This month, Twitter announced the ground-shaking news that it was actually going to offer users a way to edit tweets -- a longtime user request. And it would have been the biggest Twitter story in years, if it weren't for the Elon Musk takeover attempt. Now, it appears work on Twitter's Edit button has already begun, according to a handful of tweets from reverse engineers who have documented the feature's early beginnings.
Twitter had said that it was taking care to ensure the Edit button would not be able to be misused by bad actors looking to "alter the record of the public conversation" after the fact. But how exactly Twitter plans to ensure a tweet's edit history is highlighted has been less clear. Twitter itself may not even have a solid idea as to how this will look in the app's user interface. But the draft version of the Edit button gives us some idea as to the company's current line of thinking.
As noted by reverse engineer Nima Owji -- who helpfully tweeted a GIF of the button in action based on his findings -- access to the "Edit Tweet" feature would be found in the three-dot menu at the top-right of your tweet. From this menu today, you can do things like delete a tweet or pin it to your profile, among other things, so this would be a natural place to correct a tweet, as well.
— Nima Owji (@nima_owji) April 16, 2022
Owji shows that after you correct the tweet, you'll then click an "Update" button to post it to the Twitter timeline. But when asked about how the edit history would appear to other users, Owji didn't yet know. He told TechCrunch that from what he observed in code from the Twitter website, the Edit button didn't actually edit the tweet on the back end -- it creates another, new tweet. He suggested that it would be possible to make a list that includes the old version or versions of the prior tweet and append them to the newly revised version.
Another reverse engineer, Alessandro Paluzzi, also discovered the Edit button in development. Similarly, he noted that Twitter had not yet created the user interface where the original tweet or logs related to the edits would be shown to other users.
However, details provided by reverse engineer Jane Manchun Wong offer a bit more insight. She found code references related to the work-in-progress Edit Tweet feature in the recent build of the Twitter web app. She also noted that the Edit button wasn't actually correcting or changing the text in the original tweet -- it was creating a new tweet with the updated content.
She says that this new, edited tweet would include the list of the old tweets prior to the edit.
In other words, the new tweet would point to the original tweet and the edit history.
Looks like Twitter’s approach to Edit Tweet is immutable, as in, instead of mutating the Tweet text within the same Tweet (same ID), it re-creates a new Tweet with the amended content, along with the list of the old Tweets prior of that edit
— Jane Manchun Wong (@wongmjane) April 16, 2022
This is probably among the better ideas from a technological standpoint as to how to implement such a feature, which could otherwise be heavily abused to mislead the public as to what was originally said.
If Twitter had allowed users to edit the text in the original tweet, then both the old and new versions would point to the same tweet ID, which could complicate things from an engineering standpoint. As some commenters noted -- including a former Twitter employee, Ben Sangster, who helped to investigate the idea of an Edit feature back in 2015 -- that could be a nightmare for caching systems based on the tweet ID.
He theorized that Twitter could be working on a user interface that would provide a summary of engagement for all versions of the tweet, but would offer a user interface where people could view the pre-version engagement directly. This could help to address accountability concerns, but worryingly, it could also lead users to retweeting Twitter screenshots.
Then there's the troubling fact that Twitter's earlier investigations into the Edit tweet feature had found that, while such an option was technically feasible, it concluded the potential for abuse was too high to move forward.
Honestly, as complex as the implementation might end up being, the abuse vectors are going to be far harder to manage / mitigate. I was part of an effort to investigate this back in 2015-ish (?) and we concluded that the potential for abuse was too high to move forward.
— Ben (@sangster) April 16, 2022
Twitter has a significant challenge ahead of it, then, to come up with a user interface that makes it obvious to users that a tweet has an edit history, while also making the edits easily accessible instead of being buried under a number of clicks. It also needs to balance the fact that some tweets that undergo edits may have gone viral, and it needs to be clear to those who have clicked through to engage with the original that an edit has since been made.
In all likelihood, we could see Twitter test various versions of the Edit feature's user interface among Twitter Blue subscribers before any public version becomes broadly available. In fact, Twitter Blue has a testing platform built-in, Labs, which could be a place for such experimentation.
Reached for comment on the reverse engineers' findings, a Twitter spokesperson said the company is still experimenting and figuring out the path forward -- language that could mean the company hasn't solidified its decision for the user interface aspect to the system, either. For that reason, Twitter said it couldn't share much about the status of the Edit feature at this time, beyond what Twitter's VP of Consumer Product, Jay Sullivan, had previously said.
1/ We’ve been exploring how to build an Edit feature in a safe manner since last year and plan to begin testing it within @TwitterBlue Labs in the coming months. Sharing a few more insights on how we’re thinking about Edit 🧵 https://t.co/WbcfkUue8e
— Jay Sullivan (@jaysullivan) April 5, 2022