Hacker News new | past | comments | ask | show | jobs | submit login
Twitter prepares for cull of inactive users (bbc.com)
323 points by alexanderdmitri 51 days ago | hide | past | web | favorite | 253 comments

I wonder what will happen to the twitter accounts of folks who have passed away? Eg. Aaron Swartz: https://twitter.com/aaronsw https://twitter.com/aaronsw_hv

> The cull will include users who stopped posting to the site because they died - unless someone with that person's account details is able to log-in.

Yeah, this is bad.



Archive Team is making an effort to archive twitter accounts of dead users. Please see these links:




This was my first thought at reading the headline before clicking through. Thank you for sharing the form from archive team.

My first thought was Hal Finney, whose account I occasionally browse for a reminder of what twitter was like 10 years ago.


>Harold Thomas Finney II (May 4, 1956 – August 28, 2014) was a developer for PGP Corporation, and was the second developer hired after Phil Zimmermann. In his early career, he was credited as lead developer on several console games. He also was an early bitcoin contributor and received the first bitcoin transaction from bitcoin's creator Satoshi Nakamoto [1]

This is only one example. So many historical figures gone to free up some vanity names.

[1] https://en.wikipedia.org/wiki/Hal_Finney_(computer_scientist...


"Just back from walking the dogs"

"Running bitcoin"

This is kind of how I use Twitter, lol. I use it as a place to post a random thought, what I'm thinking about, what I'm doing, etc. I don't post much, but I've been trying to get in the habit of posting more.

I worked with a guy who died of a heart attack. Every year LinkedIn asks me to congratulate him having worked X+N years at the company. You have to wonder when they'll give up.

I follow an acquaintance on Twitter who set up an auto-tweet while he was alive. Now, after his death, he still tweets whenever the weather drops below a certain temperature in Scotland.

Now I'm wondering what your acquaintance's tech stack is, and more generally, what the most reliable way to keep running scheduled jobs after death is - since Twitter won't remove accounts that still log in, I kind of want to set up a cronjob to keep my username active for as long as I can. (Assume that the job itself doesn't change, i.e., there's no need to move away from a deprecated API.)

Maybe Lambda + CloudWatch Events? AWS probably won't go away for decades.

Does your credit card automatically close when your death is reported to the financial system, or can you direct it to stay open in your will + set aside a bit of money to pay off the couple-cent bill in perpetuity?

There was a case of a military veteran recently in the States who lay dead on his kitchen floor for up to three years. He was still getting some sort of pension/payment and the rent was getting auto-paid from that. They only found him because of a check on tenants in a building/area who weren't drawing any water.

In a case like that, I suppose a debit card or credit card with an automatic transfer would stay active for some time.

> They only found him because of a check on tenants in a building/area who weren't drawing any water.

Wow. Some leaky faucets or automatic plant watering system would have kept him undiscovered for even longer.

It’s rare but not unheard of for people to be found decades after death.


More humorously I remember reading about a German guy that forgot where he parked his car and it sat in an office garage for 20 years or something.

There was a building in my city that was constructed in the 70's to be a shopping mall, but the owner got bankrupt in the meanwhile, and the building was repurposed into a parking lot that run for a couple of years.

By 2014 is was imploded. There were around 30 cars still parked there.

Not sure it would matter if the CC was closed or not. If you're running a Lambda a few times a day you would fall into the free tier for... ever until they change the free tier.

So the question would be does your account stay active despite an expired credit card which is never invoked cos you never put your account into debt?

"The AWS free tier will expire 12 months from the date of when you first sign up."

Some AWS usage is part of the free tier even after 12 months. "The Lambda free tier does not automatically expire at the end of your 12 month AWS Free Tier term, but is available to both existing and new AWS customers indefinitely." - https://aws.amazon.com/lambda/pricing/

I made some Alexa skills years ago that are still up running on Lambda. They've never been used enough to get me charged.

What if AWS changes their privacy policy

What sort of change?

Any change that prompts you to re-accept?

Just like twitters in this instance.

I think in the case of AWS, your services would continue to run as long as your bill is settled. It would only affect you if you logged in.

So if you fell under the free tier then I'm curious if it would continue to run indefinitely.

He probably used IFTTT.

I've tried ifttt. The Twitter recipes will stop in a few months for me. Have to go in and turn it in manually every so often. Just trying to repost from Twitter at a location to WordPress. ymmv

If you setup IFTTT then until that stops running which could be never??

Unfortunately, API tokens are not the same as user logins, and very easy to separate out from user activity. Depends on some policy whim to be counted or not as user activity. Not dependable afterlife.

I have a friend who passed away from cancer a few of years ago. I still get these notifications for him. I was trying to call someone recently and my phone suggested my dad who has been gone for a couple of years. We need a better mechanism to gracefully transition accounts of individuals.

I still get a reminder in Facebook to celebrate the birthday of a buddy who passed in 2007...

I realize you probably don't want to for personal reasons, but have you removed your father from your phone book?

I didn't remove mine for the longest time, though his number has long been reused by someone else; I only did when I finally got a notice that that person joined Signal and Telegram. Too spooky to see "[your father] has joined Signal!"

I think there is a Russian Black Mirror reboot where this is the premise of the first episode.

This is kinda like the 'isMalicious' bit proposal on TCP / IP.

LinkedIn has a "report a member as deceased" page that puts the member's account into a commemorative state. Some people might find updates like those to be a painful reminder.

Thanks I did that, took less than 24hrs to remove. Kinda sad now as he's just disappeared, rather than someone I'm still connected to.

Seems like they could avoid this problem, keep old accounts, and still free up the namespace by just renaming all these accounts @_archived_originalname (or something better).

Maybe with some kind of process to re-activate the account under the original name (or a new one if it has since been taken). And some kind of hands-on resolution process for people who accidentally got deactivated. Feels tractable.

> Twitter said the effort is not, as had been suggested by some users on the network, an attempt to free up usernames.

