Twitter plans to comply with Musk’s demands for data
forbes.com
forbes.com
1. Musk is talking about percent of daily active users that are bots, but Twitter's SEC filings refer to the percent of monetizable users that are bots, i.e., users that actually get shown ads. There is no way for Elon to produce a comparable figure from the firehose data
2. Even if he could, the firehose doesn't contain the sort of PII you would need to determine whether a user be a bot.
3. But even if it did, none of this matters. He can get out of the deal only if he proves the error would cause a material adverse effect on the company, which is an extremely high bar that is basically never met and would certainly not be met due to an erroneous estimate in a SEC filing.
Anyway, I figure we should all suspect that the actual goal of this charade is not simply "to acquire Twitter" and motivation is not "to protect free speech" or "solve Twitter's bot problem" or whatever.
Eventually your answer to the "why" becomes simply: because you can.
I heard this idea from a colleague in the space industry. Going to Mars is one of Musks main goals, and as crazy as it sounds I could see him doing all this Twitter craziness just to move that ball forward.
https://www.faa.gov/space/stakeholder_engagement/spacex_star...
The results of the review were pushed back by 2 weeks from the original date and it now appears to be on track for the June 13 release. It doesn't mean it's going to be positive, but your version of it is pretty damn far-fetched. Like really super contrived.
This is just the most recent of a series of delays. The report was originally due by the end of 2021.
Even after the drops in tech stocks, he has enough money to fund the takeover if he really wants it. Did he not even consider that possibility? Twitter having bots is not surprising to anyone even a tiny bit informed. Their public records and filings have this data. Did he not even read those?
The whole thing is just bizarre and stupid, yes.
To me it seems like griefer [0] behavior. The goal is to just duck with Twitter and derive joy from others pain.
Musk is a gamer and I’ve played a lot of games. It’s pretty common to see people who are bored or max level or whatever who just mess with other players like a cat does with a mouse. They expel more of their own energy to cause discomfort to their target. An example may be spending 10 gold pieces to make the target lose 1 gold piece.
It “doesn’t make sense” rationally until you take into account the value of the confusion, pain, negative experiences for the target as a sort of currency.
So it seems perfectly common for me to see Musk spend a few billion in fines or whatever to make life worse for Twitter.
One theory I’ve seen to explain why Musk wants to buy Twitter is to shut it down. If that’s the goal then it’s actually much cheaper not to pay $40B to acquire it and shut it down, but to do this sort of thing that seeds uncertainty so when he pulls out the stock price plummets and Twitter dies from an acceleration from it’s likely path.
I think this Griefer theory explains a lot of the antisocial behavior we see out of executives and politicians who have beaten the game of life. We structure the game to make "Sociopath" the character class most likely to reach level 99 and win the game, and then we wonder why there are so many sociopaths using game-winning wealth to grief everyone else.
Whether or not that was the purpose is speculation but being held up as a champion of free speech (as if) I’m sure did great things for his ego too.
It will be interesting to see if conservative car types decide to give electric cars a go. Keeping a close eye on a famous YouTube mechanic.
Musk is not fully trusted by either side. He says a lot of things that traditional conservatives (especially those who don’t buy into climate change) oppose.
The left ideology is the dominant force in pop culture, and those views are in vogue. He’s speaking out against some of those things and that’s getting him a lot of attention.
Generally, this is far too complicated and involved to be any kind of effective scheme. He has professional managers and lawyers who could have easily come up with dozens of ways to sell tsla stock without this public legal morass.
The situation most likely is as it seems: He wanted to buy twitter for personal reasons and is in the process of doing so.
He could have justified the stock sale in plenty of other ways, including reallocating funds to his other companies, without forcing himself into a $44B deal.
https://www.latimes.com/business/story/2022-04-12/musk-is-of...
This is all a side show and negotiation tactics. Really not bizarre at all. The original offer is at least 10 billion or so too much now.
By providing all the data requested, twitter just squashed that out.
He had no out on the bot claim itself, but his legal team was comfortable that the covenant breach was a colorable claim and wrote a letter. Now they need to think of a new legally plausible claim.
It’s just the posts. Probably the most useless part of the data if your goal is bot detection.
But people seem to fall for it.
