Live data from Hacker News

Massive networks of fake accounts found on Twitter

bbc.com

251–260 of 267 posts

Re: Massive networks of fake accounts found on Twitter

#251
post #194
post #163

Earlier quoted context omitted.

Try to remember the last time you heard anything positive about Russia. Consider how it is compatible with the existence of Putin's troll army. At the very least try to imagine the relative scale of money/power involved in the propaganda in the West compared to the rest of the world. (for reference, check military spendings)

>Try to remember the last time you heard anything positive about Russia. Consider how it is compatible with the existence of Putin's troll army. This completely misses the point of what the troll army is there to do. It isn't aiming at convincing people of a point of view, but confusing the audience. You should read this story from 2015: http://www.bbc.com/news/world-europe-31962644

Also one of the threads in the latest Adam Curtis film [1].

[1] https://en.wikipedia.org/wiki/HyperNormalisation

Re: Massive networks of fake accounts found on Twitter

#252
post #95

Earlier quoted context omitted.

Why ?

It's not possible to be completely sure that 5% isn't a reality-based estimate, but it's typical for services like Twitter Audit and BotOrNot to identify half or more of a celebrity's followers as bots. For example, as I type this BotOrNot estimates that 34% of @RealDonaldTrump followers, and 48% of @BarackObama followers, are real. Just from being a nobody on Twitter who uses it almost every day, I'd also agree that…

What makes you think there is no overlap between these bots ?

Re: Massive networks of fake accounts found on Twitter

#253
post #249

Earlier quoted context omitted.

They don't "permit" this. I used to work on anti-bot technology at Google and we put a lot of effort into fighting botting, in particular on Gmail. We were quite successful at the time. Bot controlled accounts went from being a problem that was threatening the viability of the service to being fairly trivial. I don't know what it's like these days. Anti-abuse teams tend to be relatively small. We aren't talking hundr…

> From a quick check it seems Twitter is getting better at it. Prices used to be more like in the low $20/k range. Now the low end accounts are $45/k and if you want PVAd/profiled it's up near $90. Compare to Gmail where the price is more like $280/k. Or Facebook EN accounts, $120. Good to see my old colleagues doing such a good job! Perhaps there's just more demand? > Nowadays there's suddenly a bout of anti-Russian…

My experience was that prices never really changed unless the difficulty of creating/keeping accounts went up. Demand was pretty stable over time, only supply changed.

And yes, of course my experience of what bots were used for left me "biased" as to what they're used for. How is actual experience bias?

Re: Massive networks of fake accounts found on Twitter

#254

Earlier quoted context omitted.

There is blame enough for both twitter and the asinine people here who engage in the deliberately misleading communication, fraud and harassment. Capability is not a moral license to hurt others. If it were, you'd be morally responsible to kill as many people as possible. My third paragraph is saying this: The US elections were decided by and large by social media plays. There are hints of illegal use by campaign act…

Ever since email spam, we have learned that systems have to be made resistant to abuse. There is enough blame to go around, but it is reasonable to blame Twitter. It is not reasonable to expect bad actors to go away. Systems designed on this assumption will always fail. Yes, we can rightly rail against the people abusing the system, but ultimately there are 7 billion people and expecting all of them to be ethical is…

> Yes, we can rightly rail against the people abusing the system, but ultimately there are 7 billion people and expecting all of them to be ethical is foolish.

I don't really get this. You seem to agree with me, but then also seem to be chastising me for blaming individuals here proudly abusing the system?

> By all means, keep fighting the good fight, but the only thing that can realistically improve the situation is Twitter.

The HN mods could grab the IP history of the fraudster user and pass them to twitter. That'd make the world incrementally better for a small period of time. Who knows, maybe Twitter could bring a lawsuit to bear. Technically that dude is describing a felony.

Re: Massive networks of fake accounts found on Twitter

#255

Earlier quoted context omitted.

Ever since email spam, we have learned that systems have to be made resistant to abuse. There is enough blame to go around, but it is reasonable to blame Twitter. It is not reasonable to expect bad actors to go away. Systems designed on this assumption will always fail. Yes, we can rightly rail against the people abusing the system, but ultimately there are 7 billion people and expecting all of them to be ethical is…

> Yes, we can rightly rail against the people abusing the system, but ultimately there are 7 billion people and expecting all of them to be ethical is foolish. I don't really get this. You seem to agree with me, but then also seem to be chastising me for blaming individuals here proudly abusing the system? > By all means, keep fighting the good fight, but the only thing that can realistically improve the situation is…

Maybe I can explain better. There are two concerns. One is, what is ethical and right for a person to do? On this we agree. Building botnets to spam Twitter is not ethical and people shouldn't do it.

The other concern is, what should we do about it? This is where I "chastised" you, because focusing on the ethical behavior of members of the public is generally useless. If the people commenting here didn't do it, someone else would. That doesn't excuse them, but it does mean going after them isn't an effective strategy.

Imagine trying to fight email spam by shaming individual spammers. Many people have done it. It doesn't work. If those people had redirected their energy towards fixing the system, rather than the people, the spam problem could have been solved decades ago.

> The HN mods could grab the IP history of the fraudster user and pass them to twitter.

