I’ve been against edited tweets for a long time, mainly for technical reasons. The main problem is that Twitter is not like other social networking sites; unlike, say, Instagram and Facebook posts, embedded tweets spread the news in many different ways. Reposting and retweeting tweets is the main risk associated with allowing users to edit them. Let me break down a few reasons why deploying Twitter can be tricky but still work.
Since its launch, but especially in an era where a Trump presidency essentially wouldn’t exist if it wasn’t for Twitter, the platform has become the fastest way to get the word out to an audience. This has its benefits, but it also means that the information that is being shared is going everywhere. In a world where fake news is rampant, this is… a challenge.
The ability to edit tweets has always been technically trivial – you update the database, the job is done. Databases are great for updating information, especially if you don’t care about change history or (god forbid) immutability.
Twitter is a platform of instantaneousness: as soon as you hit “Tweet”, all your followers will see what you just wrote within seconds. This makes Twitter different from all other social networks. It also means that, from a policy standpoint, editing is a complex task.
The concept of instant repost content is a powerful feature of Twitter, and while opinions differ on whether a retweet is an affirmation of sentiment or just an amplification, an idea can be retweeted within seconds of being shared. This complicates the concept of editing a tweet because it is very difficult to understand what is actually being edited. means.
If you think a little about what the implications of edited tweets are, you’ll open up a whole series of questions, each with two or more perfectly reasonable solutions, but each solution comes with its own set of problems. If someone tweets something and you like that tweet, what should happen to your “like” after the tweet is edited? What if editing means you no longer agree? Or what if your “like” actually triggers a Zapier script that does something to the tweet—what should happen to the tweet?
Imagine I write something about how incredible my local coffee shop is, they retweet it, and I edit the tweet to read “I should have gone to Starbucks, not LocalCoffeeShop”?
It looks like Twitter is still trying to limit access to the ability, Twitter Blue subscribers only.
The platform was also launched editing tweets for API usersso they can access past tweets and use the tweet editing API for accounts that have this feature enabled.
It will be very interesting to see how these features are used by users – and attackers – on the Twitter platform. There are so many fancy edge cases in this one; Here are a few that make me break out in a cold sweat as a product specialist:
- What if you report an offensive tweet that has since been edited and is no longer visible to regular users?
- What if you’re displaying tweets on a website but caching the content? How to make sure you’re showing the most recent version of a tweet?
- If you embed a Tweet, you, as the publisher, assume some editorial responsibility for it. How to trigger re-moderation of potentially very old tweets?
- What if you have an open tweet and it is being edited between the time you read it and the time you click retweet?
- What if you used a scheduling tool (like Buffer) to later retweet something with a comment, but the original tweet was edited? If your comment no longer makes sense, how do you deal with it?
- Should an edited tweet be re-inserted into the timeline for users to see it on a “newest first” basis?
- What if I could go back in time and, in January 2015, Donald Trump is believed to be running for president as a joke? So, I actually did it, but now that the edit exists, it would be much easier to fake it. It still wouldn’t be funny though.
- Should there be a difference between major and minor editing? For example, a typo or a complete change in the content of a tweet? And in a world where a comma can change the worldhow does the platform know what constitutes a major or minor edit?
I’m still curious if enabling editing will cause more trouble than it’s worth, but the limited rollout in New Zealand, Canada, and Australia is about to get really interesting. The potential for abuse is great, but maybe we can finally get rid of those stupid little typos in tweets that suddenly explode.