But there is also a “public opinion” part, and that’s about politics. I have the impression that Tw is trying to at least remain in control until the upcoming elections.
If Musk has a way of running analysis on the data to do the following: 1. Level of duplicates. 2. Source of duplicates (who posts it first) 3. How often do they do that? 4. What is the political slant of these tweets 5. Which part of the world are the people tweeting from
All of this alone will reveal very interesting information.. my guess is that it will prove that a lot of tweets are orchestrated.
The legal battle will not be about whether it is all the data. Musk isn’t owed that. It’s about whether twitter is sufficiently responding to requests.
Even without this the claim was very likely to fail. Hard to see it surviving this.
Pew analysis (based on surveying humans) says "97% of tweet volume comes from the top 25% most active tweeters"[1]
So Musk is arguing about a statistic that twitter claims, which is that spambots make up less than 5% of users.
Now, if those bots are part of the most active 25%, then they make up 1/5 of the 'most active' user population. So it seems plausible that they could well be responsible for 1/5 of tweets.
Which matches exactly what Musk is claiming - that 20% of tweets are from spam bots.
Which still makes no difference to Twitter's ability to sell Advertising targeting the 95% of users who are humans, 75/95 (=79%) of which are only responsible for 3% of tweets, but... presumably... 79% of ad revenue. Because they only serve you ads when you're looking at a feed. And that's what those users are using Twitter to do.
[1] https://www.fastcompany.com/90697826/the-top-25-most-active-...
https://www.bloomberg.com/opinion/articles/2022-06-09/elon-w...
https://www.bloomberg.com/opinion/articles/2022-06-06/elon-h...
https://www.bloomberg.com/opinion/articles/2022-05-17/elon-m...
https://www.bloomberg.com/opinion/articles/2022-05-13/elon-m...
Doesn’t make sense to buy the company to do things and… not just do that.
What data would that be though ?
1) Email addresses would be valid since they are used during the verification process.
2) Location data is only available on mobile which bots obviously won't be using.
3) IP address data would be filtered through compromised hosts.
4) Interaction data is probably the key one. But this is trivially forged.
The type of bots that are most common on Twitter are those from well funded interests e.g. governments, companies or technical ones e.g. hackers.
People who know what they are doing and would be building something largely imperceptible from ordinary users.
Twitter spam bots that offer to sell me bitcoin or start following me with a beautiful profile picture who just opened their twitter account and I don't know are pretty obviously bots to me. The former could be found using the firehose, but the latter wouldn't be detected until they tweeted something in the firehose.
If the firehose included DMs, new accounts (with all available signup and verification data), etc. then the firehose would be useful. Detecting bots using the firehose would only detect bots based on suspicious tweet content and nothing else.
Also, the firehose is a 7Mbps (compressed) stream. It takes substantial compute power to run just sentiment analysis and substantial storage to keep heuristics on that volume of data. Musk would need to spend some time and money and hire some people to just make sense of the full firehose.
Source: I worked for GNIP in 2012. GNIP used to re-sell the twitter firehose and was a twitter partner before they were acquired by twitter in 2014.
But! There's another way out for Musk.
Merger agreements include covenants, promises that the parties make to one another about what they will do to close the deal. If one of the parties can prove that the other didn't fulfill a covenant obligation and therefore broke its promise to work towards closing in good faith, then that's a valid justification to walk away.
The merger agreement includes covenants that force Twitter to provide any and all information "reasonably" requested by Musk, in a timely manner.
So the game becomes: how to trick Twitter into not respecting its obligations? One way is to ask for information that is impossible to provide: "Give me all the tweets since the beginning of time, each tweet on a single piece of paper!"
But Twitter might object, and a judge might find, that this is done in bad faith...
The other, more subtle way is to ask for information that Twitter should be very anxious and reluctant to provide, such as, for example, personal information regarding users. Twitter may run afoul of many regulations if it does, and has reasons to believe Musk won't honor any NDA he might sign (as he has a history of disrespecting his own word). Yet on the other hand, Musk can argue that he's entitled to ask for such information.
That's the nature of the current game. I don't think there's any way this isn't settled in court.
Note: for more on this, cf. Matt Levine's newsletter from June 6th.
