Twitter’s deletion of its FB app caused old cross-posts to temporarily disappear
techcrunch.com
techcrunch.com
Source: https://www.axios.com/how-cross-posted-tweets-disappeared-fr...
The Hacker News title should be updated as well.
> TechCrunch also heard from a source with understanding of the situation that the deletion of the app took Facebook by surprise, as well as the fact that Twitter didn’t immediately tell them to restore the content
These statements seem to indicate that Twitter was to blame for removing their app. However, from the axios link:
> ...the Twitter app for the Facebook platform was essentially made useless earlier this month once Facebook officially removed the ability to cross-post.
> With the app's sole function eliminated, Twitter decided to delete it from the Facebook platform, having no reason to think that doing so would remove old tweets that were cross-posted.
> It's not clear whether Facebook knew this would happen, either.
Looks like Facebook did not know that Twitter would delete their app, and Twitter did not know that deleting the app would delete the content. I can understand why Facebook would have such a policy for deleted apps, and I can understand why Twitter doesn't want to have a "dead" app that's active on the Facebook platform. They probably just ought to have spoken to each other.
Present situation leads me to believe otherwise..
Twitter probably should have just done the dive to verify what deleting an app does. I would guess this is mentioned somewhere in the docs.
Perhaps someone who is a facebook app developer can chip in to say if this info is in their docs somewhere...
We simply turned the apps off (don't remember exact wording). Last I checked my dev email, I was still getting emails from Facebook warning of API changes and informing of our games' metrics and activity (!), so it is true that deleting an app is cleaner.
This is a great example of big tech companies not thinking about the consequences of their actions.
With only a couple months until election season, Facebook is now literally editing politician's pages and deleting the campaign material they've created to communicate with the voters in their districts.
In what way, AT ALL, was this this feature good for FACEBOOK?
It meant people could be “on” Facebook and “active” without ever actually logging in or using the app or engaging in any way.
Why would they want to just be a dumping ground for copies of another social network’s posts?
I guess I’m surprised this was allowed in the first place, or at least not removed many years ago.
People do regularly mirror repositories to github, and there'd be serious problems if github prevented that or somehow tried to force you to have your github repository as the "canonical" home of your project.
That’s not really what Facebook is for, it’s not supposed to be ANOTHER place to find your tweets.
I’m trying to come up with a better analogy but I’m not doing very good.
Auto posting tweets so that people can use Twitter and still appear to be on Facebook doesn’t help that mission.
Facebook was not losing by this. If someone had connected their Twitter account, it meant they cared enough so they would be visiting Facebook, even if it was sporadically to check how their posts were doing and to see the necessary ads.
http://v3.designforfun.com/images/portfolio/client/2009/face...
Anyway, I'm not particularly upset. Seems more like Facebook shooting themselves in the foot, as now I really have no reason to visit the site.
When a post from two weeks ago didn't show up, I should've listened to my gut instinct to grab a recent dump of my FB data.
As the years march on, I think we'll see more and more data protection laws in favor of users rather than companies. Data is the new uranium, not the new oil. It needs to be protected and taken seriously.
You can argue that it's Facebook's site and that they can do whatever they want. I argue that they cause material harm to users by removing years worth of social interactions among their family and friends. Emotional harm is still harm, and it should be treated as such.
In other words, it's a shame that they didn't catch this use case sooner, but that's really not the users' problem. The fact is, users were doing this. And now that they've been doing this, it's unethical to wipe all of their conversations just because you feel you can.
The largest companies are mostly monopolies. When there's nowhere else to go to have a conversation with the people you care about, what are you supposed to do when companies suddenly tell you that your conversations are violating their terms of service?
“Data is the new uranium, not the new oil”.
Looks like Facebook is sitting on the largest legal liability of the 21st century.
It’s kind of like being a major manufacturer in the US in 1970...you’re about to get screwed by 40+ years of environmental law (considering the river in Cleveland caught in fire in 1969, one can see why some of these laws were passed).
(river is the Cuyahoga)
How much were they paying for the service?
(But even a 0¢ credit card requirement would have an enormous negative impact.)
Huh? We're talking about Facebook refusing to keep reposting stuff from another site. That seems like exactly the opposite of the stuff Facebook might have some responsibility for, if it did have responsibility for stuff.
