We advertise on X and have done so for many years now (well, it was Twitter for most of that period). Indeed we can confirm that we saw a big change happen around a couple of months ago or so - sometime towards the end of last year.
Our ads started getting literally orders of magnitude more traffic - we reduced the bids by a factor of 30x (no typo - that's thirty) and were surprised to see them taken up to the budget limit completely... we ended up spending a minor fraction of what we spent before for a lot more traffic.
So far, so good, right? Not so much. None of the traffic converted anymore. All useless. I can't remember the last actual conversion from X our conversion tracking software recorded at this point... weeks ago for sure. Formerly we got several a day.
Your experience with twitter is similar to my experience advertising on reddit a couple of years ago. Their ads were dirt cheap compared to on other platforms but nothing converted. Most traffic from Reddit were roundtrip clicks with average sessions of about a second.
I would say that, from a user perspective, I do accidentally click on a lot of stuff on Reddit and Twitter. Almost every third post is an ad.
About 99% of the ads I’ve clicked on Reddit were by accident when I hit the back button and then try to click a link below an ad just as the page reflows to fit a taller ad and puts it under my cursor
which runs some great articles but also runs a stupid roundup on large HDDs every month where they don't do any research but scrape prices from AMZN.
The thing is when I am visiting that site on mobile the content is always jiggling around, I think it is no accident because it means you want to click on a link but ka-ching you clicked on an ad.
> I would say that, from a user perspective, I do accidentally click on a lot of stuff on Reddit and Twitter. Almost every third post is an ad.
As a user, and not a person with advertising dollars to spend, try using reddit in firefox with ublock origin installed as a plugin. The experience is significantly better.
Also works well with firefox on android and ublock origin.
This might partially disappear with the nitter loophole closed but the number of bots under musk has gone up an order magnitude. I used to have 5 followers now I have anywhere from 70-100 depending on the day and I guarantee you I have not become more interesting. It’s all bots.
With nitter gone a lot of burner accounts are being created which no doubt they’re going to claim as a metric win. I’m running my own scraper with a lot of burner accounts. They can’t stop us from scraping.
Why do you think nitter going away will change anything? I would expect the set of users purposefully using nitter to be completely contained in the set of users with adblock on.
I don't get what you mean here. Same technique as in loading the website? If there's money to be made, you have so many more options including paying people to register real accounts. Nitter does not have the same available motivation / resources as people who run serious scams.
Spammers will need to generate fake accounts to scrape using a different method than before. It may be more expensive but I doubt it’s prohibitively expensive
But, no one should be advertising on Twitter for any reason anymore. It’s not brand safe and it’s always been absolutely trash; but, now it’s even worse.
Shift your budget to any other meaningful platform within the channel; you’ll be glad you did.
I was in Hong Kong for two weeks late last year and every time I opened Twitter there I saw a promoted post for a voachinese.com article (something something China bad). It’s always VOA Chinese even though my language is clearly English, and my account is completely passive, never tweeted anything, liked anything or followed anyone, most of my visits are HN referrals about something tech related. It was quite surreal, like they bought the entire ad space of the region or something.
As of a few months ago, 99% of the ads I see are cryptocurrency scams. Blocking them doesn't help because the same ads are each being posted by dozens of accounts.
Anyone who has been on Twitter lately could tell you that. Click any mildly popular thread, and you get about 25% crypto scams, 25% OF, 20% irrelevant clickbait, and maybe 25% at least somewhat relevant replies. The other 5% are also bots complaining that the replies aren't about the post, odd as that sounds.
The blue checkmark is the main cause of this, or rather, the fact that it automatically boosts your comment over those who don't have one. This led to the current engagement farming meta, where none of the comments are about the OP itself (save for smaller accounts and even then.)
EDIT: Sometimes I find myself holding the spacebar, doom scrolling to reach the "peasant" comments, because at least they're actual comments.
I remember when a large portion of Hacker News commenters were praising the blue check changes. Comments such as:
> Fewer bots, less brigading, ideally no ads, less spam, etc.
> I am actually quite optimistic. Let the best ideas win. The blue check elitism is coming to an end.