> The site said it was because users who do not log-in were unable to agree to its updated privacy policies.

That.. makes sense, but is almost suspiciously proactive.

6 months is quite a bold number, much harsher than domain names would do. I’d get like two years which is probably still tens of millions.

That’s a lot of potential digital content that will go dark.

Does that make sense? Why should a privacy policy be retroactive on things I already posted?

Because Twitter potentially bears liability by hosting them into the future, and is only willing to bear that liability if you agree to their terms.

Citation needed. If they receive a copyright takedown request and don't act on it, then they're liable. Otherwise I don't see where liability comes into play. Please cite court cases showing websites are liable for content posted by its users. Nobody is using Twitter to archive copyrighted material.

It may not be for content, but for user data analysis without consent.

Then don't analyze those records... such nonsense.


This is the exact purpose of Section 230 and why it is so important to preserve it.

Except that Section 230 explicitly states that they don't bear liability.

Does European privacy law have the same provision? They are removing non-US accounts first, after all.

Don't like that idea. There are a lot of direct links across the internet to tweets that would get broken if the handle was renamed.

As long as the id is okay the user name of a link does not matter


Wow, I did not know that! I was worried about which tweet of mine you had found! :)

I regularly visit Adam West's feed. I knew Adam and going to see some of his corny posts (like Moose Wayne) are an occasional comfort and not only remind me of him but remind me of watching the tv series as a child with my late father.

Sadly, I've noticed that DMs don't seem to be forever on twitter as all of our private messages to each other are all gone, the bulk were gone before he even passed away.

Given his account is verified I would hope that Twitter keeps that content as long as the service survives, but for those with family and friends that have passed that weren't high profile, they may also take comfort in going through then posts of those accounts and this is kinda blah.

I get it from a cost view, it costs money to keep all of that data but I somewhat feel that a company that has user created content is socially obligated to maintain that content as long as they continue to have that product or until the user wants it removed whichever comes first.

The issue I've seen with archive efforts that end up on the Internet Archive whether by IA via Archive Team is problems with infinite scroll (and other forms of dynamically loading content via JS), at least in my experience. Though I recall one WARC archiving tool that could achieve it (I believe it required the user to manually scroll the page during the capture).

The Internet Archive uses various Web archiving technologies and processes to capture Web-based resources and is working to improve them on an ongoing basis. This includes support for sites/pages that support infinite scroll via JS. If you have specific requests, bug reports, issues, please send them to info@archive.org and we will try to do better.

Why is it bad? Why should dead people take up usernames? What will happen in 50 years, half of good usernames will be taken by graveyard dwellers?

I don't care about their usernames. Just their content.

Twitter won't exist that long

For every Aaron there are thousands of people who signed up, took a great all letters no numbers username, never posted, and never came back.

They could have the rule be "all inactive accounts with less than X followers".

who determines X? How is that fair to X-1?

Archive team seems like the appropriate end for these accounts. Maintaining accounts for deceased people for the remainder of the lifespan of the internet is not really a sustainable thing.

maintaining accounts for deceased people for the remainder of the lifespan of Twitter is probably sustainable, though.

To what end? You already assume Twitter has a shorter lifespan than the internet. Are you going to be upset when Twitter dies, removing not just all the accounts of live people, but dead people as well?

Also, Twitter may disagree with you on its lifespan. I wouldn't, but I do believe that unless they do start archiving, it will reduce its lifespan, perhaps considerably.

This is why I really don't like the recent trend of third party sites dynamically embedding tweets on their own page when a screenshot plus direct link could have sufficed. The internet has already lost a great deal of its past with many historical image hosts that went down or rogue (looking at you Photobucket). Why are people making the same mistake again?

BTW Facebook have had the option to memorialize the profile of the deceased for a while. Things seems to have worked out fine for them.

The most frustrating case for me is Readability, whcih swapped out the original links to articles with its own link shortener, then went titsup.com

Even migrating those to Pocket (which I hope doesn't do similarly, though it remains all but useless), I've got hundreds of link-shortened articles I need to track down, one by one.

Why a screenshot? That would mess with screenreaders.

Tweets are text, so quote them as text?

Why would is mess with Screen readers? the Alt tag has been a thing for a long time.

Why then embed tweet as an image if you're going to copy its text anyway?

So when twitter changes an API, or is down, and someone happens upon your page they can still see what you intended to include.

Also if you're documenting something like John Smith said "Kill all Muppets, they're taking human jobs" and his publicist goes "John, you really need to delete that tweet" and he does, you can show a screenshot of it to prevent revisionist history.

We're not talking about using the API, but whether to prefer copying the tweet as a screenshot+text or just the text.

Yes. And you could even quote the page with all the twitter styling. Ie quote the HTML and CSS instead of just the tweet, if you really want that twitter look and feel.

Same reason we use Archiving services

doesn't the default twitter embed include the original text of the tweet in the embed code, and then just dynamically load in the like/retweet counts and branding?

at least that's the way it looks when they load on a ridiculously slow internet connection, i've never actually looked at the embed code.

Yes, you're right. Of course, images and quoted tweets also get loaded dynamically, and the text may be hard to understand without those.

The number of links I come across from dead link shortening services is too damn high.

Why not?

If computational resources keep getting cheaper exponentially (even at a low rate), keeping eg 1 GiB around forever only costs a finite amount of money.

There is a limit to the cost of computational resources, the cost would decrease until a certain limit and then that cost will apply until the heat death of the universe.

Yes. Though in our case I feel justified in using the simpler model of just assuming exponential shrinkage forever, because I assume that the end of twitter as a company (or the Internet) will happen before the heat death of the universe.

(Similar to how we often treat integers as unit-sized in asymptotic analysis of algorithms, even though we know that we need a logarithmic number of bits for truly large numbers.)