I mean, obviously, it's ambiguous whether Facebook is a repository, a gate-keeper, a community, a blogging-site or what. But the situation of stuff being reposted from other sites seems to be the least repository like. Deleting years of stuff that might not exist in other places would certainly be much more draconian.
We're talking about how Facebook not only turned off the API that enabled reposting, but removed all past posts that were made via that API. Which, at that point, was exactly the kind of user data that they had responsibility for.
If this happened to me I'd be extremely upset: I've had many very good discussions on Facebook that it would hurt to lose.
I would not assume Facebook has any duty to maintain historical posts. That would be up to the TOS.
I would want to ensure that FB can't host data which I submit to twitter/reddit/HN without providing me the authority to be able to request its removal from FB.
I'd think that this could set some precedent to allow for this.
Clearly, though, the FB anti-stance would be that it cannot do so in such granular fashion, which clearly is not true given the fact that any data/posts are tied to both source site and source person...
I genuinely don't mean to be snarky, but that's a surprise. I have literally zero Facebook discussions I even remember, let alone "would hurt to lose".
No, they cut off the API. The old posts were still there, and tied to the apps that were used to send them. When Twitter then killed their app (apparently, because it wouldn't be usable going forward and in ignorance of the way existing posts were tied to it), then the old posts that were made through that app were deleted. Not because they had used the API, but because they were associated with a removed app.
This is exactly what the report says they did. The original post may exist on Twitter, but the conversation that happened on Facebook does not.
I just checked my feed, and my Twitter posts do seem to still be on my FB profile, at least when I look at it.
Users paid for the service with their data, which Facebook can turn into money. The idea that users don't "pay" for Facebook ignores what Facebook does.
They took the posts down, with zero warning, zero hey we're deprecating this, back up your things. If your stuff was taken down and you care about it, one would hope that you would trust facebook less.
They try to stay ahead of troves of data going dark so they can save them somewhere first.
Why would they not have the right to control content posted on their platform?
If someone lets me and all my friends communicate by posting signs on their front lawn for free, should they have an obligation to keep my signs on their lawn until I decide to remove it?
Saying it causes emotional harm is a pretty weak defense. Basically any action can cause emotional harm to someone.
My "wall" is not generally referred to as "FB's wall documenting ISL's interaction with friends", but rather, "ISL's wall".
If a society believes a thing to be true, it will tend to vote in politicians to make it so, whether it is presently contractually written that way or not.
If you're a little more permissive and let someone live in a little hut on your lawn, then after several decades, they'll own the lawn, and you won't.
I'm guessing that since their fleet of lawyers wrote that wall of legalease everyone must have clicked on, they are pretty sure exactly what their rights are w.r.t who owns the contributions voluntarily sent to their servers.
One of the central questions of our generation is, who owns the data we create? Do we, or do the corporations?
Seeing as Zuck was dragged in front of congress back in April, I think questions like these will only become more important over time.
Why should Facebook care where the post came from? It's still a post, and still gives Facebook the same access to the same data that it harvests and monetizes in the same way.
I used to use something like this (it was a third party app that only cross posted things tagged #fb) and honestly it made me use facebook more. Now my facebook is a ghost town because fuck if I'm gonna post something on two different sites when something happens to me in my life.
I can definitely see why people think this leads to lower engagement with facebook, but I actually expect that for the vast majority of users who used it, it was the only thing keeping them using it.
One of my friends only cross-posts from Twitter. He doesn't engage with Facebook, but if his posts stop appearing in my feed, I engage less.
I probably will basically not use Facebook now. I’m not super mad about it or annoyed, so hopefully this doesn’t come off that way - I’m pretty nuetral about it. I am perhaps abstractly annoyed that a comment someone made on my wall is now gone maybe. But to directly answer your question, I think at least with me, this was a very useful feature for Facebook. Again, I don’t care that I won’t be using Facebook anymore now, so just one instance of this being useful for Facebook and neither here nor there for the person.
Moving from one Mastodon instance to another involves starting over from scratch, just as much as moving from Twitter to Mastodon does. Federated portability sounds like a nice goal and I hear that some people are working on it, but nobody really knows what form it will take.
Active users? https://www.statista.com/statistics/264810/number-of-monthly...
This sums up the rationale of the platform economy rather nicely.
> It meant people could be “on” Facebook and “active” without ever actually logging in or using the app or engaging in any way.
Why would they want to just be a dumping ground for copies of another social network’s posts?
