fooled by it.
The second reason requires a bit of tech explanation. Particularly for mobile users, tweets are not always updated or posted immediately (e.g. you might be in an area with no internet connection, have a very slow connection, be in an underground tunnel…), and a lot of users also have background network activities disabled for apps to prevent using tons of bandwidth which costs money on many mobile networks.
As a result of this, whether because of a lost connection or because of mobile data requirements, many Twitter users on mobile are not seeing the exact current snapshot of Twitter, but a cached version on their phone. Thus, there is the problem that the edit will not be visible to the user for a long time. For example: let’s imagine I am going on an underground train at 5pm. At 4:57pm, you post a tweet, and my phone downloads it. I get on the train.
At 5:01pm, you edit the tweet — but I no longer have an internet connection and am browsing offline. I will see the unedited version of the tweet until the connection comes back, which could be a while away. Additionally, if edited tweets are not going to be pushed to the top of the timeline — by the time I can see it, even if I’m *not* disconnected, it will not be at the top of the timeline anymore.