Am I understanding this right? I would assume no matter how many ads you show a bot, they wouldn't end up buying anything. Or is the point that that those ad sales are not a reliable source of income since the people buying them would likely decide they weren't getting a good enough return?
If this becomes the case, Musk can use that as leverage to threaten to litigate if twitter doesn't accept a lower offer price.
That might be the plan, but what prevents Twitter from continuing to reasonably cooperate? You take it granted that there will be a point where twitter says no, but why?
Now that he can't simply walk away and pay $1B. The contract he signed has a few extremely limited circumstances where the deal may fall through, and in those few cases, he would pay the $1B. In any other case, he has to pay the full amount.
The $1B breakup was for a narrow set of cases, outside of those he's on the hook for the whole amount. Every move now is just part of the negotiation for how much he's going to pay get out of the deal.
There's also a moral hazard here. If Musk is allowed to walk away with zero consequence, he was able to hold Twitter hostage for however many months where they could have potentially had other suitors. It writes the playbook for other large companies to do the same.
The fact that Twitter seem to have tried to stonewall him will just "prove" to him that he is right and he can afford to be stubborn enough to drag this out.
That may be, but it's not part of the contract, and it's very unlikely to materially impact the future revenue of Twitter to a high enough degree that it would be a breach of the signed contract, so all he would be doing is assisting the dirty laundry of his new company in public.
I've seen this mentioned elsewhere also, but I really need help here. From my understanding, Twitter's entire business model, and existing revenue stream, is people paying them to show content/ads to other humans. If some unknown percent of users are not human, doesn't that have a pretty extreme, direct, influence on the possible revenue?
If someone wants to buy Twitter, I assume their first question would not be, what percent of users that you're currently showing ads to are bots. Rather, it would be what percent of users are real, so I can show ads to all of them. Which bring up the question, why isn't Twitter showing ads to some users?
Keep in mind that the onus is on Elon to prove to a court that the higher bot count has a large, material impact (on the order of 50% of Twitter's revenue). It's not something that is a matter of opinion, but rather based on his ability to argue his position to the court, for which he needs strong evidence.
https://www.jonesday.com/en/insights/2018/10/delaware-chance...
This is about Delaware finding that an MAE occurred, for the first time ever, like literally ever, in 2018.
What we have here is a case where Twitter's performance isn't even faltering (it's stock price is, but its 2022 revenues are actually up). There's not a chance in hell for the MAE argument.
Advertising buyers determine how much ad views are worth to them in their own way. This demand essentially sets the price since there is no marginal cost of providing advertising views.
For example, suppose that you show 1,000 ads and get 2 people buying something from your store. It doesn't really matter whether 1,000 people saw your ad, or 500 humans and 500 bots saw it. The important statistic is what revenue you got from it.
That's the biggest part of twitters business, but they do have additional revenue streams. However, the other revenue streams are also highly dependent on their status as "a place where people say things and people look at things"
I do think Twitter in its current form is essentially unusable due to the amount of scams. I see thousands on a weekly basis. Every feature, from tweets, replies, lists, etc. is a cesspool of scamming. I block 100s of users per week and yet I still get added to “Free ETH drop” lists daily.
The mDAU exclusion includes such accounts already. I think people (and Elon) use the term bot rather colloquially, so I don't think it's an important distinction anyway.
The more key metric for the integrity of the platform, it's ability to be a trustworthy indicator of real people's sentiments, vs firehoses of disinformation into a cesspool of lies and fake interactions, is not the percentage of bots (monetizable or not), but the percentage of bot vs real person posts.
2) Then they should provide data he can use to verify their figure. It is dumb they won’t provide it and are providing noise instead.
3) He’s not trying to get out of the deal, he’s calling them out on their figure. It’s dumb they aren’t giving him the data he can use to verify their figure and he wants to expose them prior to the deal to show how effective he will be when(if) he “cleans” it up.
I'm 100% on Elon's side here. He should be allowed to kick the tires. I've been in an organization which went through something similar twice, in both cases, based on material misrepresentations.
In the first case, two organizations who should have known better invested tens of millions of dollars, with one effectively defrauding the other. In another case, a $3.5B organization bought an $800M organization, and the combined entity was worth less than $800M within a year.
Serious, when spending tens or hundreds of millions, do your due diligence. People and organizations lie all the time.