I could naively answer "because you actually want to provide a useful service for your users and not just make them hooked to your platform" - but yes, in the real world, your question is justified.
If you are really that attached to your Facebook page then do a backup.
If you really think it's unacceptable that Facebook can do whatever they want with your data (and they can), just don't use it.
https://help.twitter.com/en/managing-your-account/link-twitt...
and there was an HN story about this 19 days ago
https://news.ycombinator.com/item?id=17721986
why is TechCrunch just picking up on this now?
“Reached for comment, Facebook says it’s aware of the issue and is looking into it.”
It looks like a mistake, FB is actively communicating it's trying to restore the content but still everyone is witch hunting.
I thought people here were more sensible to the fact that developers at all companies leave bugs/undesired behaviour behind them.
This may very well have been spec'd and developed 10 years ago, and now the reality have changed, but not the code. Because let's face it, it's not exactly an everyday use case.
I feel like that's a pretty obvious problem you would anticipate and write tests for, isn't it? How did something like this manage to get to public rollout?
Facebook (and others) know that giving you that feeling of ownership and customization keeps you coming back. The Twitter Avatar, the Facebook profile, the MySpace page. All those posts on your profile add to your collection, your customization. They let you control that little bit so that you feel like you own it in some way. Did you love that old car so much that you put a bumper sticker on it, or did putting a bumper sticker on it make it more "yours" and so you loved it more? Same idea.
But you don't own it. Nothing you post on those platforms belongs to you anymore. It's theirs, to do with as they wish. And in this case, Facebook decided that retroactively deleting all that content that lets you be aware of other platforms was good for their bottom line- do they did what they wanted with their property.
Facebook deleted their posts. You only thought they were yours because Facebook makes more money if you think that.
https://en.wikipedia.org/wiki/School_speech_(First_Amendment...
Alex Jones and other fringe elements with disproportionate large reach wouldn’t exist on a regulated utility because the advertising practices that allow Facebook to monetize Jones would likely not exist.
sm seems to fit out in terms of scale and importance. We don't need social media like we need roads. Call me a crotchety old guy but I think quite the opposite, we'd all be better off with less social media. No need to enshrine it with the status of "utility".
I'm still uncertain where I stand on social media being a utility though; I'm not sure what that would imply.
The clean looks of Facebook and Twitter have oddly kept them on top.
But yes, to your point, you aren't the customer. You're the product (unless you're buying adspace .. in which case you're both).
I recently started my own Mastodon instance and am just following random people on random instances and so far have some pretty funny/interesting feeds. It's actually a pretty friendly little fedverse.
Having said that, I really hope this is not the new policy from FB. Not only posts gets deleted but the conversation around it are also getting deleted. Many high profile people have setup FB accounts so everything is cross posted from Twitter. If this is new policy, it’s going to be the most boneheaded one so far from FB. So let’s hope this is just a bug.
Had this happened to just about anyone else, the response would have been "whoops this obvious mistake really sucks!"
But it's Facebook, so HN's reaction is "wow fuck these guys for this clearly purposeful event!"
So users can profit from their content? They won't profit from their content, it's nearly worthless per unit. It's only in the vast total sum that there is any value, and only because it's held by one entity. If you distribute it per post or per person, it's very close to meaningless. $20 billion is a lot per year to one entity; $20 billion per year to two billion people, is not.
If you were generous with Twitter and assumed they can generate $400m per year in profit and then chose to distribute that to the userbase, you're talking about ~$1.30 per active user per year. If you tilt the value distribution toward the most active, famous users (whose tweets are far more valuable in regards to monetization), then you're just destroying the distribution premise to begin with by concentrating the wealth in a slightly different manner: Taylor Swift doesn't need to worry about earning $35,000 per year from her tweets, and Joe Smith getting $7 from Twitter for their 1,278 economically worthless posts also does not matter.
Ultimately you just end up with a YouTube outcome if you value tilt. A small group of rich users earning millions of dollars on FB or Twitter, and everybody else entirely failing to earn even minimum wage for the time put into their economically low value posts. How does that really change the equation vs the people that are capturing most of the value today? Taylor Swift can already capture the value generated by Twitter and Facebook, she can buy the stock.
steemit.com has some ideas - but not perfect ones to be sure - to build incentive based content.
Where is the $400M twitter profit from?
