The non-disclosure order by the courts for this warrant was issued to Twitter on January 17th for six months; the day it expired, Reddit suddenly decided to purge all chat history prior to 2023:
https://news.ycombinator.com/item?id=36782915
They haven't really purged it, they just haven't copied older data to the new chat system (dumb decision). They still have that data, and you can get it by submitting a data export request: https://www.reddit.com/r/help/comments/14ee0wz/comment/joxzv...
At this point, Elon's purchase and seemingly-intentional brand-suicide looks more and more like a political favor.
I mean, yes, if we shave this with Hanlon's Razor, the outcome is the same, but the rightward swing of Elon's public statements certainly makes this look more like quid-pro-quo than an oopsie.
I can't get over how Twitter's legal team doesn't show up on time for a court appointment. As a founder, I'm certain they're still billing for the time.
Not OP but I believe the main assertion is that Elon Musk's public rhetoric has become increasingly political over the past 5 years and that his dialogue conforms with modern conservative messaging. To break it down:
* Hes posting a lot more
* The content of his posts are proportionately more "culture-war" topics versus company/nerd/geek stuff that he used to post
* Some of his posts align with populist conservative influencers
I don't have data to back this up - just my anecdotal experience.
Well the left side of the political isle has also became increasingly critical of him since 2016 when he briefly worked at the White House. He went from being a darling fighting climate change to an evil villain almost overnight.
While it's true that he's currying favor with a political group that favors gas-guzzling pickup trucks over electric vehicles and is diametrically opposed to his mission of saving humanity from ecological disaster, owning the libs (his customer base) is its own $40b reward. /s
I think modern Elon is too jaded about reality/fat with cash to still believe in saving humanity. He certainly doesn’t talk about it in the frequency and light he used to.
Why would you think he ever believed in that rather than it being a personal and corporate brand marketing tactic the way it has been for so many other startups and their CEOs, to the point of being frequently mocked?
Really I only believe that because he’s consistently shown he’s sufficiently disconnected from reality that legitimately believing he can single-handedly save humanity seems par for the course.
It's a little bit outlandish, but the most plausible theory I've heard is the Saudis were happy to spend the cash[1] to try to hinder future Arab Spring style events, which were organized largely on Twitter. US banks have a similar motive, it's not hard to draw the line that they would see disrupting low-level leftist activist communication networks on Twitter as valuable. Under that theory, Musk isn't so much doing a favor as being a useful idiot.
The Saudis were already the top shareholder in Twitter before Elon’s purchase. They were adamantly against it until he let them roll over their stake into X
If it were any other man I'd be considering your conspiracy theory but Musk has become so completely unhinged that imho Hanlon's Razor wins. The man has a terminal case of internet poisoning.
I believe that there will one day be a name for this syndrome, wherein a person who is the victim of prolonged internet trolling begins catering to the trolls with increasingly outlandish behavior, with their mind / beliefs warping into a shape that pleases the trolls.
Mechanically, I think a person who detects that they're a social pariah might lean into the social role of a jester: the tribe will keep him around for laughs, instead of fully ejecting him into the jungle to fend for himself.
I've been wondering if there is a term for 'the choir glamouring you' sort of thing?
I've been noticing odd behaviors in some public figures that make me believe instead of them 'preaching to the choir' - their 'choir' is convincing them a weird direction is a good idea.
I think this is exasperated by the bubbles of social media - I would love to be able to see what social media posts certain people have seen over the past 5 years or so - like individually, say if I could study every post that rob de niro actually saw in the past so many years.
I feel we need new terminology to describe the kind of warping of the mind and outward behavior changes these things can bring -
or is there some old descriptors I can't think of?
These are new phenomena, and they might be qualitatively different than anything that previously existed.
Historically, you could send a letter to a president. If the letter were unhinged, staff would throw it out. Thus, as a sociopath or a crazy, you'd have little or no access to a president.
Now, a president is bombarded by thousands of crazies per day, and he will actually read some portion of it. We're no longer ineffectually yelling at the TV: the man on the TV is listening. Anyone can easily gain a small part of a leader's attention, and collectively a group of crazies even without coordination can likely influence him.