I'd be against Elon if he backed out of the transaction based on a material change. Failure to provide data? He's being perfectly reasonable.
Waiving due diligence is a crazy dumb decision. The one lesson in my career is always do due diligence, whether you're making a $40B transaction, or negotiating stock options with a startup. People cheat. A lot.
His ego is at stake, he’s called many things but not often a fool.
I mean, yeah, sure on the surface it's economically rational and all, but really--you're already nominally in charge of at least 3 different large companies, yet this is how you choose to spend your time and energy? Nickel-and-diming Twitter for lulz?
edit: and before someone points out what a big difference a few dollars in the share price would make towards the acquisition price: if he really believes that Twitter has so much untapped potential (economic or otherwise), then a few dollars either way shouldn't matter much. But by going to such lengths to slightly reduce the cost and jeopardize the deal, he shows that he doesn't think it has much potential and/or he's just screwing around.
Both options are very unlikely, but this will drag on.
Unless and until such a thing happens, he owes them ~$44B.
Having worked with the Twitter API, I think he's rather underestimating how much volume this is gonna wind up being, the amount of processing it'll entail, and the difficulties of determining "this is a bot" from just the Tweet data.
Source: Former employee of one of the few companies who got the "push" firehose and not just the API.
"1 terabyte per day of data [...] - that is how much is produced by the firehose of 250 million tweets a day" [0]
i.e., ~4kb per tweet. So, 2ish Tb/day of text + metadata at current rates?
[0] https://techcrunch.com/2011/12/07/datasift-historical-tweets...
I don't see how he thinks tweet content alone will be enough to verify the 5% of monthly monetizable users
The other poor indicator was his tweet about sampling _100_ of his own replies to detect bots. That's a horribly small sample and horrible selection bias
The firehose is terabytes a day, add storage + indexes and it's pretty far from "fitting on an USB stick" in order to analyze.
Maybe from now on we should use statistics as the epitome of difficult intellectual activity instead of “rocket science”.
His modus operandi is taking quick decisions and backtracking if he is wrong.
Actually it's not only his modus operandi, SpaceX according to NASA evaluation is also working that way.
It's basically depth first search vs. breath first search.
Dr. Daniel Rasky, a NASA Senior scientist
https://youtu.be/MxIiiwD9C0E?t=1464 :
> I think there are some things and kind of going back to in some ways back to the Apollo days if you take a look at kind of the environment that NASA had in Apollo it was kind of what Elon has with his young engineers. you had you know people trying things and in what works and what doesn't work I think over the years we've gotten into a very risk-averse profile and we're I call it full matrix engineering that okay that all concerns that anyone can raise have to be chased down a hundred percent before you can make any decision. Okay, so yeah like a full matrix of all the things that may be pertinent to a certain decision and you have to fill that entire array before you can make a decision to go forward and there's times when that makes sense but it doesn't make sense for everything we do. SpaceX uses a sparse matrix okay 51% okay when do when we get to 51% okay enough we're gonna make a decision and then we're going to try it and move on.
Either way though, I expect Musk, the richest man alive, can afford to pay a handful of senior engineers a couple weeks pay to store and run some analysis on the firehose. It's a lot of data, but at "fuck-you" level money it's certainly manageable.
Some accounts are a mix of human and "bot" updates--and you can get a very different impression of their nature depending on your sample.
https://twitter.com/JeffLindner1, for example, posts automatic alerts from the Harris County flood warning system (such as https://twitter.com/JeffLindner1/status/1529386618938920960). If you look at his account when it isn't pouring outside, you'll see human updates and retweets. If you look during a three-day downpour, you may see dozens of automatic tweets per human post.
NOTE: For the ones operating legitimately I mean.
While I do understand your point about the difference, neither type should be measured for things like (e.g.) advertising revenue. Weatherbots don't buy goods or services.
If people aren't viewing those tweets or deriving value, they're consuming resources without adding much to the ecosystem. If people are viewing those tweets and deriving value from them, the bot is increasing ad revenue.
What a terrible cultural icon Musk has become.
We should all aspire to be fentanyl dropping, pregnant women torturing street trash.
