As long as we can see the original tweet, I don't see how it's a bad idea.
Of course, this will only work if old tweets can't be edited. Can't imagine anyone being happy to have their Twitter embeds suddenly say something that they didn't before.
Alternatively, an embed function could embed a specific version.
If Twitter doesn't implement this, it sounds like a great product to built a small SAAS around: blogs, newspapers, even governments often embed tweets so being able to embed without change, could be a sought after product.
I'm incredibly excited to see what changes Elon brings to twitter, if only for the entertainment value. His response to the CTO of facebook, "facebook gives me the willies", was hilarious
Twitter's announcement says they have been working on the Edit Button since last year. It has nothing to do with Elon's poll. Let's give credit where it's due.
Nothing, it's just funny they decide to announce it directly after his poll. Maybe they rushed to make it clear he didn't have anything to do with it? Maybe it's just a coincidence. Neither way matters.
Twitter? Adding an edit button is not exactly rocket science. Meanwhile at SpaceX...
edit: I liked your subtle edit below from "paid board seat" to "(ridiculously expensive) board seat". Strange to make it a negative to invest in a company you want to change, like other board members are allowed to invest for their board seats, but Elon isn't. If Twitter does good it will be lucrative, not expensive!
My point is to poke fun at the implication that any improvements made are solely the result of one investor who now has their (ridiculously expensive) board seat.
Is the 'edit' button feature actually a hard feature to implement in twitter (due to their scaling architecture)? I remember reading a passage in that "Designing Data Intensive Applications" that explained how their architecture was different if the account was say, a celebrity account which needed to push out to a lot of followers, or an account that had few followers and tweeted very little etc..
Difficulty was not the reason why it took so long. The excuse has been that they want tweets to be immutable, so you don’t share a tweet for it to be changed later.
* Do you reset the likes / retweets after an edit? If not, what happens if folks start selling the ‘edit’ on viral tweets?
* What happens if the original tweet violates the TOS but then the edited version does not? Does the original tweet in violation still get persisted? Does the edit get persisted?
You’re onto something. I’m pretty sure that’s going to happen if edits are not limited. We already see “wow this went viral, here’s a link to my album”
There's all the political decisions to make as to what do you do with existing Likes, Retweets, etc, what oversight will there be, if any, on monitoring if an edit was just say for a typo or it's a completely different text after the edit (and do they implement a system to auto-detect the % change and then auto-flag for manual review), do you notify users who previously viewed it or perhaps only interacted with it in some way that it was edited "x%", etc.
I assume they'll have to have things like version control, so all prior versions are readable for crowdsourced moderation, if not then that's a bad decision IMHO; some past decisions of theirs haven't been rational, and seem more fear based - so maybe they won't even have version control in the name of "privacy" - pretending that the person didn't previously said what they said, and that somehow it's deleted off the internet and everyone's memory if they delete the tweet.
Technically, it should be no harder than the "delete" button, which has been the de facto "edit" button up until now. The article itself talks about how defining the feature from a product perspective was really the hard part, and I'd say it makes sense to take that at face value. You really do want to be able to see when a text has been edited, there do need to be some limits, etc. and defining those things is not trivial by any means.
They probably made assumptions in their architecture that tweets can never be updated, and that they can only be created or deleted. Depending on how ingrained this assumption is, it could be quite difficult to change. Caching, replication, eventual consistency - lots of hard stuff. Until relatively recently, I believe AWS S3 didn't have read-after-write consistency for updates to existing objects - I doubt that was easy for them to do.
I've seen lots of bad hot takes from Twitter users who have no idea how large systems like this are designed. Dumb stuff like "just do a SQL update" or "use React".
I hope it is going to be similar to editable comments on Github, with a dropdown to see older versions.
Or perhaps like Gmail: stores the message but doesn't send until a few minutes later, at which point it's no longer editable. This would be easier to implement.
Unlike GitHub, tweets should not be editable forever. I think 5 minutes is reasonable. Even HN has an edit time limit. I hope they also limit how much you can tweets (even though, in reality, adding 4 characters (“not ”) is destructive enough.