Additionally, as a crazy prior to social media, you'd have to settle at best for periodicals from your extremist group of choice. Most of your life would be consuming mainstream information and interacting with normal people. This would balance you out a bit.
Now, you can consume extremist/conspiracy content all day long without ever coming close to running out, and your entire social group can be fellow crazies from around the world. You can create for yourself as pure an echo chamber as a cult in a commune.
It's a recipe for radicalization and induces the same strange-belief-conforming that can be seen in any and all religions.
This can play out at any scale. For instance, the majority of Republicans believe the election was stolen, liberal cities are overwhelmed by constant looting, and immigrants hoards are surging across the border. They truly live in an alternate universe, and it's driven by the information+social bubbles of social media.
I dub the phenomenon "bubbling". Whether the bubble is an internet troll funhouse or a fear-and-outrage feed or the welcoming arms of a coven of internet crazies, the bubble will shape most humans who enter one into the bubble's own image. You can and will be bubbled.
I'll go with the term bubbling for now, and suggest that we should have transparency into the bubbling that our elected officials are stuffed with.
I'll add to your thoughts with the interesting battles even to the courts where some officials could not block a user, and be forced to listen to crazies they tried to filter out.
I would actually pay money to see the feed that deniro (and some others) has been bubbled with the past so many years.
There was a time when I saw distant fam getting a deluge of questionable memes and such that cobbled partial news with partial non truths and feigned outrage mobs grew into real outrage mobs that believed even if shown their sources were suspect, they cared more about the bubble than sorting whats right and wrong.
I assume this has happened to many people and very few have access to the bubble layers for each person, we may never know the influences others have been fed that has been guiding them.
This is definitely a thing I have noticed - part of it seems to be, if you refuse to react to trolls, the trolling can actually get much worse - naturally encouraging a reaction that the trolls want.
Any other time and HN would be cheering that user data was attempted to be kept private and not disclosed. But one whiff of it being on the wrong side of a political divide and it becomes a witch hunt.
For warrantless yes like when government gets logs from network providers and mobile phone operators without a warrant. I don’t think most on here disapprove of gathering evidence on a specified target via a judge approved warrant.
I don't think that's true and I'm surprised the GP is flagged.
I think that many of us believe that all personal messages should be E2EE. When user data is not encrypted, companies should fight tooth and nail before handing it over to the government.
(I'm not American and have a highly negative view of Trump, but that should be irrelevant.)
I mean, if it’s a favour, then it isn’t being particularly well-reciprocated; ol’ mini-hands is still on Truth Social, not Twitter. I think Hanlon’s Razor probably still applies here.
If you're MBS and you determine Twitter is the most likely source a the black swan event leading to your regime's overthrow maybe it makes sense to pay the money.
Here's the government attorney explaining how Twitter should do it:
> To the extent that Twitter does not, it should be
a simple process.
You have cookies associated with an
account; you have the email for the subscriber information;
perhaps you have the phone number; you definitely have the
IP addresses. Control F that through your system to see what other accounts have come from those IP addresses, are linked to that email address, are linked to the phone
number, are linked to the same cookies.
I don't profess to be a technological wizard, but
it does not seem to be a complex issue.
Reminds me of freelancing nightmares where the client would being with: "This should only take a few hours but I need a website with a simple store front and shopping cart...". The best response I came up with to these clients was: if you know exactly how long it will take then you must know exactly the steps involved and how long each takes so you can do the work yourself and you don't need me :P
So following your analogy (which I'll state upfront is a poor one), where do you think the government attorney is wrong? Because from where I sit the government attorney, if not completely correct, is at least on the right track. Other companies can piece this together (how else do they know that you are logged in somewhere else?), and I would be shocked if there isn't some means on Twitter's backend to do the same.
And from reading the transcript, it would seem that Twitter's objection is about analyzing the data, as opposed to simply providing it.
I just think it will realistically take Twitter more than two weeks to come up with this data. Seems like a long time but my experience at companies of 1,000+ is that things move very slow.
Not when it's for a court case or a legal requirement, in which case they put one or more people on it full time until it's done and make sure whoever else needs to get them the data is in the loop and knows it's a priority.
