Point Nine Land
Published in

Point Nine Land

Twitter is not working on an edit button

A few weeks ago, I put out a few tweets about our investment in Heygo. When I noticed a misleading typo in one of the tweets a few hours later, I was reminded that Twitter doesn’t allow you to edit a tweet once it’s published. It wasn’t the first time that I wished I could edit a tweet, but this time the lack of an “edit tweet” button was particularly annoying.

The world has many, many much bigger problems than tweet editing, and you might be wondering why I’m writing a blog post about the topic. The reason is that it leads to some interesting product design and UX questions and I love geeking out on this kind of stuff. Still, I wasn’t sure if I should publish this post, which I had drafted a couple of weeks ago, but when I saw Twitter’s April 1 tweet getting hundreds of thousands of comments, likes, and retweets, I figured there are enough people who are interested in it. If you’ve come to this place for practical tips on how to build a SaaS or B2B marketplace, read at your own risk. :-)

In this post, I’ll take a look at the following questions:

Why aren’t tweets editable?

One possibility is that Twitter just doesn’t care. There might be hundreds of other features and improvements on Twitter’s backlog that have higher priority. However, given how many times the feature has been requested by users, I can’t imagine that Twitter hasn’t considered building it. Someone at Twitter must have seen the jokes à-la “we can fly to the moon, why can’t we have editable tweets?”.

My guess is that from Twitter’s perspective, the costs of making tweets editable (which are significant, more on that in a second) outweigh the advantages (which benefit only a vocal minority). I haven’t talked to anyone at Twitter about it, though, so it’s all speculation. If someone from Twitter reads this, please feel free to chime in.

So what are the costs of making tweets editable? It can’t be the engineering effort that’s preventing the feature from being built. LinkedIn posts are editable, so are posts on Instagram or Facebook. I know it’s hard to imagine for startup folks like me just how expensive it is for a large organization, which runs a product that is used by tens or hundreds of millions of users, to make seemingly minor product changes. But in this case, I don’t think that’s the reason.

The real problem is that if you allow users to edit a tweet after it’s been published, you compromise the integrity of the conversation that it may be a part of. If a user can send out a tweet, get thousands of likes, comments, and retweets, and then change the content of the original tweet, you’re opening a pandora’s box full of problems. In the best case, it can lead to mild confusion. In the worst cases, it will be misused by people who want to spread misinformation or push their agendas. Imagine you retweet or like a tweet, and later that tweet gets changed to a statement promoting racism or violence. Horrible.

What could a solution look like?

If Twitter wants to make tweets editable, it will have to address two issues specifically:

The first issue, I think, is easily solvable in a fairly unobtrusive way by adding a “warning” note or icon that tells users that a tweet has been edited. This is what LinkedIn does, for example. If you wanted to take it a step further you could let users switch back to the original tweet and highlight the changes, along the lines of a very simple version history. This seems like an easy to solve UI challenge.

How edited tweets might be indicated

The second issue is much more complicated. Imagine a tweet, which got hundreds of thousands of likes and retweets, gets edited. If that triggers a notification to everyone who interacted with the tweet, you’d bother hundreds of thousands of people. Even if most of them don’t feel bothered — user attention and screen real estate are limited and valuable, so the price that you’d pay for improving the experience of one user, the user who published the tweet, would be way too high.

Now, in the majority of cases, it may not even be necessary to notify those people because the edit isn’t material but only a correction of a typo. But in some cases (e.g. the misuse scenarios mentioned in the beginning) it absolutely is.

An obvious idea would be to use AI to determine if a tweet has been edited in a material way. If the AI thinks that the edit is material, people are notified. If not, then not. This could be part of the solution, but given the very low tolerance for false negatives (i.e. a tweet is edited materially and the AI didn’t detect it) I think it’s not sufficient.

One possible solution might be to notify a random, statistically significant subset of the relevant users. If almost none of them react to the notification by un-retweeting, un-liking, or deleting comments, you can conclude that the tweet edit wasn’t critical. If a significant number of them does take action, everybody else is notified as well.

Using a combination of AI, notifications to samples of users, and maybe some editors at Twitter, one could minimize the number of notifications but you will still have a trade-off between optimizing the experience of the tweet-editing user and other users. Another challenge comes from the fact that countless third-party applications use Twitter’s APIs. So for any of the changes that I’ve described above, you’ll have to think through the implications on the various API use cases.

Is tweet editing a problem worth solving, or are the costs higher than the upside? I don’t know. In any case, next time I’m annoyed about the lack of this seemingly simple feature, I’ll keep in mind that it’s a lot more complicated than adding an “edit tweet” button. :-)



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Christoph Janz

Internet entrepreneur turned angel investor turned micro VC. Managing Partner at http://t.co/5WJ3Pepbcv.