Do you think HTML pages should be uneditable after a period of time? Regardless of your opinion there (since I can see the merits), the rest of the web isn't exactly founded on immutability.
HN has edits, in conversation and flow, and it does break sometimes, but in general it's a good.
Discussion forums have had edits forever; reddit famously lets you go and delete all your existing posts at anytime by replacing them with "edited by autodeleter9000".
But Twitter doesn’t want to be just another webpage. Tweets and its dates are used in court and by historians. That’s basically why Twitter has been hesitant to allow edits.
I kinda wish they let people add some text instead of allowing editing, like:
Wouldnt it be nice?
Added: * wouldn’t
I know deletions are still allowed, but they don’t lead to different content.
Maybe if you're a famous celebrity etc. but it's scraped mine 5 times since 2007 (and the "expand conversation" links don't work) and I promise you that it has been fairly active ever since I signed up.
Note that this isn't really an "undo", just a 5-60 second time delay before posting during which the user can change their mind before it's actually published.
Since this is a verge article, can I also say I like their comment system? You get a minute or so to fix any typos after submitting after which it is locked in.
> I hope it is going to be similar to editable comments on Github, with a dropdown to see older versions.
It took a while for them to add this. And amusingly, comments (including those opening issues and PRs) are a bit more like a wiki than personal comments—if you have the appropriate permissions, you can edit comments by others. For a long time you could do so with no visible record that it was done. Which is objectively terrible but was hilarious if you worked on a team with a good sense of humor and reasonable restraint.
I get "If I had asked people what they wanted, they would have said faster horses" vibes from this.
What people really want is not send tweets with typos and grammatical errors, so a good spell checker is what twitter should have built.
At the very least they could have started with that, because that's something that is easy to launch and un-launch, unlike editable tweets (which has database-level implications because they will store the history of edits).
I don't think it's grammar errors that's at the root of this. I think it's more subtly irritation over a valid criticism of a tweet that's got exposure, and wanting to change wording to alter the tweet's connotation to address that criticism while still making the same point by and large.
Judging by the number of comments suggesting to make comments "editable for only a few minutes after they’re first posted", I don't think that's right.
Well those people recognize the problem with letting people change their words indefinitely, but do see the value an edit button brings in fixing grammatical errors. A pretty reasonable feature I do admit.
But the amount of bellyaching over this by many is not because people want grammatical errors in their tweets but because it has the potential to totally alter how twitter is a primary source record of public statements, and words can be possibly changed over time and thus the connotation/dennotations of those statements can change. Given how much political discussion takes place on twitter and how much of it about how someone said something stupid, that could be a huge shift in the platform's usage.
Newspapers change titles and content of existing posts all the time. There are bots that track this (at least for the big ones like NYT [0]). Bloggers can change content/title at any time. There are no web-wide rules that you must mention your changes and provide a log. Yet people still link to blogs and newspaper articles.
There's also already Twitter username change trackers [1] and Twitter deleted post trackers [2].
And maybe even more commonly on Twitter I just see people taking screenshots of tweets.
So I don't see this being much bigger a deal for transparency. There'll just be new systems for tracking edits if archive.org and screenshots aren't good enough.
I assume twitter will absolutely show a 'this tweet has been edited' button that brings up all the revisions of it, opens itself up to too much bait and switch nonsense otherwise.
They've done the user centric research & every user says they want to be able to change what they say online & escape their bad mistakes. The team has resounding data pushing them this direction. And it will be immensely immensely detrimental to the platform/service/medium.
The fact that users don't want their edit history visible is precisely why making it visible would be a useful tradeoff. Sure, you can fix your mistakes — but you have to balance whether the fix is worth sending up a flag (e.g. "This tweet has been edited") that may draw even more attention and scrutiny to the original tweet.
It's like the "Hidden replies" feature [0]. Sure someone can maliciously use it to hide valid criticism. But in doing so, they're essentially doing a public curation of "these are the replies which really rankled me"
I think Reddit gets this right and Facebook Groups gets this wrong:
Facebook Groups: member-visible changelog
Reddit: you can edit the comment within 3 minutes with no "edited" sticker, but after 3 minutes if you edit then the comment's timestamp gets an asterisk next to it.