> Given twitters known bot problem, the number of bluechecks will likely be muuuuch lower as real people start to pay for them, and the real people behind the identities have to pay for them instead of their PR staff running them, but this in turn will make them more valuable and meaningful.
Control Panel for Twitter [1] can automatically hide boosted blue replies for you
…although if an account is big enough, you're safer looking at the Quote Tweets instead for actual comments (it also restores the old direct link to those in the focused Tweet), e.g. you're lucky to get more than a handful of non-blue replies under an Elon Musk tweet among the engagement farmers before you hit the maximum number of replies Twitter will load
I think one limiting factor is the fact that most posts aren't particularly deep or insightful, so the intellectual space from which one can engage on the subject is pretty limited. So most people who engage with content of this sort are just clout chasers trying to get attention or scammers.
Is it really surprising that if Elon amplifies some post with a single exclamation mark all of the replies will be completely inane and full of spam?
Another factor is account size. You can still find interesting replies in accounts that are relatively small, especially when the author takes time to manually hide spam and scams. But once an account goes beyond a certain size they just become a magnet for nonsense. It's pure quantity over quality. Elon has the biggest number of followers but the quality is abysmal.
Ironically discussions tend to be better in circles where nobody wants to pay for the blue checkmark on principle, because the comments there are still sorted using the old ranking system. But as soon as bluechecks start posting they get shunted to the top regardless of quality, burying all the good replies from non-bluechecks.
I mean, my Facebook, which I visited for the first time yesterday after months, was almost entirely "fan" pages showing AI generated images of famous actresses in skimpy underwear. Like 90%. I suppose the algorithm has decided that if I read the posts by sci-fi author David Brin, then I'm in a group that also wants dirty pictures of actresses. o_0
I went through a phase where it randomly showed me what I assumed were Thai prostitutes because a spammer created an insta account with my email. After I reset the password on the account it eventually subsided.
I recently had a fake account follow me where they replaced the letter L with an I in the user name. Whatever font Twitter website uses, makes it impossible to visually see the difference. The fake account looks 100% the same as the real, even has thousands of 'followers'. I reported it, and it is still up.
It is unreal to me that the platform hasn't developed a way to automatically deal with fake accounts like this. Just check to see if the profile image is the same!?
I had a fake Yann LeCun (Meta's AI chief) follow me. It looked and read like the real thing. I was happy about it for a couple of days until I realized it was fake. It fooled a lot of people. I didn't report it because obviously Musk doesn't care.
I can't imagine working for Twitter, in a position to fix things like this, and having to listen to Musk tell me to 'stay the course'. I know that everyone has a price, so their salary must be insanely good.
I suspect that only the true believers are left. There were _so_ many actions and warning signs over the last year I can’t imagine anyone thinking it was a good/stable place to work unless they really liked Musk.
I think there's still a bunch of people who have an audience and don't really care about the drama. They're there not to engage, or sell anything, but just to use it as a broadcast medium. But yes why would I ever talk to anyone there? I can just use nitter to read my neets and call it good.
For creators, I’ve heard other services offer much more useful interaction per follower, which would mean either Twitter followers often aren’t shown posts of a large cube of followers don’t care (likely bots).
I don’t think Musks behavior and obvious preferences should be ignored. But even without that it sounds like it’s dead at the core and more people figure it out every day.
I can't imagine the hell it must be to be stuck this way at X. Can you imagine being the engineer that gets the email "Elon wants Séamas O’Reilly suspended because he said some rude things about X" and you happen to be Irish?
Journalist says he finds it ‘surreal’ to have account on X suspended after writing critique of platform
Dunno how “new” it is… I’ve definitely got a big price difference and I’m pretty sure others will to…
Twitter, Tesla, Neuralink, Boring… all the non space Musk corps…. Quarter million if I don’t have to move from Australia, half million to a million depending on where I was going if it requires moving to the USA… I’d do it… but I’d basically have the resignation letter already drafted from day one…
For the rockets… I’d probably be happy with their normal pay offer… not being a US citizen basically makes it impossible for me to apply for a job with SpaceX so I’d take what I could get lol… and since Elon is clearly just glued to his phone tweeting… it’s not going to be the Elon show… Gwynne Shotwell is running the company like a level headed and competent management executive, the place is full of competent engineers who seem like pretty well adjusted people when they talk about their experiences after leaving, and they all seem to leave for normal reasons…
Was the fact this was not as significant an issue due to Dorsey and then Agrawal being really good moderators as individuals?
