Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Twitter says Musk’s spam analysis used tool that called his own account a bot

arstechnica.com

Post reply on HN
Print view View original

461–470 of 585 posts

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#461

Post by vba616 »

Banana699 wrote: post #320
dragontamer wrote: post #202

Earlier quoted context omitted.

> Legally ? I don't know, I'm not a lawyer, and law can frequently have surprising implications in the twists and turns of the its logical structure, which - as a bonus - differ by time and exact place. The contract states the place of the law. Specifically: in the Delaware Chancery Court. The contract also states the time and conditions of the contract. Have you read the contract? Its public. Furthermore, the date o…

How does any of this change or contradict what I said? And doesn't the US legal system allow for escalation to the Supreme Court, which can override lower courts? So technically whatever the Delware Court is not the final say on whatever happens. >Trial date set to October This is so so awesome, I will get to see Twatter defenders rage and fume for 2 more months? ^__^

An expert commenter on the topic (not me) wrote that there is only one potential step in the appeals process, and that is a "Supreme Court" which is in Delaware, not the famous "US Supreme Court" you may have heard of.

First Google hit for "delaware chancery court appeal":

"The Supreme Court is the State's appellate court which receives direct appeals from the Court of Chancery, the Superior Court, and the Family Court"

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#462

Post by RandomBK »

convery wrote: post #10
mattbrewsbytes wrote: post #7

> "To the contrary, Musk forwent all due diligence—giving Twitter twenty-four hours to accept his take-it-or-leave-it offer before he would present it directly to Twitter's stockholders," Twitter wrote. Bot/spam account analysis is just trying to do PR spin. Musk has no legal standing to use any of that to back out of the merger. > The five-day trial is now scheduled to begin on October 17 So we will see about 2 more…

> Musk has no legal standing to use any of that to back out of the merger. The claim is that Twitters SEC filings had bots at ~5%, and Musk made the investment based on that. From the leaked recordings of staff, everyone knew that the percentage was much higher.

Twitter doesn't say that 5% of all accounts are bots. While I don't have inside knowledge of how Twitter does things, the industry standard is to:

1. Build a classification model to determine whether an account is a bot or not.

2. Measure the precision/recall curve of that model, and pick a threshold where the precision would be >95% (i.e. >95% of what the model labels as "humans" are really humans).

(Higher recall leads to lower precision and vice versa. Higher recall means a higher mDAU, so it's a balancing act between a high number versus a high confidence in that number.)

3. Let the model loose on all accounts.

4. The number the model tags as humans is the "mDAU", though While this model takes a lot of training data to build, verifying its continued precision/recall characteristics can be completed with very little ongoing sampling - easily handled by the "100 a day" figure Musk has claimed.

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#463

Post by bentcorner »

Zpalmtree wrote: post #439
bentcorner wrote: post #185

Earlier quoted context omitted.

> This gets my curiosity going, so now I'm paying a lot more attention to conservatives than is normal for me (on Substack, YouTube, and Instagram). This is a phrase right out of this playbook. No thank you. https://medium.com/@DeoTasDevil/the-rhetoric-tricks-traps-an...

I don't get how his comment relates at all to your article. Paying attention to conservatives makes him a white nationalist?

Paying attention to them is fine, I think curiosity is an excellent trait. GP's comment is extremely similar to the "thanks for pushing normal people even further to the right" phrase in my linked article. GP is implying that right-wing talking points are being indirectly validated by twitter, and GP values are being changed because of that.

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#464

Post by dragonwriter »

vba616 wrote: post #461
Banana699 wrote: post #320

Earlier quoted context omitted.

How does any of this change or contradict what I said? And doesn't the US legal system allow for escalation to the Supreme Court, which can override lower courts? So technically whatever the Delware Court is not the final say on whatever happens. >Trial date set to October This is so so awesome, I will get to see Twatter defenders rage and fume for 2 more months? ^__^

An expert commenter on the topic (not me) wrote that there is only one potential step in the appeals process, and that is a "Supreme Court" which is in Delaware, not the famous "US Supreme Court" you may have heard of. First Google hit for "delaware chancery court appeal": "The Supreme Court is the State's appellate court which receives direct appeals from the Court of Chancery, the Superior Court, and the Family Cou…

If there is an issue of federal law raised (which can emerge during the case, e.g., if there is a due process issue with the handling of the case, even if it isn't initially part of the case) state Supreme Court decisions can be appealed directly to the US Supreme Court, though (like all appeals to the Supreme Court) such appeals are heard at the discretion of the US Supreme Court, not by right.

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#465

Post by vba616 »

koheripbal wrote: post #161
rchaud wrote: post #156

Earlier quoted context omitted.