> Newspapers change titles and content of existing posts all the time.
In fact, launching with a headline that matches the URL, and waiting until the page gets scraped/indexed, in order to then start a series of A/B tests for narrative, ranking, clickability, etc is the de-facto process for news orgs.
They will also stand to benefit from this change more than your average shit-poster.
It's really amusing when the URL is something like cats_are_better_than_dogs_science_shows and the article has been updated to be titled "Study shows dogs, cats basically the same".
The biggest difference has to do with how Twitter has been used. Your retweets and likes show up on your profile and thus edits made to tweets may change what appears when someone looks at your profile. Unless they do it in some smarter way than most[0], this could personally devastating to folks because likes and retweets are viewed as endorsements (despite what some think).
The second biggest difference is newspapers tend not to change the entire point of the article in their edits, and aren't out to troll their engaged audience.
0: ie show the edited content on the page alongside the previous version(s) with the same prominence and, most importantly, without having to click anything to show an old version.
>Newspapers change titles and content of existing posts all the time.
Sure, but I trust a newspaper to not swap their article out for something repulsive to troll people. I don't have that same trust in a random Twitter user.
I think the problem with Twitter is that it's built around retweets and response tweets. But if the thing you are responding to upstream can change it creates an issue where something can change after it's been broadcast and retweeted all over the place.
It's entirely different though, you can't compare short form to long form journalism. Twitter is a real time item platform, things happen in an instant. Most people don't even know that NYTs changes stories and it's actually a big deal that they do and that they are generally not transparent about it, which is why such third parties do track it. An edit button on Twitter is a horrible idea. If they do it, they better put a time limit on it, that's 30 seconds to a minute.
You're talking about transparency, which is important, but it's not user experience. We know how to make edit buttons that have transparency, just have an 'Edited' link and make the edit log accessible. Easy. This is not an engineering problem.
The problem with the edit button is the abuse vector, what the low friction path is, the effect of the feature on the community, the whole user experience.
It wouldn't even be right to compare Facebook's edit feature to anything Twitter does, because the communities and the mediums are wildly different.
As an example of unintended consequences, viral tweets are often embedded on articles. Editing them via an account takeover or rogue Twitter apps (if not handled correctly) would lead to a lot of crypto ads reaching Buzzfeed.
This will absolutely be a self-own by Twitter: publishers will switch to screenshots rather than risk inadvertently showing Goatse to their audience.
Free business idea to HN: create a WordPress plugin that autogenerates Twitter screenshots & includes the alt text for embedding. Perhaps the @picaso_me bot can grow into this niche.
There was an amazing account on reddit for a while whose name was something like EditedToMakeYouLookStupid (that wasn't quite exactly it, though, I don't think) that would set up traps and then, well, edit their comment to make your response look stupid ;P.
Edit, as a Twitter feature, has been weaponized. It's now a symbol of how Twitter has stagnated and failed to evolve into a more relevant (and potentially profitable) product. Even though it has been in the works for more than a year, once edits are launched many people will credit Elon for "shipping". The Twitter team would do well to launch this ASAP and mark that as a new chapter for Twitter.
The thread yesterday about Elon getting added to the board was full of people saying that Elon should push to get an "Edit" feature added and then leave. I would be astonished if people aren't already crediting Elon for this.
Is it, though? It's not just about exposing an extra API endpoint which updates a DB row. There's definitely a lot of other business & technical design decisions to be made, such as retweets and quote tweets. "Since last year" could mean about six months, which doesn't sound that unreasonable for a pretty major feature (and one that gets a lot of press).
Infuriatingly they “announced” it on April 1st. Everyone thought it was a prank. And then even a product manager started another thread asking what people wanted and referenced the April 1st tweet so nobody could take him seriously.
The fear of an edit button is that you quote someone saying you totally agree and then they edit it to say "Aborted babies taste great" or something absurd.
I just hope they just keep a complete change log of any edits.