Space gets cheap. But the namespace for user names gets ever more crowded.

not just the namespace, but the content-space. a service like twitter thives on a feeling if immediacy and involvement, and if you were to constantly be coming across dead accounts (whether the owner is deceased or just otherwise left twitter) i think it would detract from the usefulness of twitter.

whether the accounts are moved to the internet archive or some twitter-hosted archive, the value in twitter is not in old tweets. i think most people would be better off if all of twitter had an automatic expiry.

On a site that only actively presents the newest content, there’s no way you would be constantly coming across dead accounts.

I also find the insinuation that dead users provide no value absurd. Aside from current events, the human condition doesn’t really change, which is why we can still enjoy books written hundreds of years ago.

Why is that bad? Most people probably wouldn't want their twitter account to be their most lasting legacy.

Awful move. If they just remove users with all their tweets it breaks internet and I hate when companies do that. The stated reason (not logged in can't agree on terms changes) seems feeble - there's million ways to handle it better. And why do that starting with Dec 11th, what's up with the rush? Much more sensible way could be something like: set a policy before hand that requires login for example once a year (6 months seems too short) and if account didn't have any tweets or had like 3 tweets and 2 followers, delete it. That will allow to shed completely inactive users and free up some squatted handles, but will not be detrimental to the content.

>And why do that starting Dec 11th

It's tied to someone's year-end KPIs.

This is awfull.

6 months is too short for any account.

I can understand the need to cull some never really used accounts. There are a lot of never really logged on accounts filling up the space. So some weeding of those is understandable.

But not ones that have been used but not that active any more. E.g. dead, seriously ill in hospital, backpacking for a year or travelling to Mars.

Or simple just stopped tweeting for a while or permanently. I am sure many now dormant accounts have tweets linked from pages across the internet.

I do hate services that recycle usernames. So open for abuse.

I have a couple that is merely business or project related that tweet once per year if that. They should not be deleted.

I also have a few that I just created to demo something. After a year or so I can not argue against them being deleted.

I registered a few accounts for my kids, who are still too young to use it but thought I'd ensure they have the option of choosing their name as their handle. Never tweeted, so probably a prime candidate for deletion. Unfortunately.

I/we don't know the criteria they use specifically internally to decide who might get deleted. But I hope it is some sort of scoring based on several criteria. Not just when they last logged in. like:

* how many tweets have they sent

* where the account used over time, not just after registering

* last login

* how many times did they login

* does it have a profile picture

* detailed profile

* have they requested not to be deleted

* are any of their tweets backlinked from other sites

* have x followers

* following/followers is more than x

Twitter is just an internet forum with a handful of famous users, internet forums come and go so do user names.

Twitter is not the real world, despite what Twitter addicts and journalists (who benefit from Twitter because it makes their jobs easier) would like you to believe, most normal people barely use it.

If Twitter disappeared tomorrow I think you'd be surprised how little it would matter.

Twitter et al IS a historical archive of information however - Imagine if Donald Trump's handle was to EVER be recycled, who'd get it? what would the consequences be? What about news pieces that embed or screenshot his tweets?

I feel context is vitally important here, usernames are an important identify in who said what at what time, and can be used "on the record".

> what would the consequences be?

Twitter is not and has never been an officially sanctioned government press release channel. The consequences to government and policy would be trivial to none. Since most of the tweets from Trump are void of fact/research or are bizarre conspiracy theories I suspect a troll taking over the account would go largely unnoticed.

DoJ lawyers have argued otherwise in court.


But what did the court confirm?

In a more practical context I was in CENTCOM supporting troops in Syria when news of complete withdraw stunned the world via Twitter. The military did not act on a tweet. No formal change in policy in the government occurs via tweet.

In a more clear example remember back when Trump ordered, via tweet, the military will ban all transgender persons from military service? The military ignored that tweet without repercussion.

The court confirmed that the Twitter account was used for official communications with the public (and therefore couldn't be arbitrarily restricted).

Does the military act on press releases? I would think both have importance based on how everyone else reacts to them, not because they serve as internal communication channels for the government.

A tweet is not a press release and neither of those are lawful orders. Congressmen, for example, have numerous options for communicating officially with their constituents. That communication is not an indication of lawful actions or policy though.

Its like comparing the correspondence of a legal action from the government to a campaign rally. Just because many people in the public cannot tell the difference does not make them equivalent.

What's the difference between a press release and a tweet? Letterhead? A press release is just an email. What if a press release is published by tweet?

This sounds like an equivocation fallacy. Official letterhead can have legal ramifications that a tweet cannot. A more absurd example is: What's the difference between a tweet and a medical prescription? Letterhead?

The Secretary of the Navy tried to ignore an order-by-tweet last week and he's been fired.

People who don't act on Trump's tweets can do so, but when he follows up by firing people who don't, that's something of a signal that they're meant to be taken as orders.

That's not what happened. The Navy secretary was not fired by Trump or anything related to Twitter. He was fired by the Secretary of Defense after violating a verbal order and making contradictory public statements.

Trump tweeted that Esper fired Spencer but Esper very likely didn't actually have the authority to do so. Probably Esper asked Spencer to resign.

Point is, when you have 1) a tweet about an issue, 2) an official who says "that's not an order, we are going to wait until there's an order" followed by 3) that official being fired, it sure looks like the tweet was meant to be taken as an order.


Esper fired Spenser. Esper told Spenser to write a resignation letter on Sunday and deliver it to Trump the same day. This happened before Trump tweeting anything related. Trump became aware of the termination as a result of the resignation letter and tweeted as a result.

Here is an article where Esper explains why he fired Spenser to Pentagon officials. https://townhall.com/tipsheet/cortneyobrien/2019/11/25/esper...

You clearly aren't an active, engaged user on Twitter - quite the ignorant take you present.

I use it 22 minutes a day on average according to iOS screentime. So I'm well aware of what it is.

You might want to reference average time use of users if you want your "22 minutes a day" to make any point relating to time use.

And to go back to your original comment, please define "normal" people, e.g. your saying normal people don't use it; I'm not going to put more energy into breaking down your statements further.