That will be almost impossible to prove. Especially as he provided an all-cash, take-it-or-leave-it offer based on publicly available financial data. He has access to lawyers and financial advisors to warn him against potential misstatemrnts on Twitter's part. If they signed off on it, he has no case. If they warned him against it, and he went ahead anyway, he also has no case. The judge will rightfully wonder how th…

What happens in cases like this is that Musk will request the court for discovery of internal Twitter documents/emails/reports/etc to backup his claim that Twitter was aware that more bots existed in their BAU number than they reported. Twitter will object, but ultimately some degree of discovery will be allowed (with a more narrow scope), but it will pressure Twitter to settle for a lower purchase price because god…

Is it really possible to pursue a claim, if nothing you find in discovery can possibly be relevant to the case?

The experts seem to think that it's virtually impossible a hypothetical misrepresentation rises to the level of a "material adverse event".

It can't even be his motivation, since he is inherently admitting he doesn't have the evidence yet.

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#466

Post by sethaurus »

patrickdavey wrote: post #458

If I bought Twitter shares today ($42) would Musk be forced to also buy my shares at $52 if the sale was forced through?

Yes — and this possibility has been factored by the market into that $42 price. It's among the reasons why the stock isn't trading significantly lower.

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#467

Post by winternett »

justapassenger wrote: post #459
winternett wrote: post #444

Earlier quoted context omitted.

It sounds like a diabolical idea, but why would someone risk that much money on something that is pretty much guaranteed to get tied up in long-term and very public litigation? There's no way to simply back out of a very public offer for a major social app without litigation that costs time out of your life (which is in that case more of a painful thing than losing the money). Paying capital gains (after huge deducti…

Because Musk thinks he’s a god, he can distort the reality around him and do whatever he wants with no consequences. And TBH, he has all the reasons to think so. He’s been openly lying about so many aspects of his businesses for a long time, manipulating markets, ignoring all the rules and regulations and only thing it resulted in, was making him wealthiest man in the world.

I'd never give bad behavior a pass, but I think moral points are superfluous to the discussion, and they only serve to show emotional bias, they don't really serve the right method of evaluating the overall situation.

Expressing character traits and flaws puts a harmful spin on discussions. Think of the situation as if an unknown person did the very same things perhaps?

The actual actions taken and specific words written and recorded are the only things that will be considered most by a court ultimately (if correct process is followed).

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#468

Post by sangnoir »

dahdum wrote: post #195
kergonath wrote: post #62

Earlier quoted context omitted.

From your source: > our estimates suggest that between 9% and 15% of active Twitter accounts are bots From Twitter: “roughly 5% of our monetizable daily users are bots”. Both can be true at the same time, and it does not mean that Twitter lied.

Twitter's lawsuit mentions this, stating only they know who is monetizable. IIRC Musk's lawyers deliberately push the same false equivalency in their filing.

Musks lawyers aren't fans of being disbarred or other legal sanctions, so they carefully couched the language in their filing to say "Musk believes [the intentionally confusing, possibly false narrative about bots]" without providing evidence, which is weak sauce.

Re: Twitter says Musk’s spam analysis used tool that called his own account a bot

#469

Post by Banana699 »

tehwebguy wrote: post #411
Banana699 wrote: post #230

Earlier quoted context omitted.

Quoted post unavailable.

Your other post got nuked but to explain more: It’s okay for some opinions to be suppressed in most places! Some opinions are almost universally considered to be bad, there are some really easy & gross examples that I think most of us can think of. Some are considered bad in a particular context. The ones you mentioned are great examples of reasonable policy: - banning covid misinformation to reduce deaths is good, l…

Contradicting the official narrative on Covid19 isn't automatically 'misinformation', the official narrative held that masks are useless in the first months, then insisted on it religiously the following months, when, in fact, they are indeed mostly useless in the face of latest variants. Both of those convictions were held against evidence, and intelligent and knowledgeable commenters and organizations tried to point out what was wrong in them only to be met by "debonking" and censorship.

Reality is not harassment. If somebody is upset they can't control someone's else words or perception, they can go to therapy about it. This remains true even if you invent a couple of dumb words to describe the words or perceptions that upset you, like 'deadnaming' or 'misgendering'.

There is a difference between deliberate selectivity based on ideology, clearly and nakedly what twatter does, and simple errors and human variability.

Yes, I'm entirely free to think anything I want and disparage any website or any other group\institution of people for not agreeing with me, and being happy that dumb anti-free-speechers are dragged to court and toyed with for months. What I don't understand is how your dumb role-playing of an imagined strawman of me in your first reply says anything about me or my opinions.

Post reply on HN