I don't think that'd work - a lot of what drives twitter traffic is the immediacy of the timeline. If you see a tweet and can't react to it, chances are you'll never see it again.
I’ve never understood the constant calls for this. It feels like a meme more than a feature people really need. Is it that hard to proofread 140 characters? What other common use cases are there that don’t impact the conversation in replies?
> It feels like a meme more than a feature people really need.
And anyone can always delete and retweet, too. The "problem" this solves is that narcissists measure their self-worth by tweet metrics and are desperate not to lose them.
It's not a coincidence that this is happening as Musk buys his way onto the Twitter board and riles up his followers about his first pet feature.
Twitter says it's been working on it since last year.
But as for the whole "3 days before he bought the shares" - He's been accumulating for far longer than that and even the CEO of Twitter mentioned they were in talks with Elon about being on the board "in recent weeks" [1]
With that said, yes it may be manufactured coincidence mostly driven by the new CEO change and not by Elon buying shares
An edit button without reseting the likes/retweets counts is dangerous as it can be used to bait and switch or even change the complete history of a user depending on who's looking at it. An edit button that does reset the likes/retweets is equivalent to delete/re-post which already exists.
How long can a Twitter edit button take to develop, test, implement and rollout? Twitter claims to have been working on this since last year and it’s now April so they’ve already been working on this for 3+ months. It’s hard to understand how such a simple feature can take so long to implement for a company like Twitter which has 7,500 staff. https://www.macrotrends.net/stocks/charts/TWTR/twitter/numbe...
A feature like this has three main variables to figure out and any functional PM or company should be able to answer these questions and test the preferred options with users in a few weeks at most.
1) period of time a Tweet is editable for: a few mins or forever? As Tweets are retweeted by others and embeddable, making them editable for only a few minutes after they’re first posted seems like the best approach as users who are retweeting don’t want a Tweet to be able to be edited to have a very different meaning.
2) is the Tweet ‘live’ and able to be retweeted while it’s editable or is it only live after the Tweet is no longer able to be edited? I prefer the latter approach as it makes the implementation a lot easier and essentially maintains Twitter as it currently works.
3) how to indicate a Tweet has been edited - adding an “edit history” link to Tweets only if they’ve been edited and linking these to a page showing the edit history of the page URL in simple coloured diffs seems to answer this question.
Well, for one, not all 7500 of Twitter is working on this :P And this kind of thing is less “hey users which do you like?” and more “ok what would the worse trolls on Twitter do with this feature? How can we minimize misunderstandings while providing value? Should we ship an edit button or some other implementation?” These conversations take time.
There's a ton of social aspects and impacts that have to be considered and tested, and I think you've only scratched the surface. An edit function would add a substantial amount of new UI/UX and interface elements. There are still Twitter features (Spaces, bookmarks, many of Twitter Blue's additions) that have been on the smartphone app for year(s) that are still inconsistent or half-baked on the web app (and sometimes vice versa). And there would have to be substantial additions/reworking to the API.
On a more technical issue, no doubt much of Twitter's architecture (e.g. caching) has been built with the assumption that a tweet's content is immutable. I imagine that breaking that assumption impacts a lot of underlying systems.
* How do you show that a like, Retweet or reply was in response to an earlier version of a tweet that might be vastly different from the tweet you’re seeing now.
* Do you reset a tweet’s likes and retweets when it’s edited? If not, what’s the policy on selling the edit on a viral tweet?
* How many times can you edit?
* Are older edit versions persisted or deleted?
* What’s the policy on a tweet that violates the ToS and then is edited to not violate the ToS?
I’m sure there are more, but it does get complex at scale I’m sure.
Yes - but most of these are only issues if you allow Tweets to be editable forever. If they’re only editable for a few mins after posting and are not live and able to be retweeted and embedded until the editing window has closed, then most of these issues can be avoided.
Wouldn't that handicap Twitter as a source of breaking news? I suppose you could have a "Make live" button to skip the edit window, but that's what the site effectively has currently.