I feel that you are over thinking this. The email I received (apparently I have a lot of accounts I registered for demos but no longer use) says this:

"you need to log in and follow the on-screen prompts before Dec. 11, 2019, otherwise your account will be removed from Twitter."

So they are sending the email to all inactive accounts, and leaving it up to the owner to perform an action if they want to continue to using it. Apart from the rather short timeframe, it seems like a good way to handle it.

Yes, so far the emails I have received has been for account handles I don't mind if they get deleted.

But like Niemöller's prose, which are next?

I am hoping they are using sensible scoring as I suggested.

Otherwise, if they will keep sending these emails every 6 months, what happens if you miss one of these emails for an account that actually matters. Easily done even if you are not dead, ill, travelling.

I have received the email.

This is actually useful for me. The name of my new company is taken by a 10 year old account that has never tweeted, and I haven’t managed to find somebody at Twitter who can get the name for me.

Does anybody know what’s going to happen to the names? Are they going to become available to register on 11th Dec? I hope somebody isn’t waiting there with a dictionary to squat them all...

Yeah honestly I understand all of the negative effects people are saying but I’ll be so glad if the username I want is released from the user that signed up, tweeted once, got hit by a spam virus in 2011, never tweeted again.

I’ve even got a few reminders in my calendar to check on the username availability on the 11th and 12th. Really hoping it’s a full on user delete rather than just making them inactive

There are lots of permanently suspended accounts that have basically been squatting on a useless name for a decade.

There's zero downside to releasing those, since their old tweets are already gone.

> previously unavailable usernames will start coming up for grabs after the 11 December cut-off - though Twitter said it would be a gradual process, beginning with users outside of the US.

Conversely, it's problematic for me - my Twitter account is up for grabs and someone might impersonate me, assisted by several mentions of the account on old pages.

Why is your account up for grabs? Are you unable to login and confirm you're active?

I'm not active and haven't been for several years ever since Twitter started asking for phone numbers obnoxiously.

Yeah, seems like they have a habit of suspending accounts and asking for numbers. I've been lucky until now.

Seems like you can contact support to get it active again without having to give out your phone, but I understand if you'd rather not use the platform at all.

I’m in a similar situation.

Does anyone know of a service that can be made to automatically pick up a Twitter name as it becomes available?

Same situation here. I've emailed them several times about it with no success. Hoping this pans out.

You just gave out the idea to someone to do just that

I mean domain squatting has been a thing since the inception of the net. Not too far off to think this will happen over and over again.

Apparently, I'm the only one who likes the move. Two reasons:

1. I actually think services should store less about people. It's good that stuff gets removed from parts of the internet.

2. I fully understand the idea of archiving a part of history. But in a way, I treat the websites more like a shopping window and less like articles and artifacts of time: They will be rearranged, stuff is removed, new stuff is added. Nobody would argue that redecorating a shopping window somehow destroys culture or history. If that is important to someone: go ahead and take pictures of store fronts. Make an archive if you want. But don't expect the store owner to keep an accurate history of their shopping window.

After all, most websites are actually the fronts of shops that are open 24/7.

I haven't logged in to twitter in years. I'm not sure if it qualifies but I sure hope my account gets removed without me having to log in and do it myself.

But Twitter isn't a shop. If in 10 years Amazon goes bankrupt (or any other big shop) and it won't get archived, I don't really care. But Twitter has a lot of discussions or other information that may still be important. I think Twitter is more of a library than a shop and you wouldn't throw away books from a library either.

I think there are several perspectives on this:

- For some, Twitter is a chat and chats usually are meant to be short-term with an expiration date.

- Twitter is a store front, because it is used to promote services or products and only the stuff that got posted recently is relevant. Furthermore, some people auto-delete their older Tweets with tools for various reasons.

I think it makes sense to archive Tweets of public figures, especially politicans to hold them accountable in a way or just for historic purposes. This can, should and is done by third party services. But this isn't the only way to see it.

> you wouldn't throw away books from a library either.

While i wouldn't, the libraries do throw books away which aren't being borrowed.

I worked at Twitter back in the early days. I left because I had an inkling there were tons of fake accounts and bot accounts. I thought at some point this would nosedive the company because Facebook’s growth at the time was all about how many active users they had. Twitter seemed to have too many accounts that weren’t justifiably real.

A few years later I Had conversation at another job with coworkers and told them how easy it was to fake followers with $40. Sure enough I found a site that promised a certain amount of followers for $40. I paid the $40 and went from a couple dozen followers to 4,000 over the course of a few days. I think for an extra $60 I could have them retweet something for me as well. At that point I knew my suspicions were correct and knew I made the right choice to leave.

So you left Twitter 5+ years ago expecting them to soon fail, and you feel vindicated by the very-much-still-existing Twitter of today purging some accounts?

In a sense, yes. When I was there the company had a ton of other issues too. Mostly how to monetize the service. They couldn’t figure out how to make money, but they were spending money like they were Facebook.

Facebook at least had ad revenue. Twitter really couldn’t figure out how to sell ads.

The concern was to me, fake accounts and fake retweets/followers creating fake user growth. That’s purely my perspective and likely not how the company saw it. I wasn’t comfortable investing my time into a place that gave me that feeling.

I always wondered why doesn't Twitter do this to purge those bot accounts. They can even do a money-back request to the payment processor afterwards as the service is breaking laws.

Breaking TOS, not necessarily breaking laws.

While talking with my parents I realized how little of their generation is available to us. What did they eat in there travels? Where? What music did they listen to? What kind of cloths did they wear? I have a few fragments of photos, anecdots and not much more. I was thinking about how highly detailed memories of our generation would be available to future generations. Facebook, Twitter etc are in a way digital monuments of our entire generation. Looking at these misguided arbitrary corporate decisions of culling data isn't comforting. Archive.org is trying to do its best but there needs be better efforts in preserving memories of who passed away and we need to be more sensitive about that even though they are going to produce ad click revenues.