We should all aspire to accomplish great things like AOC when she ...uh she did ...um ...drop Adderall and start a podcast about putting on makeup! Yeah! That podcast alone employs 50,000 engineers and has saved or created 100,000,000 jobs a month, at least.
He’ll likely have to sample the data anyway before he can do any meaningful analysis, so the whole exercise is pointless.
Edit: to be clear, I'm saying this about what Elon said, not you.
The hypothesis is that sampling those buffers at scale should be enough to detect significant bot activity on the platform, without needing any additional metadata.
That may very well be the case, so why the snide?
If the link is bad we’ll fix it with votes, not admin actionz
If we tried to go by upvotes alone, we'd end up with the same small number of hot topics and sensational controversies dominating the front page. This is the weakness of the upvoting system. There wouldn't be any room for the unpredictable sorts of stories that best fulfill the site mandate of intellectual curiosity.
I know it's irritating when moderators do something you don't like, but in case it helps at all, I think the irritation is pretty evenly distributed across the different subgroups who populate this place.
As for the OP, you're not going to get much support from the community here if you think that we should have left a hard-paywalled link on the front page. That's no doubt why users heavily flagged your comment (mods didn't touch it, btw, and I've unkilled it now).
If you have the firehose, couldn't you just only sample every 10th tweet and make a "decahose"[1] of your own to work with?
[1] Shouldn't this be "decihose" since it's 1/10 not 10x?
What kind of system architecture would be good to load & search through such dataset. We tried to explore Mongo Atlas but it is coming out very expensive. Other alternative is to throw away most metadata & just keep ID & Tweet.
By the way, you have to pay for this access or there are some other options available? It is interesting to ingest all this data by ourselves
Tweets come with a lot more data than that.
Tweet object model: https://developer.twitter.com/en/docs/twitter-api/data-dicti...
Example payloads: https://developer.twitter.com/en/docs/twitter-api/data-dicti...
So now they just have to play full defense. Musk says he'll walk because they won't provide data? They'll provide data. Musk says the numbers are false? Let him prove it. At the end of the day it's a signed contract.
The question is how quickly does this come down to "Okay let's cut a deal" - and my feeling is that since this entire thing has been an ego trip for Musk there is no deal that's acceptable for him to walk away without some sort of: Musk pays $XB and Twitter admits the bot problem is bigger than stated in some sort of way that doesn't cause legal troubles for Twitter. It has to be a "I didn't want Twitter anyway" narrative, rather than "Musk just stuck his dick into a meat grinder" type narrative.
They will let him walk away, but the starting number is $44B. Now they are just negotiating.
The stuff about bots and whatever else is a distraction.
Or, somehow, becoming the richest person in the world has more to do with other features like luck, than the ability to asses the market?
In the beginning Musk stated in interviews that he wasn't concerned about the monetary amount. I can't pull it up on my VPN, but I recall him taking the stance of it being a social-impact-y investment and not one that is designed to make a profit.
More to disagree with the argument than to defend Musk. He's certainly been doing a lot of things to really undermine the brilliant-at-all-trades image.
And luck is always underrated.
> Musk waived due diligence when he agreed to buy Twitter at $54.20 per share in April.
> We estimate that the average of false or spam accounts during the fourth quarter of 2021 continued to represent fewer than 5% of our mDAU during the quarter. However, this estimate is based on an internal review of a sample of accounts and we apply significant judgment in making this determination. As such, our estimation of false or spam accounts may not accurately represent the actual number of such accounts, and the actual number of false or spam accounts could be higher than we have currently estimated.
> Musk waived due diligence when he moved to buy the company, seemingly to hasten the acceptance of his bid.
> He wrote in a letter to Twitter Chairman Bret Taylor on April 24, "As we discussed, $54.20 has been and will remain my best and final offer, period. This is binary – my offer will either be accepted or I will exit my position."
> Twitter announced the company had accepted his offer the next day.
> But in the weeks that followed, Musk vocally criticized Twitter during media interviews and on Twitter, where he has tens of millions of followers.
When you buy a home, even if you waive all contingencies, including a home inspection, if the seller fails to disclose material problems with the house, the sellers will be liable. This liability extends for years past the close of sale.
So you accidentally gave a perfect counter example as to why you are wrong.
It seems like the people who are supporting this move assume Elon has some secret plan we're not privy too, but if so why would his actions be so erratic?