Also, I think people typically want to edit in response to something pointed out in a reply. Like if someone tweets "The Earth is crowded! 7 billion people and counting..." and in a reply someone links to https://www.census.gov/popclock/world and points out there are ~7.8B people, the author might want an Edit button to keep the intent of the Tweet, but clarify the info:. For example, revising it to say: "The Earth is crowded! 7.8 billion people and counting..." An edit button that disappears once the tweet goes live wouldn't address that problem.
This completely ignores the complexity of scale. For one given tweets up until now were immutable there are likely tons of systems with that assumption baked in that you now have to reevaluate/rework. Think caching layers and data storage. Do they need to store an audit history (I assume this requires a few different legal teams in different countries)? How does that scale? The picture is far bigger than the three main variables you mention.
Better be sure about it before you roll that out, because there is no going back.
Maybe this will be an unpopular opinion, but I don’t think Twitter should do it. I don’t see people leaving Twitter because of this occasional inconvenience.
It seems like there is little upside for Twitter doing this, and there is potential for misuse. People are already used to how tweets work and are mostly satisfied.
I prefer knowing that something I already read (and potentially liked) will not be changed later. If the author wants to make a change, it is easy enough to delete and tweet again, or just reply to your own tweet saying “whoops, I left out this word” or whatever.
I think edit option only for 1-2 min on creation of tweet might not be such a bad option. I have myself found wanting this feature to fix grammer or typo or forgot to add a certain #hashtag, etc
Not the original commenter but I guess depending on the implementation many edits might have already been made. Are those rolled back as well? What about tweets that reference the edited tweets, which might not make sense now?
176 comments
[ 3.4 ms ] story [ 210 ms ] threadOf course, this will only work if old tweets can't be edited. Can't imagine anyone being happy to have their Twitter embeds suddenly say something that they didn't before.
If Twitter doesn't implement this, it sounds like a great product to built a small SAAS around: blogs, newspapers, even governments often embed tweets so being able to embed without change, could be a sought after product.
Twitter? Adding an edit button is not exactly rocket science. Meanwhile at SpaceX...
edit: I liked your subtle edit below from "paid board seat" to "(ridiculously expensive) board seat". Strange to make it a negative to invest in a company you want to change, like other board members are allowed to invest for their board seats, but Elon isn't. If Twitter does good it will be lucrative, not expensive!
* Do you reset the likes / retweets after an edit? If not, what happens if folks start selling the ‘edit’ on viral tweets?
* What happens if the original tweet violates the TOS but then the edited version does not? Does the original tweet in violation still get persisted? Does the edit get persisted?
You’re onto something. I’m pretty sure that’s going to happen if edits are not limited. We already see “wow this went viral, here’s a link to my album”
I assume they'll have to have things like version control, so all prior versions are readable for crowdsourced moderation, if not then that's a bad decision IMHO; some past decisions of theirs haven't been rational, and seem more fear based - so maybe they won't even have version control in the name of "privacy" - pretending that the person didn't previously said what they said, and that somehow it's deleted off the internet and everyone's memory if they delete the tweet.
I've seen lots of bad hot takes from Twitter users who have no idea how large systems like this are designed. Dumb stuff like "just do a SQL update" or "use React".
Or perhaps like Gmail: stores the message but doesn't send until a few minutes later, at which point it's no longer editable. This would be easier to implement.
Discussion forums have had edits forever; reddit famously lets you go and delete all your existing posts at anytime by replacing them with "edited by autodeleter9000".
I kinda wish they let people add some text instead of allowing editing, like:
I know deletions are still allowed, but they don’t lead to different content.Maybe if you're a famous celebrity etc. but it's scraped mine 5 times since 2007 (and the "expand conversation" links don't work) and I promise you that it has been fairly active ever since I signed up.
Links and uploaded media should not be editable.
https://help.twitter.com/en/using-twitter/twitter-blue-featu...
It took a while for them to add this. And amusingly, comments (including those opening issues and PRs) are a bit more like a wiki than personal comments—if you have the appropriate permissions, you can edit comments by others. For a long time you could do so with no visible record that it was done. Which is objectively terrible but was hilarious if you worked on a team with a good sense of humor and reasonable restraint.