> arbitrary corporate decisions

There was a HN thread about an open source Twitter implemented in Haskell with the full stack in one page (it's called something specific) and that made me think that maybe decentrialisation is really just having decentrialised habits.

If you can figure out storage, the rest is all just done on the client side. It would kind of be like a exodus back to 1997, when everyone just slapped together some HTML and had a web page (I think; back then I was playing SC2 and didn't know what HTML was).

I thought of this example yesterday: Often people lament something like "the master craftsmen are all old, what would happen in the next generation". But I think the next generation is busy inventing the craft independently. Much like leathercraft or book binding on Reddit. I do think people can get lazy because of YouTwitFace. But I do think that a lazy person can become very efficient and prolific precisely because of revolt against their self-imposed boredom.

It does not require a lot of people (e.g.: archive.org) to make a rather large difference. GNU/Linux and the rest are examples of that 90% viewers/lurkers, 9% interaction, 1% creators. I don't know if I am misquoting that, but maybe people should be active in choosing what they preserve. That is, you are your own 100%; there is no 90-9-1 with regards to your legacy. You have to store it yourself.

Edit: SC1, obviously. I am sorry to concede that I didn't have a future copy of SC2 back in 1997.

For preserving at least your own data, check out one of my projects, Timeliner: https://github.com/mholt/timeliner

It downloads your digital life and stores it locally on a single, unified timeline, for archival and family history purposes.

That hit me recently - was sat in an airport the other side of the world having just sent my mum a WhatsApp saying we were through security and sharing some snaps from that leg of the journey.

Even within my lifetime, that'd have been a £10 phonecall and a postcard that might or might not arrive in a week or so. It took 20 seconds to have an immediate, searchable, chronological update on what we'd done and where we were. Incredible.

That being said, if Google Photos and WhatsApp go the way of the dodo, we're a bit screwed.

I have the opposite opinion on culling the data. This "preserve everything" attitude just smells like digital hoarding to me. Especially in the context of Twitter, which is mostly the noise with rare blips of the signal. And even in those cases the signal often is very specific to that moment in time and not that important later. The vision of a huge pile of digital trash with a few gold nuggets inside has zero appeal to me.

My wife has started making photobooks of significant family events, having observed that having an arbitrarily large pile of digital photos that never get curated or culled actually means we're less likely to revisit our past.

I find it interesting that what Twitter is doing here would have been absolute standard procedure in any small forum 10 years ago - yet everyone seems to panic when Twitter does it.

I think the reaction shows how much the internet has changed - and, to be honest - how much the tech crowd has been seduced by the idea that all data should be around forever.

I think this is a great descision by Twitter in terms of user privacy.

The difference being that content posted to forum from a deleted account is still accessible unless explicitly removed - just marked as from deleted user, while Twitter feed or post is not.

I'm not sure that's better or not.

Wow, this is happening on December 11th? That seems rather abrupt.

Why not freeze and hide the accounts instead of deleting? I.e. don't push any tweets to them, or anything else computationally intensive. Then when they log in again, have them click a button to "catch up." Probably not as profitable as annoying a lot of surprised users who try and log in after Dec 12...

>>Wow, this is happening on December 11th? That seems rather abrupt.

Just in time for Christmas, image wanting to look back over your deceased father tweets on the holiday only to discover Twitter black-holed them for you...

it never cease to amaze how incompetent management is at Large corporations, it seems the bigger a company gets more incompetent is management becomes

It’s good that they’re giving notice so you can download them.

It is not Twitter’s responsibility to archive for all eternity.

You're missing the point.

I always love comments like this.

At no point did I say they have a responsibility to archive them for all eternity.

Further giving people 2 weeks of notice over a heavy holiday period were most people are consumed with other activities is not "giving notice"

Further still it is just bad optics, PR is not about legal responsibility, not about technically being right, it is about how it looks to normal everyday people. Moves this like look bad to normal everyday people.

It is moronic to pull a move like this over the year end holidays. They have chosen this because they think it will fly under the radar, that they can do it while everyone is busy with other things. I have a feeling this is going to back fire on them spectacularly

Which is great because personally I think twitter is a net negative for society and its collapse will be good for humanity

I imagine the goal is to free up handles for new users.

The article explicitly states the goal, which is not to free up handles:

"The site said it [the upcoming deletions] was because users who do not log-in were unable to agree to its updated privacy policies."

Although, to your credit, the handles will indeed become available:

"...previously unavailable usernames will start coming up for grabs after the 11 December cut-off - though Twitter said it would be a gradual process, beginning with users outside of the US."

The evidence suggests that the suspect committed the crime. Oh nevermind, the suspect explicitly stated he did not commit the crime.

I don't even think freeing up usernames is necessarily a bad thing, but trusting the stated motivation of big companies for their actions seems awfully naive.

The fact that twitter has a size limit of 280 bytes (used to be 140 bytes), they are probably trying to free up some inodes to save some disk space, as greedy as it sounds.

They neither store each tweet in a single file, nor is is 140 bytes but 140 characters (think ASCII vs UTF-8)

Also pictures/videos attached to the tweet

it was sarcasm, and it was never implied that each tweet was stored as a single inode.

Why would they store each tweet as one file on an inode limited file system?

it was sarcasm, but anyone who takes twitter seriously should have their head checked

Oh, ok.

Storing each tweet as one file might even make sense. But you need to pick a filesystem that supports that use case at scale, if you are twitter.

https://en.wikipedia.org/wiki/Maildir is a good example of how this can make sense---at least at a small local scale.

They've already started deleting old tweets for some reason. I only tweeted 4 times and the 2 oldest tweets are gone :(

I had a tweet from 2006, right after launch. It just said "this is dumb".

It's gone now.

Wait, is this really happening? Can you confirm or post evidence? This is a big deal if so.