Analyzing the raw feed isn't all that hard for someone with Musk's resources. A fat pipe and a ClickHouse cluster could boil this down pretty efficiently. At least the metadata.
Analyzing tweet content would involve more effort, obviously. But bot operators aren't all that bright and much of their activity will be obvious.
The real question is would any conclusions of such analysis be actionable? Could Musk produce an analysis credible enough in the time available to lead to renegotiation? That is highly dubious.
It was just a few months ago that he had to sell stock to pay his taxes and he played it off like he was doing it because of some ridiculous online poll. I think this Twitter play is just more cover to sell more stock.
It will be very easy to see if that's what it was. If they deal won't go thru (I anticipated both parties will be stuck in court to fight over $1B for close to 5 years, then Musk will settle for some $200MM), Musk can simply buy Tesla stock back! Why he needs to all this cash sitting in a bank right???
2. Musk likely cares about all users that keep spamming his feed and others. The platform currently feels like armies of bots from political and ad/marketing agencies and some active users who think someone is reading their content. If one correlates the number of followers they have (especially if you’ve had twitter for a long time) vs number of reads on each of your recent tweets, the % is very low.
3. A firehouse of all tweets coming out may show a lot of bot activity (accounts with more numbers than letters in the username commenting in reply to popular accounts with a lot of followers) but they are not near the number active users who are just reading and that is where Musk’s claim and Twitter’s defensively worded SEC statements pass each-other like ships in the night.
“how do I get out of buying twitter?”
One point of time, it used to be a great discussion platform. But the low quality answers & hi-quality gossip, overtly aggressive "Be nice" policy (to the point every sob story has a happy ending, with cupcakes & unicorns, without taking any names or identifiable information) & login wall to view answers has made the site so so poor.
Pinterest & Quora are piles of flaming junk at present.
You would have to be sure that the sample is statistically representative of the overall population, which would require prior analysis with access to the whole data.
Example of biases that you would need to adjust the sample for:
- Maybe the part of active users receiving ads change shape with time, or is biased toward a certain type of user.
- As more users are active on the site, a random sampling would overrepresent recent messages over previous ones. Or maybe the reverse if Twitter lost users.
- Maybe %users being bots is not that interesting, versus %messages posted by bots. 5% of users being bots can represent 30% of messages being sent by bots.
- If you're willing to train some ML model to detect bots, you're most likely going to need as much out of sample as you could have.
- Etc
Asking for a custom-made sample is really the same as requesting the whole population and sampling yourself.
Musk agreed to pay $54/share. But after the tech selloff, Twitter might be worth $25-30. Maybe less. Musk can, in fact, get out of the deal for $1 billion or so in fees. But that would be awful for Twitter too, it's stock would tank overnight without a bidder (it already dropped from $50+ to under $40 quite rapidly). So perhaps Musk can get the board to agree to a deal at around $32-35/share, it would be a win/win for both parties.
For the first, they all know they’re getting fired the minute Elon can find a way. Maybe they care about money, but I bet they also would prefer to keep their positions. I bet a lot of them believe in and like Twitter, and it’s not about money.
As for the second question… they can’t be seen as the reason it doesn’t close. I firmly believe they’ve been banking on Elon doing exactly what he’s doing. They likely can’t be successfully sued if they say “hey, we accepted the deal and did everything Elon asked…”
Not gonna lie, I'm torn on the whole deal, bit I still think it'll be a good thing as long as there's actual free speech, but with realistic limits and not just one-sided moderation. He better not mess with Twitter design though. It's really good and honestly appears to influence most of the other apps in some way, shape or form, but I'd also be excited to see some design changes that are new in the industry. I also want the app to be faster.
[0] btw I bet Elon and Twitter disagree on the definition of "bot." Is a low paid human worker who tweets spam and scams all day a "bot*?
Not like, 5% fake. Not 15% fake.
No it’s more in the range of “most clicks are fake but we just live with it anyway.”
If you run PPC campaigns you also know how easy it is to spot this crap.
I for one am interested in learning what’s going on there. I don’t buy the “it’s soooo complicated oh it’s just impossible to know for sure!” Narrative that Twitter is peddling.