Bureaucracy is what slows down larger companies, but that's because incentives aren't aligned. When incentives are aligned and priority is understood, often that bureaucracy fades away to a large extent, even if for short periods.
> Nonetheless, the district court gave Twitter an
opportunity to purge its contempt by producing the account
information. When the court asked Twitter's counsel whether
the company could produce the required materials by 5:00 p.m. that evening, counsel answered: "I believe we are prepared to do that. Yes, Your Honor."
If I'm understanding the transcript correctly, the major issue wasn't that Twitter couldn't collect the information or needed to do extra work to collect the information; Twitter just didn't want to provide the information, seemingly so that they could Trump and have him lodge a (bogus) executive privilege claim.
Not really. It's not like this is the first time they are issued a subpoena. Any decent big tech has a decent system for attending to legal requests and, probably, a technical team on call with the power to run custom queries if necessary.
> The best response I came up with to these clients was: if you know exactly how long it will take then you must know exactly the steps involved and how long each takes so you can do the work yourself and you don't need me
I don’t think Twitter really would want the government to take physical custody of Twitter’s assets and do the search themselves, but, sure, they could have tried that response.
Google seems to be able to find associated accounts by seeing if one device is signed in multiple times. Same with Meta.
The court specifically asked for other accounts used by trump on the same device. This SHOULD be as simple as control-F. Twitter almost certainly stores some sort of device identifier that can accomplish this.
> Twitter almost certainly stores some sort of device identifier that can accomplish this.
They definitely do, and use this as part of when they ban a user permanently - all other accounts (i.e. "backup" accounts) that have ever used the device in question get banned as well. Happened way too often in my circles.
> By better understanding how different browsers and devices are related, we can use information from one browser or device to help personalize the Twitter experience on another. For example, if you commonly use Twitter for Android around the same time and from the same network where you browse sports websites with embedded Tweets on a computer, we may infer that your Android device and laptop are related and later suggest sports-related Tweets and serve sports-related advertising on your Android device.
If you make a data-access request in GDPR you might actually get a massive CSV file with this data in it. Obviously some automated system put it together, not control F, but the legal team totally could be using CSV
What's funny is while the judge might look silly suggesting such a basic technique, Twitter's lawyers were clearly "playing dumb" saying that a search for "realDonaldTrump" produced "millions of emails" and therefore was unreasonable.
Well I think they were saying that (1) internally at Twitter there were millions of references to the realDonaldTrump account in slack/email (plausible) and (2) there were a ton of external communications about the account which were probably not material to this request.
> Among the data the search warrant commanded Twitter to produce: Accounts associated with @realdonaldtrump that the former president might have used in the same device.
They know that doesn't mean "anyone who @mentioned him publicly".
A public @mention of Trump by a random person is not a "communication between Twitter [and that person] regarding the [@realDonaldTrump] account" by any sane reading.
That item means they have to hand over things like support tickets or emails sent to the @realDonaldTrump account or Trump's social media team.
A brief summary: lawyers for Twitter were busy obfuscating and dissembling to the judge, pretending that Twitter was completely unable to formulate a SQL query to identify multiple accounts used by the same device and otherwise playing dumb, trying to throw enough technobabble to confuse a judge. The lawyers were also assembling a bunch of nonsense 1st Amendment arguments and hurling those at the wall too, hoping something would stick.
The judge and prosecutors were not fooled. Eventually the judge fined Twitter for all their nonsense.
So, contrary to some of the comments posted here already about dumb judges or whatever, this case shows a judge seeing through a bunch of techno-nonsense that lawyers tried to use to sow confusion.
HOLTZBLATT: I wanted to confirm that information because I don't want to make an incorrect representation to Your Honor.
THE COURT: Of course not. We have already been through that.
It's shocking to see the data that has been collected on a platform as simple as Twitter.
During collage I was intern at a legal firm, and one thing I've learned is the more info someone can get on you the stronger the case. Regardless of the content weather good or bad.
The DOJ asked for access to Trump's twitter account, including DMs, deleted items, and drafts. Twitter stalled on producing this information so this hearing was the DOJ asking the judge to start hitting Twitter with fines until they could produce.