I can't provide proof, but I _know_ that I tweeted @adom_dev around the time of it's kickstarter and one of the Baldurs Gate EE devs about gemrb around the time they announced that project.

Both tweets are gone.

I don't think those tweets are gone.

The "Tweets & replies" section of your profile (https://twitter.com/eggsome/with_replies ) has two tweets that seem to match your description:



I can't prove that it was deleted. My first tweet appears to be from 2012 but I joined in April 2007 and definitely tweeted once then. (I was wrong about it being 2006)

I'm pretty sure that I changed my handle at some point. I'll try to remember what it was and search the internet archive.

How does one prove a negative? A deleted tweet from 2006 where nobody has taken a screen shot. It's not an active account like POTUS with the entire world watching specifically for catching deletes. I'm seriously asking

Internet Archive. Account metrics. Cached search results. There are probably a number of potential ways.

That doesn't seem like a big deal at all. They don't do blanket removals of tweets.

jack probably messed up while testing

That may not be intentional. Or permanent.

They may have had an accidental data loss. But more likely it is data that is rarely read so not stored in a hot cache near your edge.

Are they really deleting the tweets and data, or are they just hiding them from plain view?

It seems that deleting any old data is a negative for the company. Data is the new oil, right.

Are they gone or just held in cold storage?

If the usernames free up, I wonder if this going to lead to something synonymous to domain squatting?

A solution could be to make the usernames available after a random period of time (for up to a year), so popular usernames aren't released all at once to be grabbed by a squatter. With a staggered release this at least allows the chance for someone legitimate to stumble upon the username before a squatter arrives.

If squatting Twitter names becomes even marginally profitable you can be sure that squatters will be setting up bots that will monitor the availability of deactivated usernames that were popular in the past and that as soon as these are released they will be taken. A delay of the kind you suggested will not stop the bot writers.

Yes better to put the account names up for auction and bring in some revenue.

Donate the revenue to some charity, if the auction would be a PR problem otherwise.

Eg https://fiscal.treasury.gov/public/gifts-to-government.html

Terrible idea.

On the other hand, my username (arp242) is already "squatted" by an account that never tweeted or had any other apparently activity. I don't use Twitter much, but people have erroneously "mentioned" this account on a few occasions.

They already have this problem now. This may actually be one of the few positive things that come with this change

I doubt they'd free up the usernames. Username reuse has problems with reputation hijacking, either for paid likes, spam, or making it look like the user said something they didn't. It also creates a bad experience with historic @users in tweets, and no one wants that bad PR tweets pretending to be from someone deceased.

What you said is true, but they will free up old usernames

FTA- "previously unavailable usernames will start coming up for grabs after the 11 December cut-off"

Twitter handles are worth a lot of money and hackers have spent much effort to steal them. See



Amazing if Twitter is just going to release valuable namespaces worth millions all on the same day.

Color me incorrect. I wish them luck with that.

They could rewrite @mention urls on recycled accounts to a /recycled page that explains this.

I imagine they want to free up accounts that never did anything. You should be able to safely make a handle available after deleting an account if it has never tweeted, has not been tweeted at, and hasn’t logged in in a long time.

> In future, the firm said it would also look at accounts where people have logged in but don't "do anything" on the platform.

Yes, Twitter is going to purge a lot of accounts, the same accounts Twitter boasted about having pre IPO in order to fake growth.

That sucks!

I have an account. I log in but I never post. I only use it to follow some interesting accounts —that is all. I don’t contact customer service i don’t feign outrage I don’t virtue signal I don’t post memes... it’s by all means passive.

WTF, I just want to follow some accounts which disseminate useful information.

That represents probably the overwhelming majority of Twitter usage. Your account will be fine.

I think that counts as doing something

The best type of twitter account is the one that brings eyeballs to the platform. Yours is the second best. You're fine.

I do the same with Reddit and HN - I have an account because I'm on the platform, but I rarely comment. This probably won't be a great move for Twitter, especially given how little time they've provided.

> Twitter will begin deleting accounts that have been inactive for more than six months, unless they log in before an 11 December deadline.

I’m responding to my parent poster who quotes: “In future, the firm said it would also look at accounts where people have logged in but don't "do anything" on the platform.”

I guess it depends on what “don’t do anything” translates to in Twitterspeak.

Maybe "on the platform" should be emphasised. Like if you just use your twitter account to log into another site or something.

The old Yahoo Mail strategy. See how well that one worked out.

I don't get this reference: what happened at Yahoo mail and how did it work out?

They removed inactive accounts. And made the handles available to register again by someone else.

Do you come into contact with Twitter advertising? Then you’re probably fine.

They probably have ways of getting the metric usage for lurkers

> The cull will not affect Twitter's reported user numbers, as the firm bases its usage level only on users who log-in at least once a day. According to its latest earnings report, from September, Twitter has 145m "monetisable" daily active users (users who come into contact with Twitter's advertising on a daily basis).

There are so many now deceased person's account which contains very interesting, and are definitely considered as an important historical material. These historical materials will be lost on this move.

Yes, we can't expect the for-profit companies to preserve the history. But these are so hard to preserve.

CCPA is probably at least part of the reason behind this. It takes effect on Jan 1. The article mentions users not being about to accept their privacy policy.

There is a nonzero chance that various political figures will have their handles 'accidentally' deactivated and inserted into the available pool. An enterprising body should set some bots up right now just in case there's lulz to be had.

Should also have a tweet lined up, in case they catch their mistake fast. Something like 'we're nuking wales tmr'.

Yeah, bots will stay - they post like crazy and make me not use Twitter. Another idiotic decision from Twitter!

Are tweets being deleted or just accounts?

What happens to historically important tweets (e.g. from a government office holder) who happens to have died, 5, 10, or 50 years from now?

Reading this news, I remembered I had an old secondary account (for a musical project) that I hadn't used in years but wanted to keep.

I logged in using this old account, and Twitter asked me to prove I was not a bot. It required I would fill a recaptcha but also that I would give them a phone number for SMS verification.