Besides setting a bad precedent for HN, this would be useless. Twitter already has access to much more information than the BBC did when they did their investigation. Twitter simply doesn't care to act on it. If it was a priority for Twitter, they would fix it, certainly not by bringing a lawsuit, but by applying a technical solution across the board.

By blaming the bad actors, you take heat off of Twitter, and encourage people to waste their efforts doing things that will never make a difference.

Re: Massive networks of fake accounts found on Twitter

#256

Earlier quoted context omitted.

> Yes, we can rightly rail against the people abusing the system, but ultimately there are 7 billion people and expecting all of them to be ethical is foolish. I don't really get this. You seem to agree with me, but then also seem to be chastising me for blaming individuals here proudly abusing the system? > By all means, keep fighting the good fight, but the only thing that can realistically improve the situation is…

Maybe I can explain better. There are two concerns. One is, what is ethical and right for a person to do? On this we agree. Building botnets to spam Twitter is not ethical and people shouldn't do it. The other concern is, what should we do about it? This is where I "chastised" you, because focusing on the ethical behavior of members of the public is generally useless. If the people commenting here didn't do it, someo…

> because focusing on the ethical behavior of members of the public is generally useless. If the people commenting here didn't do it, someone else would. That doesn't excuse them, but it does mean going after them isn't an effective strategy.

But... we're talking to the people that did. It is no hypothetical, no question. Past tense. Done. Did.

I'm not expecting this to axiomatically solve all twitter abuse. I do hope it will spark a conversation about how absolutely ruthlessly amoral many people here are. I hope young people looking for funding for a startup will think twice when they see what kind of environment HN creates. Do you really want to plant your flag and take money from people who enable fraudsters from the very ecosystem you plan to engage with?

I point out people like that all the time in many venues. Because I think that raising awareness of the utter amorality of the software industry means more amoral assholes will be penalized for their actions. Perhaps this is naïve, but it is a conceit I will not easily give up.

> Besides setting a bad precedent for HN,

Yeah the bad precedent of, "Don't boast about how you're destroying the ecosystem we're trying to build a living on." We wouldn't want to disenfranchise the wretched thieving toe-rags who think making political twitterbomb networks, now would we?

> Twitter already has access to much more information than the BBC did when they did their investigation. Twitter simply doesn't care to act on it. If it was a priority for Twitter, they would fix it, certainly not by bringing a lawsuit, but by applying a technical solution across the board.

It's curious how the instant I mention sending data to twitter for enforcement suddenly that, too, is useless. We should focus on making twitter do something but since they don't care we shouldn't bother. Lower our arms from supplication and sink into the mud, I suppose?

> By blaming the bad actors, you take heat off of Twitter,

No see, I have enough incandescent bile that I can do both. But you seem to want to do neither.

Re: Massive networks of fake accounts found on Twitter

#258
post #257

Earlier quoted context omitted.

Sure, I believe you. /s

It's not like my ties are not public. Check my profile.

Check my profile idiot.

Regardless of how infuriated you may be by the behavior of other users, please don't resort to name-calling.

Re: Massive networks of fake accounts found on Twitter

#259
post #111
post #75

Earlier quoted context omitted.

Almost like one of those RSA/Google Authenticator/etc rotating numbers... always 6 characters, seemingly random, somewhat pronounceable. Weird.

As it is twitted exactly every 2 minutes it can actually be a kind of current confirmation or authentication code.

Dead man's switch?

Re: Massive networks of fake accounts found on Twitter

#260

Earlier quoted context omitted.

Maybe I can explain better. There are two concerns. One is, what is ethical and right for a person to do? On this we agree. Building botnets to spam Twitter is not ethical and people shouldn't do it. The other concern is, what should we do about it? This is where I "chastised" you, because focusing on the ethical behavior of members of the public is generally useless. If the people commenting here didn't do it, someo…

> because focusing on the ethical behavior of members of the public is generally useless. If the people commenting here didn't do it, someone else would. That doesn't excuse them, but it does mean going after them isn't an effective strategy. But... we're talking to the people that did. It is no hypothetical, no question. Past tense. Done. Did. I'm not expecting this to axiomatically solve all twitter abuse. I do hop…

I agree Twitter should be taken to task, but it seems naive to think Twitter cares and is unable to solve the problem when so many indications point to spam and bots being a massive part of their traffic and them looking the other way.

Around the election, I noticed on every Trump tweet the second or third reply would be someone selling a "liberal tears" mug. They had a set sequence of posts and replies leading up to that one with the link and they hit every single one of Trump's posts like clockwork. This went on for weeks. We are talking about blatant and totally unambiguous commercial spam on one of the highest-profile Twitter accounts. If they don't even deal with the trivially obvious, what are they going to do about the sophisticated attacks which just so happen to inflate the numbers that affect their revenue?

The precedent I am talking about would be HN moderators getting involved in detective work and in reporting people, a huge and unrewarding task, in this case, for a company that perfectly obviously doesn't care, and is just going to drop the information on the floor. And what are they going to do with one IP address, anyway? I understand your outrage, but it just isn't practical.

If you want to make Twitter do something, outrage directed at their top management, where it belongs, may have an effect, if it gets picked up by the press and starts to affect their share price or advertisers. Even that is a long shot.

Honestly, if you really care, I hope you can find some people building something better and collaborate with them, rather than trying to make Twitter better, which I think is unlikely to ever work.

Post reply on HN