What people really want is not send tweets with typos and grammatical errors, so a good spell checker is what twitter should have built. At the very least they could have started with that, because that's something that is easy to launch and un-launch, unlike editable tweets (which has database-level implications because they will store the history of edits).
I don’t think such a change really matters at a company like Twitter. The problem is more about principle than technicalities.
But the amount of bellyaching over this by many is not because people want grammatical errors in their tweets but because it has the potential to totally alter how twitter is a primary source record of public statements, and words can be possibly changed over time and thus the connotation/dennotations of those statements can change. Given how much political discussion takes place on twitter and how much of it about how someone said something stupid, that could be a huge shift in the platform's usage.
There's also already Twitter username change trackers [1] and Twitter deleted post trackers [2].
And maybe even more commonly on Twitter I just see people taking screenshots of tweets.
So I don't see this being much bigger a deal for transparency. There'll just be new systems for tracking edits if archive.org and screenshots aren't good enough.
[0] https://twitter.com/nyt_diff
[1] https://twitter.com/travisbrown/status/1510985771343351818
[2] https://polititweet.org/
They've done the user centric research & every user says they want to be able to change what they say online & escape their bad mistakes. The team has resounding data pushing them this direction. And it will be immensely immensely detrimental to the platform/service/medium.
It's like the "Hidden replies" feature [0]. Sure someone can maliciously use it to hide valid criticism. But in doing so, they're essentially doing a public curation of "these are the replies which really rankled me"
[0] https://techcrunch.com/2019/11/21/twitter-rolls-out-its-hide...
i.e. more like a built-in 'delete and re-post', but with preserved metrics.
Facebook Groups: member-visible changelog
Reddit: you can edit the comment within 3 minutes with no "edited" sticker, but after 3 minutes if you edit then the comment's timestamp gets an asterisk next to it.
In fact, launching with a headline that matches the URL, and waiting until the page gets scraped/indexed, in order to then start a series of A/B tests for narrative, ranking, clickability, etc is the de-facto process for news orgs.
They will also stand to benefit from this change more than your average shit-poster.
Screenshots of tweets seems like the worst way to "archive" them, since the screenshot can be trivially faked.
https://shashiirk.github.io/fake-tweet-generator/
The second biggest difference is newspapers tend not to change the entire point of the article in their edits, and aren't out to troll their engaged audience.
0: ie show the edited content on the page alongside the previous version(s) with the same prominence and, most importantly, without having to click anything to show an old version.
ETA: typos.
Good. Maybe people should spend a moment figuring out whether the person they're retweeting is an asshole before giving power to their voice.
Sure, but I trust a newspaper to not swap their article out for something repulsive to troll people. I don't have that same trust in a random Twitter user.
https://www.boredpanda.com/people-edit-original-questions-an...
https://www.reddit.com/r/CasualUK/comments/fo6pki/ask_me_a_q...
You're talking about transparency, which is important, but it's not user experience. We know how to make edit buttons that have transparency, just have an 'Edited' link and make the edit log accessible. Easy. This is not an engineering problem.
The problem with the edit button is the abuse vector, what the low friction path is, the effect of the feature on the community, the whole user experience.
It wouldn't even be right to compare Facebook's edit feature to anything Twitter does, because the communities and the mediums are wildly different.
Free business idea to HN: create a WordPress plugin that autogenerates Twitter screenshots & includes the alt text for embedding. Perhaps the @picaso_me bot can grow into this niche.
For example, display added text in italics and deleted text in strikethrough. (Like some legislatures do: https://www.ncsl.org/legislators-staff/legislative-staff/res...)
And show this view by default, so that users must read through the old text when looking at the new text.
Things get more complicated with multiple edits to a single tweet. You could possibly extend the approach, or you could allow only one edit.
https://www.reddit.com/r/Music/comments/56cdgm/ama_im_really...
https://twitter.com/elonmusk/status/1511143607385874434
Because we live in a hell world.
That gives me a strong feeling of "if you're paying us, we'll let you rewrite history."
Just an absolute clown show.