So in order to keep this old account, I have to disclose my phone number to Twitter. I'm furious.

I also refused and replied to the email saying so and they enabled the account. This was a few months ago.

I whish they put more effort on blocking the thousands of propaganda bots that are created apparently with lots of ease https://www.theverge.com/2019/11/18/20970888/bot-campaign-tw...

An interesting move... at an interesting time!!! I posted few days ago here, on HN "Ask HN: Do you know a regular Twitter user?" (https://news.ycombinator.com/item?id=21584586) ;-)

I see it as an interesting move because it will allow Twitter to have a more "honest" number of real used account.

Moreover, for Twitter, it's a way to build stronger engagement: a lot of people will connect to unused account to avoid the risk of loosing it (even if they don't really use it, just to avoid loosing it).

So the active accounts number will not be that honest! In a way it's a kind of FUD...

And it reminds everybody that Twitter is about "live and news", not building some kind reference.

This overall seems extremely poorly considered. Hopefully, archiving services and efforts are going into overdrive to fix Twitter's errors here.

Allowing hijacking of old, valuable historical Tweets makes no sense — even if it would be nice on some level to reclaim the handles. It may not be worth it.

I'm wondering about those users who are under govt clamped internet shutdown since months and were not able to login since months e.g. Kashmir (Indian Administered) where internet shutdown is continuing since Aug 5 2019?

Ooh! A story about me! I'm a very inactive Twitter 'user' (Can you really be an inactive user? I'm not using it at all.) Ages ago I made a Twitter account[0] mostly to test some social media links for a website I was working on.

I don't even know what half of my tweets are about. Spammy contests apparently? But I do stand by my Scala comment. (Actually, thanks to EcmaScript 6 it makes more sense to me now.)

[0] https://twitter.com/MartijnCVos

Super dumb move. This will cause all kinds of impersonation and fraud troubles.

The removed handles should be quarantined for say 10 years after the removal I think. That makes it at least no harder to create a fake account that's similar to the original than it already is.

x@y should be unique, forever. It's not as though we're running out of bits.

Why should it be?

We might not be running out bits, but keeping those bits maintained takes effort; it doesn't happen by magic.

and I can't think of any form of identifier that guarantees uniqueness; even the stone markings on graveyards eventually weather away.

Because bits aren't stone markings on graveyards. If you want they can last as long as the service does, which is a fine cut-off. Otherwise you get the same trouble with every other kind of method of authentication: after your ID is recycled whoever picks it up again even after it has been carefully scrubbed will inherit all the inbound links and so on. That doesn't really happen with gravestones.

I'm not trying to be argumentative, but I'm looking for something more than "it's possible to do this."

Or rather, I don't think there's a need to go to such an extent to preserve old usernames and histories. I think it'd be sufficient to let archivists at whatever data they wanted, and return the names to the churn.

They're not names, they're identifiers. There is absolutely no shortage of those and Twitter threads in their original form have all kinds of uses. Court cases, keeping the web in one piece, reference material, claims of prior art and so on. The archives can not be searched and that makes them much harder to use for that purpose.

Our company is pretty handy in digging up stuff but you'd be surprised how much of the web is rotting away while you're staring at it and every time that happens there is a fair chance we lose something, forever. It need not be that way.

Imagine we'd have a day-to-day account of the lives of the people from 500 years ago, and that you could home in on whatever individual from that time stuck on the web. I personally think that unless the author removes it these content farms owe it to the creators that gave them the content in the first place to at least support and store that content until copyright on the content runs out.

Thank you for elaborating. I'm actually more sympathetic to preserving identifiers now!

I do think Twitter could handle this better. One problem with your distinction between names and identifiers is that, for Twitter, they're largely the same. And, Twitter lets any user change their name AND identifier at will. The history on Twitter is already badly muddled; this doesn't change much, except for a handful of high-profile accounts that people want to keep as memorials.

Twitter, I think, is not the right venue for that. They should make these accounts available for archival, or spin them off into a separate namespace, but I don't believe it makes sense for Twitter to keep inactive accounts preserved as they are indefinitely, just because the whole twitter namespace is already in such a flux (and it's not a good venue for memorializing the deceased).

Still, thanks for clarifying and changing my mind somewhat.

I'd be ok with it if they preserved the history, timestamped the change of ownership and made sure that from the page of the current registrant you could backtrack to the history of the account over time. That would be an acceptable middle-ground, maybe even show (#2) or so behind the name linking to the account history to indicate the nth re-incarnation of that particular account. That way all the old content would remain linked and you'd be able to find out if someone has re-purposed an account right up front.

I hope I'll get my real name when this happens. The account that has it now hasn't tweeted in 7 years. They also appear to have been a fake female profile.

I don't even get why. My name is pretty clearly male.

In related news, all the people who quit Twitter years ago but never actually deleted their accounts will be getting together on December 10 to collectively shed one single tear.

They're backpedaling until they figure out an archive/memorialization strategy for people who passed away: https://www.theverge.com/2019/11/27/20986084/twitter-inactiv...

> The cull will include users who stopped posting to the site because they died...

Indeed, your tweets die with you. I wonder about really popular personalities?

They suspended Robin Williams's account after he died, so they probably wouldn't be against it.

I think that's really sad.

Maybe someone will make a tweet in peace site.

How about removing the nazis?

They can't remove the nazis, because apparently if they did, it would remove too many Republican accounts[0].

Also because the nazis are, unfortunately, quite active.



You created an account to say you look like a nazi?

In April of 2011 I made a twitter account to post wherever I had updates to post on a certain subject. I haven't had updates to post in a few years on that subject (I last posted in 2017) but I don't think I should have my account culled.

EDIT: And it doesn't even make any sense. It's a publishing platform. The author already agreed to anything necessary.

