Make Twitter’s Edit Button Obvious, Ugly and Fleeting

Elon Musk wants to reinvent the social media site in his own, disruptive image. Don’t let him

Elon Musk’s recent acquisition of 9.2% of Twitter, and his elevation to the company’s board, was accompanied by a typically impish Twitter poll from him asking: “Do you want an edit button?”
By Ben Schott
April 08, 2022 | 11:13 AM

Bloomberg Opinion — Ah, the Twitter edit button.

For nigh on two decades, hot-take miners at the comment coalface have debated the merits of this mythical feature. Opponents doubt it will functionalize the site and fear it will muddy already brackish waters; advocates hope it will elevate the discourse while simultaneously eliminating typos.

(The surest way to proofread prose is to publish, at which point every infelicity becomes glaringly obvious while a dozen more elegant phrasings of your bien pensée appear like so many esprits de l’escalier.)

But old order changeth, yielding place to new, and Iron Man fulfils himself in many ways.


Elon Musk’s recent acquisition of 9.2% of Twitter (TWRT), and his elevation to the company’s board, was accompanied by a typically impish Twitter poll from him asking: “Do you want an edit button?” At the time of writing 4,406,764 votes had been cast: 73.6% in favor, 26.4% against.

Twitter reacted defensively to the ensuing commotion, protesting that it had been working on an edit button for months before Musk dropped $2.9 billion to become a “passive investor.”

Musk’s Twitter Fans Find Value Analysts Don’t See

Never mind that such protestations sit awkwardly with a retweet of Musk’s tantalization by Twitter’s CEO, Parag Agrawal, who said: “The consequences of this poll will be important. Please vote carefully.”


Some sites have editing built into their DNA. Wikipedia is premised not just on the continual modernization and moderation of its pages, but on the public availability of their (sometimes tortured) evolution. Donald Trump’s 16,806-word entry, for example, averages 7.3 edits a day — each of which is visible and comparable, should you have the humanity to click though the links. Indeed in a triumph of meta that’s worthy of Facebook, you can even explore the page history of the Wikipedia page dedicated to explaining Wikipedia’s page history:

Page History.dfd

LinkedIn takes a less accountable approach where both posts and articles are editable — though neither tracks changes, and while the former displays an “edited” warning, the latter does not:

“Your connections won’t be notified when you update your article, and there won’t be an indication that the article has been edited. Once your edits have been saved, the original version of the article will no longer be available.”

When it comes to Twitter, however, any such silent emendations to published tweets would be hazardous to the public sphere and calamitous for private reputations. Imagine liking, retweeting or commenting favorably on a video of adorable otters holding paws, only for it to be stealthily updated to a tirade of racist abuse. Cancel culture may be a buzzword, but it’s also a buzz saw — assassinating characters at a keystroke, and rarely with due process.

If Twitter is to embrace editability, a number of tweaks might ameliorate the change.

Stylistically, edits should be immediately conspicuous (to render changes obvious and incontrovertible) and also somewhat inelegant (to deter all but the most pressing). One approach might be that used by the Twitter bot Editing The Gray Lady which deploys clear but clunky color-coded tracking to highlight changes in the main page of the New York Times.

Bored Ape Yacht Club Warns Users Not to Mint NFTs After Discord Hack

Such legalistic highlight-and-strikeout formatting would inevitably have an impact on the character and character-count of edited tweets. Yet in the volatile vortex of social media, unambiguous tracking must surely be safer than click-through links to edit logs or hover-over pop-ups to version histories.

In terms of technology, it makes sense to restrict the number of edits (say, to one) to prevent endless revisional ping-pong, and expedient to curtail the editorial window (say, to a minute) to enable typo correction while preventing long-tail manipulation.


(Twitter Blue, the company’s premium subscription service, already approximates this functionality with “Undo Tweet” which offers a pre-publication preview period of up to 60 seconds before a post goes live.)

To the extent that editing tweets will empower the author, it should also protect the audience. And so, even with such limitations in place, it would be desirable for Twitter to notify any user who has interacted with an edited post that a change has been made (providing an opportunity to revisit the interaction) and to offer a setting that would auto-delete interactions from any edited post. It should additionally be possible to indicate pre- and post-edit interactions (with an icon or text color) to protect users from accidental misunderstanding or cynical misattribution.

Is the U.S. Losing the War for Global Talent?

The risk of Twitter introducing a freewheeling edit function absent such stylistic and technical guardrails would be to destabilize further an already precarious discourse. Just as lies can girdle the world before truth has been shod, so a correction tweet rarely “does the numbers” of its fallacious forebear. How much more perilous (and poisonous) will Twitter become if every post can be weaponized retroactively in an endless fandango of bait and switch?

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Ben Schott is Bloomberg Opinion’s advertising and brands columnist. He created the Schott’s Original Miscellany and Schott’s Almanac series, and writes for newspapers and magazines around the world.