I just hope they just keep a complete change log of any edits.
Not a bad idea.
Another alternative, you get say 5 minutes to edit then it locks.
Then it's not different from status quo of deleting the tweet and rewriting it, is it?
I suppose the event of being amputated is appended to your life, but your original limb was deleted.
And anyone can always delete and retweet, too. The "problem" this solves is that narcissists measure their self-worth by tweet metrics and are desperate not to lose them.
It's not a coincidence that this is happening as Musk buys his way onto the Twitter board and riles up his followers about his first pet feature.
But as for the whole "3 days before he bought the shares" - He's been accumulating for far longer than that and even the CEO of Twitter mentioned they were in talks with Elon about being on the board "in recent weeks" [1]
With that said, yes it may be manufactured coincidence mostly driven by the new CEO change and not by Elon buying shares
1: https://www.cnbc.com/2022/04/05/elon-musk-spent-2point64-bil...
This is dumb.
A feature like this has three main variables to figure out and any functional PM or company should be able to answer these questions and test the preferred options with users in a few weeks at most.
1) period of time a Tweet is editable for: a few mins or forever? As Tweets are retweeted by others and embeddable, making them editable for only a few minutes after they’re first posted seems like the best approach as users who are retweeting don’t want a Tweet to be able to be edited to have a very different meaning.
2) is the Tweet ‘live’ and able to be retweeted while it’s editable or is it only live after the Tweet is no longer able to be edited? I prefer the latter approach as it makes the implementation a lot easier and essentially maintains Twitter as it currently works.
3) how to indicate a Tweet has been edited - adding an “edit history” link to Tweets only if they’ve been edited and linking these to a page showing the edit history of the page URL in simple coloured diffs seems to answer this question.
On a more technical issue, no doubt much of Twitter's architecture (e.g. caching) has been built with the assumption that a tweet's content is immutable. I imagine that breaking that assumption impacts a lot of underlying systems.
* How do you show that a like, Retweet or reply was in response to an earlier version of a tweet that might be vastly different from the tweet you’re seeing now.
* Do you reset a tweet’s likes and retweets when it’s edited? If not, what’s the policy on selling the edit on a viral tweet?
* How many times can you edit?
* Are older edit versions persisted or deleted?
* What’s the policy on a tweet that violates the ToS and then is edited to not violate the ToS?
I’m sure there are more, but it does get complex at scale I’m sure.
For someone like Kanye, I’m sure that’s enough time to get pretty good traction.
Wouldn't that handicap Twitter as a source of breaking news? I suppose you could have a "Make live" button to skip the edit window, but that's what the site effectively has currently.
Also, I think people typically want to edit in response to something pointed out in a reply. Like if someone tweets "The Earth is crowded! 7 billion people and counting..." and in a reply someone links to https://www.census.gov/popclock/world and points out there are ~7.8B people, the author might want an Edit button to keep the intent of the Tweet, but clarify the info:. For example, revising it to say: "The Earth is crowded! 7.8 billion people and counting..." An edit button that disappears once the tweet goes live wouldn't address that problem.
Better be sure about it before you roll that out, because there is no going back.
Maybe this will be an unpopular opinion, but I don’t think Twitter should do it. I don’t see people leaving Twitter because of this occasional inconvenience.
It seems like there is little upside for Twitter doing this, and there is potential for misuse. People are already used to how tweets work and are mostly satisfied.
I prefer knowing that something I already read (and potentially liked) will not be changed later. If the author wants to make a change, it is easy enough to delete and tweet again, or just reply to your own tweet saying “whoops, I left out this word” or whatever.
1. Copy the old tweet content
2. Delete the tweet
3. Start creating new the tweet (but not finalized)
4. Paste the content
5. Edit the content
6. Finalize the tweet (equivalent of Submit button)
VS
Edit will look like:
1. Press edit
2. Edit the content
3. Finalize (or Submit)
But, like FB, there needs to be a visible edit history and we need to know to what version of the post did people respond to.
I hope this new infusion of ownership and oversight brings fresh ideas to a service that has remained stagnant and is very risk averse.