I'd argue real people rarely click ads
for facebook I read that most of the advertising money is spent on clickfarms, which just click anything to seem real e.g. follow 100 real brands and sell 50 fake follows on that acc.
He's liquidated nearly $10B for this deal already, has a remaining ~125B in Tesla stock and ~55B in SpaceX, plus a long list of billionaire partners for this M&A. I bet he has a few more options than going "bankwupt".
The thought may be titillating some, but it's wishful thinking.
It's only after the price of both Tesla and Twitter fell significantly that he suddenly became very concerned about the number of bots (a matter which he had been claiming is false in Twitter's filings for years before signing the contract to buy Twitter without further investigations).
Will that help him find the bots though or will he just find a Bacofoil Tent Roasted Turkey masquerading as a Tesla Cyber Truck?
Anyway, anyone know what address we need to use to submit a GDPR request as this is him personally, not Tesla or any other entity, ie personal wealth so does anyone know his home address so I can submit a GDPR request in person?
If you gave me 100, could I spot the 5 that are bots? If you gave a team of paralegals a pile of 10k random accounts that are actively writing content, could they find the 500 that are supposedly bots?
Perhaps I’ve answered my own question: the bot accounts that get people hot under the collar are the ones that simply like and retweet, without ever posting any unique content?
Do we need proof? You can safely assume that nobody who behaves like that is as smart as they advertise. Don't wait for the hard evidence.
If you have > 2,000 followers, you pay $0.01 per 10,000 followers per tweet. At current follower counts, Elon Musk pays $97/tweet. CNN pays $58. McDonalds pays $4.50.
Pay the same amount to reply to a tweet/thread of an account with that many followers.
Bot problem solved. No need for an ad platform, ad targeting, etc. The whole thing is just paid ads.
Don't get me wrong, Twitter is a garbage fire, but this idea is just... bad.
The poster is saying - why keep up this charade of being a social media platform at all, worrying about bots harassing your ad-viewers and so on? Why not embrace how the platform is already going, and get more direct monetization from those looking to broadcast?
If you want to see platforms for voicing opinions of the masses, Facebook and Reddit are doing much better at that, witch much more organic conversation and interaction.
Musk is in this case - not always - a troll (he even said, early on, he was going into 'goblin mode' or something like that). The primary aim of trolls is to disrupt and gain attention as much as possible, not to accomplish anything. Accomplishments undermine trolling; they complete something; they are constructive or anti-disruptive; they end the trolling. From that perspective, if the deal went smoothly and quickly, it would be a failure; if it concludes, its value is lost. The troll would want to drag this out as long as possible, with as many disruptive and attention-getting events as possible. The deal is a vehicle for trolling.
Would Musk risk $40 billion on it? Would you be surprised if he thought he was smart enough to play this highest-stakes poker game and get away with it? If he relished it?
Maybe this whole thing was a way to show everyone it was worthless and now he's going to spend $1B on getting out of the deal and another $1B in starting a Twitter clone.
Musk cannot just wake up one morning, decide to not buy Twitter, and pay the breakup fee. He has already agreed to buy Twitter and cannot back out.
That is why the bot issue is so important. If Musk can prove that Twitter made false claims to the SEC, he is allowed to get out of the deal. But if not, he must buy Twitter
Saying something confidently does not make it so.
And indeed, Musk's own letter to the SEC confirms that his sole reason for seeking the data is that it relates to his ability to acquire financing for the deal.
https://www.sec.gov/Archives/edgar/data/1418091/000110465922...
They are billionaires. They will come out on top because at the end of it they will still be a billionaire.
Name one billionaire that isn't on top.
Donald Trump.
2. It seems entirely unreasonable to assume that Elon doesn't want to own Twitter, or that he tried to buy it on a lark and changed his mind. If your premise is that Elon did want to own Twitter a few weeks ago and doesn't want to own Twitter now you should rethink your opinion.
500e6/(24 * 60 * 60)~= 6k tweets per seconds. It's not nothing, but not that hard to handle either.
> it will take Musk considerable time to do an analysis of bots on Twitter, a task requiring a team of researchers who will need to laboriously construct software to review the tweets.
...is just not true. A few engineers can bring up infra to first filter & sample the firehose, and then store only what they need.