There's also some confusion at the start because DOJ asked for other possible accounts that Trump might have used as a sockpuppet/backchannel means of communication and Twitter seems to either not be able to discover that, or can't find it.
> Twitter seems to either not be able to discover that, or can't find it
They clearly CAN, but don't want to. Their (X's lawyers) objections in the transcript are ridiculous (they claimed they don't have an "affiliated account" category - like really?!?) and when pressed for timeframes suddenly they can produce "impossible" data in less than 24h.
Was Twitter really as sophisticated as Facebook in this regard?
If @realdonaldtrump and @johnbarron are both tweeting from the same IP in the same time frame, could Twitter make the connection that both are the same user? Was there enough backtracking stored in their systems to know that?
They may not have a literal field for the account, but they are asking twitter to run a simple join across database fields for matching IP addresses in the time range. The jusge is clearly fed up with twitters bullshit - the control F comment does actually make a lot of sense, it should be that simple. For example, SELECT * FROM dms WHERE ip=donaldtrumpsaccountip
If the the IP is "the White House" there may be dozens of clients running out that single NAT IP. So you'd need a bit more correlation before you hit all of them with a warrant, I'd think. Some tracking cookie or some ID to show it's the same person behind the keyboard. As the transcript shows, the judge asked about payment (like a credit card # for correlation) but Twitter is free.
Not saying it's a viable technique but I get the idea that judges hate warrants that vague.
Sure, but as others have said, you have device IDs, you have when Trump was traveling and using cellular data, etc. So there is a huge amount of info that can be used to relate accounts.
Legally, it seems asking for a company to run essentially joins across their DB tables is accepted practice. Otherwise, how could warrant be obtained from google for all devices within a specific geolocation? Requires at least a few DB queries I'd imagine.
They should be able to trivially determine, after the fact, that the two accounts were using the same IP, even if not at the time. (Of course, that doesn’t _necessarily_ mean they’re the same person).
Twitter is playing games about turning over all the records requested by the government.
They are pretending that a database search for an IP address, for example, in two different tables is a big challenge and requires them to "make" the data because they don't have all the matching IP addresses pre-queried.
Twitter is likely doing this because musk is now in charge, and wants to help the criminal conspiracy by hiding and denying the legally requested evidence
Legendary where? His Twitter password was guessed[1].
Even worse, he used that password across several more services and didn't change them when his Twitter account was compromised. Legendarily bad, perhaps.
Many HN users have been on the other side of doing CCPA and GDPR data compliance and have more than passing familiarity with what is required and how that differs from what is desired.
The four word summary of "right to be forgotten" does not properly encompass the PII that can be requested to be deleted nor differentiate it from other data which can be retained.
We're not talking about passing regulations down on HN: we're merely suggesting capability that almost every other similar forum platform supports: being able to delete your posts & account.
You can delete posts within two hours and dang is usually good about deleting stuff if you email him directly from what others have said in the past. If you can’t delete it, you can edit it to be blank.
This is not a technological matter, it's product knowledge of a website the judge has no reason to use. This isn't a trial about Twitter, there's no reason for the judge to have any specialized knowledge here. She asked a totally normal question relevant to the discussion (did Twitter provide the requested credit card and bank account information) and then moved on.
It's a central and ongoing misunderstanding throughout the document that, beyond verified users, the Court does not realize Twitter account is not associated with a verified human identity. A judge should not be walking into this case completely ignorant of the subject matter that will be presented to them, but this one is.
It is unreasonable to expect a judge to be a subject matter expert on every case they preside over. That's why they ask questions. That is a reasonable question considering there are ways of paying for twitter and ways of getting verified on twitter.
“Oh, shit, we have a lawsuit about oil extraction rights up next week. Anyone know a judge who’s also an amateur geologist?”
That’s not how it works. The lawyers on both sides are responsible for explaining the salient points; in most legal systems the judge is not expected to be an expert on the non-legal aspects of the matter at hand.
Creating a Twitter account in recent years requires a valid email address and valid phone number, or at least it did until recently. Twitter definitely has a pretty good idea of which human being is behind the majority of its accounts.