Or do you think maybe business leaders are accountable for the behavior of the system beneath them, especially when they personally overhauled exactly the system in question?
> Whatever font Twitter website uses, makes it impossible to visually see the difference.
Doesn’t it? When I view the profiles you listed (before that horrible login wall pops up), the ‘I’ is of the crossbar variety, and the ‘l’ has a finial, making the characters visibly different. In fact, I recall that Twitter started using that font (which also visually distinguishes 0 and O) after the Musk takeover as an anti‐spoofing measure. Spoofed usernames were completely impossible to detect before the font was changed.
Sure enough, 17 new bot followers today after blocking 14 only a few days ago. There's precisely zero reason any legit person should be following my entirely passive, unengaged account. Twitter truly is a cesspool.
It's interesting to me how hot and cold descriptions are of Twitter. You get posts like these, but then someone chimes in about how much better the conversations/etc are on Twitter since Musk took over.
I'm not on Twitter so i can't really make sense of it. I feel like i see more negative than positive.. but still.. it's bizarre to me that there's people in both camps. More than likely some of them are biased.. but still, i find it "interesting".
There are a lot of people for whom everything Musk touches is the worst thing ever created, then a lot of people for whom everything he touches is gold, then some people who just like a pretty reasonably stable social media site that's not overrun by bots and pay-for-engagement morons.
IMO it seems objectively true that the bot problem is worse than it's ever been. It's definitely objectively true that boosting paid-for comments above organically high-engagement comments means your signal:noise ratio is way, way worse.
It's not really clear at all to me, since, eg some people here are talking about getting several bot account follows daily, while for me it is the opposite, where I used to get a lot more bot follows than I have gotten over the past year or so (down from 4-5 per week to less than 1 per month).
On the other hand, I do see a lot more bots in replies (haven't had any reply to me yet though).
Overall my opinion has been slightly positive, my usage of the site is up and I'm hearing far less of people I follow (or those they follow) who have been unreasonably suspended or forced to delete tweets. It feels a lot "wilder" than before, but that's kind of what I enjoy anyway.
But I also don't really engage with political Twitter, I only use it to keep up on space related reporting and weeb stuff. The worst part has just been a lot more engagement farming through dumb statements from content creators who are circling the drain.
Yeah, I also engage little on Twitter, but so far I’ve seen no notable worsening of spam. When I clicked to see popular announcements (from HN or somewhere else) I also saw crypto spams and other stuff but from my perspective, it’s not much different from the situation on FB, for example. When I replied to some interesting GTP-chess posts, I still got replies clearly from humans (ask questions, do some follow up etc.). Thus, I don’t have a clear feeling whether it got better or worse.
The most striking change I have seen is the percentage of Musk’s tweets on my feed. It looks like he is everywhere now! That was definitely not the case before. But yeah, he is the owner and he likes to tweet :)
Please share some examples, I never see crypto scams anymore and pre-acquisition you could go in the replies of any popular account and it would be nothing but that. I'm guessing it's still there but I don't see it because I don't scroll for hours on end and verified posts being boosted at the top suppresses the visibility of any bots. Either way considering how bad the bot problem was before the change in ownership I feel like people have a very selective memory about this problem. I did get a handful of catfishing bot accounts following me a few months ago, but that's about the extent of what I've seen as far as spam goes recently and feeds are much cleaner than they were a couple years back. If you're comparing to how it used to be several years ago then it's a different story.
> It's definitely objectively true that boosting paid-for comments above organically high-engagement comments means your signal:noise ratio is way, way worse.
I'd like to see an objective proof for this alleged objective truth.
The bigger problem that I see constantly nowadays is cherry-picked information and people outright lying about a picture or video that they're sharing, but thankfully the community notes help a lot with that.