EDIT: I'd candidly love to know how what I'm saying is completely off base, if downvoters wouldn't mind educating me.
It's not something Musk can do algorithmically or through brute force data analysis, because that sort of superficial analysis would be very inferior to Twitter's very involved process, and Musk would have to demonstrate that the algorithm that he or his team came up with on short notice is more accurate than the process Twitter has been using for several years. Which isn't going to happen. See: pretty much every Musk "first principles" claim ever.
Musk will need to hire a lot of people to do this job, but he can't just drag in SpaceX and Tesla employees to do this for him, since he would be (rightfully) sued by the other shareholders of these companies for abusing/misuing company resources. Also, Musk is pretty well known as a horrible boss that likes to badmouth his employees. It's going to be very difficult for him to assemble a team of qualified people to handle this task for him, especially in the time frame he has.
You got downvoted for doing the programmer thing of thinking that a very difficult problem can be trivially solved with some light programming, and completely ignoring that people who actually deal with this sort of thing for a living think that this is a very large and difficult task that will take a great deal of effort by a great many people.
Even if 100% of all tweets were generated by bots, as long as real people come to Twitter to read them and occasionally click on ads that Twitter shows them, Twitter's revenue is secured.
Now I think Elon has been acting in extremely bad faith throughout this process and I hope Twitter sticks to their guns, but I don't have a lot of faith.
unless I'm missing something it seems pretty cut and dry.
We know that. Twitter makes it clear publicly in their annual reports.
It's to prove that there is a material fraction of fake tweets significantly above their number.
I suspect that Musk was willing to purchase Twitter if legitimate, and knew if Twitter was fraudulent, this process would shed light on it. If the deal is void due to Twitter misrepresenting itself, Musk can invest in a legitimate venture instead.
I don't see a whole lot of reason for speculation. The deal will close, or it will not. Chances are, Musk wasn't kidding when he made the offer, which means if it doesn't close, Twitter misrepresented itself.
Their filing on active users are extremely narrow in their claims - the only way they could realistically be shown to be fraudulent is by discovering in internal emails that they were crafted to deceive; otherwise, they are very specifically saying that the 5% number they give is only an estimate and that it involves a judgment call in how it was decided. So if you find some way of calculating bots and come up with 20%, that still doesn't mean they misrepresented the numbers: they never claimed 5% is a precise number, just their judgment based on some sampling they did.
Futhermore, even if the number is found to be a lie, that still wouldn't mean they materially lied to the SEC - you still have to then prove that this impacted their revenue projections significantly (apparently the standard is "at least by 40%"), which comes with even bigger burdens of proof. You would have to show that Twitter's clients (advertisers, mostly) were relying on the reported numbers to decide if they were getting a good deal, and that they would significantly change the price/number of ads on Twitter based on this difference. This in itself is unlikely, as these companies have their own metrics for judging whether an ad was effective, and in the many years Twitter's official bot estimates have stayed the same, they would have had time to notice if their revenue from advertising on Twitter was worth it.
Even worse, Musk has been claiming that Twitter's reported bot numbers are much lower than reality for many years before he made the offer to buy Twitter, so his belief can't have materially impacted his willingness to buy. He even publically talked about cleaning up Twitter of it's bot problems as a specific target after the acquisition. These public statements will again significantly affect any claim that Musk makes that any new information he finds on the number of Twitter bots would have impacted his decision to sign the contract if it had been disclosed before hand.
Finally, note that the information he asked Twitter to provide (access to tweets) has almost nothing to do with the discussion about revenue, as Twitter's revenue comes from users reading tweets, not writing them. The 5% number is about users who request the Twitter feed and receive ads, but are in fact bots and thus can't buy products. This is completely uncorrelated to the number of bots and other fake accounts who might be posting tweets, as there are no ads in the tweet posting pages.
In short, if Musk analyzes the data he received now from Twitter and discovers that 100% of all tweets are actually made by bots, it will not make one iota of a difference to his obligation of paying the $44B he singed on. Twitter makes no claims to the SEC about the number of real VS fake users posting tweets, and this number has no direct impact on revenue. The only claims by Twitter are that by their own internal process and judgment, up to 5% of monetizable daily users are bots, and this can only refer to users reading tweets, not posting them, based on how Twitter works.