Ultimately it is not about the direct monetary ROI for this in my mind. But empowering choice of you as to who can use your data (and knowing what they are legally able to do with it) is critical. Thinking about reducing the supply of data will likely spike the cost of that data as well. Most of what is sold is not content - but your use of the platform and _who you are and what you do there_ and predicting things about you - not what you post.
All your content is still yours, and with the GDPR you can request a copy of it at any moment.
However, Facebook has no obligation to preserve copies of your posts, there is no provision in that respect in the TOS, so if you don't actively preserve your posts, that's your problem.
I’m mostly joking. Who makes their own websites anymore?
Frustrating seeing users caught in the crossfire of policy updates for tech giants.
Like the one all the content was actually posted on originally.
I guess I don’t see why this is such a big deal other than as a generic opportunity to be mad at Facebook.
GAE/Github + Hugo is remarkably friendly and low maintenance.
And how do you make your static blog social? No Mastodon and other decentralized networks are the real alternative to Twitter,Facebook, Reddit and co.
You can't really. That's the price you pay for running your own stuff.
You can fake it a bit with widgets or injected comments, but ultimately that's a pretty pale imitation of the social experience.
The upside, though, is that you get to keep all of your content forever, and you get to display it the way you want.
Google stopped mining it last year: https://blog.google/products/gmail/g-suite-gains-traction-in...
> Google stopped mining it last year: https://blog.google/products/gmail/g-suite-gains-traction-in....
Google stopped mining it for ads, but it's disingenuous to say they stopped mining, period [0]. How do you think those Gmail suggested replies are generated?
[0]: https://variety.com/2017/digital/news/google-gmail-ads-email...
Would you stop using Protonmail if they offered the same feature?
I don't really see how running the contents through an quick reply generation function is different to running it through whatever function prepares the body for display, here. An implementation of that which involves wide-scale mining of email content is imaginable, but not necessary.
You can be locked out of your account for arbitrary-reason-of-the-day and good luck finding a contact number or email address for support.
If email is important to you, you probably shouldn’t be using a “free” service.
This, in turn, means that even when considering non-free email service providers, you need some reassurance that provider will be around and will not institute crazy policies in the future. Which favors large providers like Google over smaller ones, and reinforces the lock-in.
- Notify your contacts of your address change.
- Set your old address to forward to your new address and always reply to emails from your new one.
- After 2 months, set your old email to auto-reply to all messages saying that you have changed your address and the old one will soon be deactivated.
- Finally, turn deactivate the old email.
It's work, but I've done it on multiple occasions. I wonder if there could be a service that could automate the entire process, since it could all be done via IMAP.
I have documented a few cases of such unnoticed termination in my blog post:
https://paradite.com/2016/02/18/stay-independent-problems-wi...
If you can't be bothered, just use a provider who will set it for you... like Google.
“just use an existing service / do what everyone else is doing”
probably isn’t very representative of the entrepreneurial / hacker spirit.
Over time, I've been replacing various Google services with smaller alternatives. At this point, it's down to Youtube, which is pretty hard to beat. But it feels a lot better to provide usership to smaller businesses that could use it more.
While this change was definitely made in the interest of preventing spam, it also made Facebook worse as a product. We've had plenty of requests from our users who want a "share an image of my book passage to FB" feature. Not possible at all with their new stricter APIs.
Facebook was the most useful pre 2012. You could create widgets that changed the look of part of your profile. Facebook games were interactive and not siloed like they are now. People posted status updates and they appeared in the news feeds of others.
/one can dream
On the other hand there has been a lot of concerns of illegitimate accounts spreading propaganda, affecting US elections, and it may be harder for Facebook to validate accounts that basically only exist on Twitter.
Sounds like it may be a bug?
"In theory, the API changes should only have prevented Twitter users from continuing to cross-post their tweets to Facebook automatically. It shouldn’t have also deleted the existing posts from Facebook users’ profiles and business users’ Facebook Pages."
oops
Also, cross-posts from 3rd parties make up the majority of crap on my timeline. Happy to se it go and happy to see people being exposed to the fragility of relying on these free services as a backup service for anything they consider important. I hope it prompts people to consider their relationship with these types of platforms and what is being exchanged between you and them exactly.
/s
People will notice things just being different. Good object lesson, and I hope robust discussion.
If yes, why is nobody sueing?
Either this has been fixed, or my account hat not yet been affected.
And then there's often FB-based discussion about them.
Microcorn!