Go ahead and flag this comment for not being prefixed with "Grr I hate Elon Musk!"
I don't care about Musk one way or the other and I don't see any evidence to support your claim. The level of bot activity is no higher than before in terms of comments on tweets by accounts that I follow. But why bother reading comments anyway? That has always been mostly a waste of time.
To the Nazis that were kicked off or fled when Twitter had decent moderation to let them know Nazi conversations are flowing like wine and they're welcome back.
Rephrasing for people who will react negatively to that. You're making an important point but "dog whistle" assumes some things:
Those are people speaking abstractly about how much more free their speech is*.
I've never heard that common phrasing used, in any setting, to refer to a change in Twitter user makeup, or it becoming a kinder place. Whether I'm in the company of team left or team right, and coding as same team or opposite team.
* yr humble author refuses comment and does not endorse this viewpoint, or that anything changed on this front
I ended up in a really unfortunate position, pre-elon I never really "got" twitter and so barely used it. When he took over I'd log in every now and then to see how much it was plummeting, and in the process of doing so found the (ever shrinking) value of the platform, so I only started enjoying it when it started going downhill (for me at least).
I guess though that's less the platform or the people running it and more the users. There's nothing particularly unique about twitter that I like other than the other people who use it
Is this based on the kind of politics you are into? E.g. if you are from US you either have one side or the other where one side likes it and the other dislikes it?
For people with interests or opinions that were heavily censored by pre-Musk Twitter, the end of this censorship alone compensates for the countless new annoyances.
This seems vibes based. Twitter censored in the past and continues to censor. If you were in the out group and now are in the in group you’re happy even tho you were unlikely to have been censored
I've always found Twitter to be kind of dumb, most of my interaction with it is when twitter bullshit gets plastered over UFC PPVs or "news" articles report on what people say on twitter.
Censorship hasn't ended, it is simply impossible not to censor.
There is obviously illegal content (child porn, ...) and copyright infringement, you have to censor it or you will get sued and lose. Related is libel, doxxing, harassement, revenge porn, etc... that may also get you in legal trouble. And there are countries other than the US with different laws, for example many European countries ban some categories of "hate speech", and if you want to do business with them, you need to follow their laws.
And there is spam, if you let all of it pass, it will simply drown everything else.
And at some point, you may want to make money. Usually, you start with advertising, and if you want people to advertise on your platform, so you have to be at least somewhat socially acceptable. You also have to please payment processors, if you bring the wrong kind of people, Visa and the likes won't want to deal with you. On that last part, considering Elon Musk past, he may have workarounds, but the reason Visa has a problem with porn is not just because they are prudes. That's because there is a lot of fraud happening here, and it will happen too on a socially unacceptable network.
I somehow doubt that the audience that's going to make Twitter money with subscriptions or advertisers are Nazis, MAGA, Crypto Bros, and OF performers. But you do you.
Instagram is the same. They have bots to give you the feeling that you jumped right into a group of people, while it's actually a desert like environment.
And the stupid optimization tricks even legitimate users do to increase their clout: post several pictures - "Which one is your favorite? Comment below" so they get engagement points by the number of comments when it's just a stream of numbers that no-one (probably not even the creator) cares about.
I gave Musk a lot of leeway on this for taking over right as open-source LLMs started ~passing the turing test, but that they can't even stop this is eating most of it.
The paid blue checks basically created a bot-engagement tier. for $8/mo you can protect your account from getting banned(profit motive to retain paid checks) AND get your spam to the top of every post.
I saw an account with a $1000 gold checkmark impersonating a genuine goldcheck account to promote a crypto scam, that's the deluxe spam boosting package.
AND now you get paid for views, so just have your bot account post a random stolen video and spam it under every single post that goes anywhere near viral. Boom, easy money.
And at the same time they put streaming API access behind a $5000/mo paywall so there's no way for independent researchers to easily research/report on botnets, astroturfing campaigns etc.
Musk Twitter has not done any better than Old Twitter at containing spam, but it has managed to at least monetize it, which you have to admit is a small step up.
To me the weirdest part is that all the bots I see have blue checkmarks. How does it make sense for the bot operators to pay money to run their bots? Are they using stolen credit cards?
It would be interesting to know how well X detects signups with stolen credit cards. If scammers have millions of credit cards from data dumps and make a $8 charge on X, what portion of the victims would even notice? Even if X receives a fraud report and acts decisively to shut down the offending account, it might just be an inconvenience to scammers who have signed up for many more. At scale, fraudsters could be moving a substantial amount of stolen money into X membership fees.
You don’t need literal stolen credit cards to dodge CCN bans any more. There are so many services today that allow you — without KYC! — to create an online wallet that is assigned a real CCN, and which also don’t present as a “gift card” on systems like Stripe. You just steal or scam people out of liquid assets of any kind (e.g. crypto), and you can then turn those into as many virtual credit cards as you want — all under different names, with different addresses, etc.
Many of these services also allow spending limits to be set on these cards on a per-card basis — which in practice allows the defrauding of any postpaid service, by limiting each card to spend just enough for the initial card verification, and then denying the actual postpaid charges when they come. (Then you close the acct, and open a new one with a new name + card.)
To combat this, we’ve resorted to using grey-market(!) “BIN lists” to determine what CCN prefixes (BINs) are used by the banks backing these non-KYCed online wallets, and just blacklisting all of them. Somehow, I doubt bigcorps are doing the same.
Man this is gold. Thanks for sharing. Would be curious for any naming you can do re: where to look for said BIN lists, as well as, a company or two that does this CCN thing.
It let's twitter run fake organic ads without having to tag them as ads. A paid account is basically an ad post account boost as long as you don't go too hard against the spam rules.
Sure. But I would expect bot operators to have a vast array of accounts. If you have 250,000 accounts and you paid for a blue checkmark for 10,000 of those accounts. That would be $70,000 per month.
You only need 10,000 accounts if you expect bans. Bluechecks don't get banned so you only need 1,000 not 10,000 (made up numbers) and in that world, you just steel some crypto to fund a thousand credit cards (or similar) and you're gold, forever because Twitter won't ban paying customers -- it's too desperate for that.
It's more a question on how much they make per account though, right? It's either above or below the profit line on average. Or they've found a way to reduce costs, like stolen cards or owning Twitter (hah).
> a question on how much they make per account though, right?
Idk, but intuitively I would think that the profits they get are spread over a great number of accounts.
If 98% of your spam earned income comes from a single account, that would make you very vulnerable to losing all of your income if that single account is banned.
If on the other hand, 80% of your income comes from 10,000 accounts then it doesn’t matter if 1 or 10 or 100 or 1,000 of those accounts get banned. You’ll still have the income from the remaining 9,999 or 9,990 or 9,900 or 9,000 accounts. And you’ll make a bunch of new accounts all the time so that for all of those account of yours that are banned it’s not really significantly reducing the total number of accounts under your control either.
Of course, I could be wrong and it could be that they have a handful of accounts that make all of the income and all of the other accounts only serve to fake engagement for those few accounts so that those few accounts get boosted and seen by people. I’m not in the spam making “business” so I don’t know how it actually works.
There's so many weird things about that specific abuse.
The frequency...the inability to stop it after 2 months...
The weirdest thing is they seem to get some special treatment that renders them invisible, yet, not prevent them from being posted altogether, or have any impact on the account itself.
It's not even particularly productive spam. It's like watching someone thumb their noise in front of a steamroller. Makes no sense.
“Stop the bots” was a made-up rationale when he wanted out of the deal.
The original intention was to buy Twitter so Musk would be the most popular man in the world, his comedy would be appreciated by a billion followers soon, and as a nice bonus they’d buy whatever crypto he decided to pump. The Twitter buyout was maybe the very last major business decision executed entirely on 2021 moon logic.
> they’d buy whatever crypto [Musk] decided to pump.
For a moment, I'd forgotten that was a thing Musk actually did. I got so used to bots impersonating him pumping cryptocurrencies, my memories had rewritten the times he did do it himself into more impersonators doing it instead.
Lol @ the Twitter purchase being a "jump the shark" moment for the gogo COVID era
Lol'ing at its accuracy, and it being obvious once you hear it, but something that didn't cross my mind until now. Even the constant rolling firings at BigCos: I find it hard to believe companies would have cut as quickly and aggressively and continually as they did without Musk's...housecleaning?...making them look reasonable in comparison.
> The Twitter buyout was maybe the very last major business decision executed entirely on 2021 moon logic.
You know, a part of me hopes he makes it to Mars in my lifetime and just proceeds to take over the planet, running an alternative society there. Planet Musk: The Reality Show™ would be quite the thing to watch... from 140 million miles away.
I feel like this is how we get a much dumber version of the MCR from The Expanse. Like, they wouldn't be justifying their weird space fascism as a means to terraform Mars, it would all just be one big joke to fuel Musk's ego, and somehow a bunch of people end up being stupid enough to buy into it.
In the past few years, I’ve read at least two sci-fi novels (an Alastair Reynolds one and one other, maybe Ken MacLeod) which made passing reference to a failed Musk Mars base. Seems to be a popular idea.
I always did wonder if the Musk Dogs species from Pushing Ice - who trick the colonists into blowing up their planet or something along those lines - was a vague dig at Musk...
I have installed plug-ins that totally strip all recommendations, shorts and other stuff from youtube to the point where it only displays the one video that I was interested in. This probably saved me more hours than I care to admit.
'unhook' and 'hide youtube shorts' are the main ingredients.
Not defending Twitter per se but I’ve seen partial to full nudity in YT recommendations after watching fairly innocuous but popular videos.
For lack of a better term to describe it if you watch enough “normie” videos it seems to start showing you anything with a high view count. Watch 100 silly cat videos and it will start recommending you celebrities and football and ofc naked girls. YMMV but that’s what I’ve seen. Worse if you aren’t logged in.
At least it’s better than Instagram where softcore porn is the rule rather than the exception
> CHEQ also provided Mashable with fake traffic data from the entire month of January 2024. TikTok, Facebook, and Instagram all had very similar stats to each platform's respective Super Bowl weekend numbers. Slightly more than 2.8 percent of the 306 million visits sent from TikTok were determined to be fake. Out of the 90 million visits that came from Facebook, a bit more than 2 percent were fake. And Instagram's traffic was only 0.96 percent fake, based on 749,000 visits.
The relative scale on visits here doesn't make any sense: TikTok 306M, Facebook 90M, Twitter 759K, Instagram 749K.
This seems like marketing for a snake-oil bot detection product masquerading as a political hit piece to get attention.
>The relative scale on visits here doesn't make any sense: TikTok 306M, Facebook 90M, Twitter 759K, Instagram 749K.
While I agree that this feels like an ad for CHEQ, their scale would presumably be based on how many ads their partners placed on various platforms. They could be buying far more ads on TikTok than on Instagram for some reason.
'Ad fraud' and 'bot traffic' tools really don't have any true insight into these walled gardens.
None of them are running their own code on anyone's device to track anything, at best they're processing data given to them from a Meta/TikTok/X etc. API that might have a bit more data other people don't have access to.
Maybe that uncovers some new insight and some egregious cases of fraud occasionally, but ultimately the big platforms are basically just saying 'trust me bro' and getting rubber stamped by these vendors. I don't doubt X is full of bot traffic, but Meta might just be better at fudging their numbers to CHEQ for all we know.
I would really like to know more about how they determine what is "fake" versus real.
> CHEQ monitors bots and fake users across the internet in order to minimize online ad fraud for its clients. Tytunovich's company accomplishes this by tracking how visitors from different sources, such as X, interact with a client's page after they click one of their links. The company can also tell when a bot is passing itself off as a real user, such as when a fraudulent user is faking what type of operating system they are using to view a website.
This does not seem anywhere near foolproof, and seems like it would have a small sample pool to analyze. It seems to me the overwhelming majority of bots are not following links... meaning they are sampling only a specific type of bot behavior and attempting to extrapolate that to the entire platform. That seems fallible.
Additionally, the compared sites are not... comparable. It's not fair to compare twitter with Mashable, TikTok, Instagram, or even Facebook. The type of community is different, and the type of content is different. Twitter is much more of a broadcasting/informational platform (ie. shouting into the wind) and therefore lends itself to bot accounts a lot more than your average Facebook account, for instance.
This is why for years the Twitter API (aka. "Firehose") has been such a contentious issue and other social platforms like Facebook's have not.
Can confirm. I don't have cable so I relied on the "latest" tab in X. Almost all the traffic was for sites purporting to be showing it live, but weren't because they were some sort of scam.
You get this for Premier League football games too - hundreds of accounts posting "WATCH LIVE!" with hundreds of sketchy URLs. In defence of the current owner, it was like this before he took over and it's not noticeably worse (from my observations - probably is if you do proper analysis, etc.)
I think it is noticeably worse, because some content is now absent (especially cybersec, at least in my timeline).
The only use of twitter i have left is following US sports, and i guarantee you it's worse.
Also, i hve the distinct impression that women left the platform. I think it was already one of the most masculine coded social media, but now with the rebrand (both logo and new name) i would be surprised if the number of non-activist women on it are really, really low.
I got a similar call telling me the Superbowl was today. I watched it, and the 49ers had an amazing victory!! The halftime ads for some crypto thing weren't very good though :(
That's because Gavin Newsom called up Roger Goodell in what was a perfect phone call, and demanded: "What I want to do is this. I just want to find, uh, four points, which is one more than we have, because we won the game."
The absence of specific information regarding the publication of data reports could suggest that the dissemination of detailed reports or analyses to the public might not be within its standard practice or could be subject to certain conditions.
Wasn't Musk supposed to end all the bots? How are they getting around the limits?
Even Nitter is packing its bags because the site and API are so restrictive.
And I had to remove twitter title fetching from my IRC bot ages ago because the API shut down and would've been too expensive for a hobby project. It literally didn't do anything except use the API to grab info about a url to a tweet.
I honestly think removing API access was done to prevent researchers from getting accurate data out of Twitter, making Twitter themselves the only source of information around platform usage.
Spam has obviously not been reduced in any way by the API changes. I think they already had an easy time tracking API account misuse and knew it wouldn't reduce spam in any meaningful way.
Maybe a bit of a conspiracy theory. Also possible Mr Musk just decided API access = spam and shut it down on a whim. Definitely not easy having a boss who makes snap decisions like that and doesn't tolerate any pushback or dissent.
I'd imagine that Elon and Twitter would deliberately keep bot numbers in their reports to make it seem like the website is doing better than ever.
Can Elon / Twitter straight up lie and share fake or misleading numbers? Would that be illegal and result in any kind of legal action, or are would they be allowed to lie and mislead because they're a private company? It's not like these reports would be audited by any external sources. Asking genuinely if anyone is familiarized with this legal domain.
Being public didn't stop him from lying about Tesla information. This would generally be fraud, whether public or private, but that is extremely under prosecuted.
He should just sell twitter and focus on real things. This whole saving the town square free speech thing was fun for a bit and now everyone knows twitter was never the town square and free speech was never a problem. Tesla stocks are tumbling because of his bullshit.
Tangentially related but I recently signed up for a Twitter account post Elon acquisition and the last step of the process requires you to follow one account to start using the site, the first option is Elon Musk and everything else is random based on your selected interests. Are any numbers on this site not juiced at this point?
Yeah, this is reportedly a measure they put in place after he bought the company because he was unhappy with engagement on his posts. It doesn't exclusively boost him AFAIK.
That’s actually a pretty good reason to flag an article. It’s not much different than any article mentioning politicians. People aren’t able to engage on articles surrounding musk or any of his companies intelligently.
Our ads started getting literally orders of magnitude more traffic - we reduced the bids by a factor of 30x (no typo - that's thirty) and were surprised to see them taken up to the budget limit completely... we ended up spending a minor fraction of what we spent before for a lot more traffic.
So far, so good, right? Not so much. None of the traffic converted anymore. All useless. I can't remember the last actual conversion from X our conversion tracking software recorded at this point... weeks ago for sure. Formerly we got several a day.