So log in to it now.. literally just log in

This is both good and bad, gets rid of squatters but also could delete accounts of those that have passed away and have generated great content. Why not just change it to...deleting inactive accounts that have under X amount of engagement and overall tweets. That would save most accounts of the deceased while getting rid of the other problem.

Wish I could just pay an annual fee for a Twitter name like I do for domains so I don't lose them.

I wish they would clean up the honeypots that keep adding people. I report and block about a dozen fake accounts that add me with an account created within 30 days.

I know 1 guy who actually fell for a honeypot, even sent her money. Its gotta be profitable for scammers with as bad as its getting.

Are we calling those "honeypots" these days?

I wonder if this will help combat the bots? They often sit dormant for years and then start posting.

They would do a world of good if they just culled one particular, very active account instead.

Tangentially, has anyone obtained a username via trademark infringement? How did that go?

Yes. It works sometimes, and not others.

I got notified by this because I have a few old accounts, like, I set up a ton of accounts just because I needed to test something.

And actually, if anyone wants to own `@INeedToTest` (maybe a QA oriented account?) let me know :)

I hope Facebook does the same thing at least for deleted users. I’ve deleted my FB account for ages and I still receive emails from FB asking me to go back while I’m not subscribed to any email notification from FB.

Are there any Twitter folks here... how can i can access to my 12 year old account? I lost the email its associated with and really would like to regain access!

Has anyone been able to recover their account that was in this state?

If you remember what the email was, check for if it's ever been in a data breach, then run a password cracker on the hash.

Wow this is quite smart. Never thought of that

When I was younger, it was a fun hobby to help people with.

Cool idea, thanks!

I actually just gained access to that email and unfortunately since I only used it once years ago it was wiped.

Darn, I hope they don't wipe my account... there is no support .. premium support even as Id pay to have access and use it again.

If it was wiped but you now have access to it, just send the recovery email anyway. Companies can't check if an address was deleted and then remade.

If the goal is to make follower counts more accurate, why not just clear the old accounts' follows? That would dodge the many issues with this that people are pointing out.

As long as they don't require me to login before I can read some tweet.

I think that's the only reason I have an account.

> That said, previously unavailable usernames will start coming up for grabs after the 11 December cut-off - though Twitter said it would be a gradual process, beginning with users outside of the US.

> beginning with users outside of the US.

> outside of the US.

Why is Twitter discriminating against US users? Also, how do I ensure I can get that username that's been squatted and has no activity at all?

This seems like a great opportunity to grab that twitter handle you always wanted.

Rather of the opinion that Twitter should have an annual Big Reset.

They could just tombstone the user and not treat them like a normal user. I imagine they are trying to clean up indexes and evidence that their network is just tie with abuse.

Time for a regular reminder that Twitter's namespace is the sole property of Twitter, Inc.

We should really get about replacing it. How's Mastodon holding up? I could never get over the ridiculous name or "toots", but if it scales we should really be using it instead.

Another thing that shocks me is that ads actually pay for Twitter. How did they build such a massive company (expensive employees and outsized market cap) on ad revenue? It seems like the wrong demographics for ads to actually work and persuade people into buying. I wouldn't be surprised if Twitter becomes ground zero for the advertising bubble to burst.

The advertising bubble will burst.

Mastodon is holding up just fine and growing rapidly, the atmosphere is much nicer and every few weeks, there's a massive influx of new users whenever Twitter is upsetting its user base.

The Fediverse (which Mastodon is just one particular implementation of) is shockingly healthy right now. I surmise that it will look quite a bit different once you have, oh, a bbc.com implementing its underlying protocol and pumping out content onto it. At that point who knows what'll happen.

>I surmise that it will look quite a bit different once you have, oh, a bbc.com implementing its underlying protocol and pumping out content onto it.

I hope this never happens. It's rare that something doesn't become worse when it becomes mainstream online.

It's pubsub so if you don't like the BBC's content then you just don't follow them.

Mastodon itself as a service and protocol, is pretty wonderful.

But in terms of namespace, domain-name, and instance issues, you've got pretty much all your traditional Internet concerns, with a few new twists added.

My first primary instance has gone the way of the dodo, and its domain is now parked somewhere entirely different (https://mammouth.cafe). Numerous other instances have also come and gone.

My current instance, run initially by an anonymous administrator, was transferred to a Japanese concern, who don't do much to keep anyone advised of service outages or their causes/resolution. This is an instance with 56k registered users, large by Mastodon standards.

Then there are the issues with blocking and federation, which don't immediately concern namespace, but could as blocked instances die, and potentially others occupy their namespace, or they change their identity to evade blocks.

There's been a lot of agitation over the dot-org PIR self-dealing and corruption, as that's a fairly commonly-used TLD, either directly by instances or for related resources. The whole DNS / registry system is a bit of a mess.

The fact that between the Internet and DNS there's no real allowance made for either space or time is becoming an increasing stress point. What solutions might exist isn't clear, but assumptions of the 1970s and 1980s are not being sustained 30-40 years later.

As I've noted a few times recently, a dictionary bibliographic entry typically consists of little more than name, dates, nationality, and profession. Webster's 11 Collegiate includes some 6,000 or so such entries. There are roughly 1 million times more people than that now living, and another 100 or 200x more who have ever lived.

In digital systems, there is no distinction between use and mention -- the name is the thing and the thing is the name. This is convenient in narrow scopes (time, place, scale), and ... exceedingly inconvenient over broad scopes (time, place, scale). UUIDs address the collision problem but not the convenience. Convenient shortcuts address convenience, but not uniqueness requirements.

Since meatspace identity is who you are, and other references (names, nicknames, registration numbers, etc.) are not identities but identifiers, the problems aren't as manifest. Your actual identity cannot be merged with another (though numerous related aspects and relations can be). Online, if A and B are suddenly both called "A", then they are both "A".

I think we'll be sorting this out for a while yet.

Mastodon instances randomly disappear forever too which doesn't solve the issue with Twitter deleting inactive accounts, unfortunately.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact