Hacker News new | past | comments | ask | show | jobs | submit login
Twitter Bots Use Likes, RTs for Intimidation (krebsonsecurity.com)
125 points by willvarfar on Sept 4, 2017 | hide | past | favorite | 39 comments



I've pared down my twitter usage and who I follow, and increased the use of words to mute. I regularly delete old tweets. In a way, the golden days of Twitter occured when we all were on it, the early adopters. It appears to me that the rapid growth of the platform has not been accompanied by a rapid growth in, well, growth management. Spammers, politics, ideological motivated manual censorship / safety council, bots, spam, flame wars etc, It used to be about broadcasting your current status - the status like what you would update on your instant messaging client. The chase to get as many people on it as possible leaves huge holes for bots.


I made my first Twitter account shortly after it launched, so there's a good chance I'm an earlier adopter than most people who complain about how it's changed.

Many would, unfortunately, read any of my LGBTQ friends (for example) venting about the harassment they get for who they are as "political." What's political/ideological/junk/whatever to one person is a central and valuable function of the platform to others. This is their "current status," which they broadcast, and it's unavoidably political because certain groups have decided to politicize innocent and harmless attributes.

I would not have met my best friend, who I can talk to about this stuff when Twitter seems inappropriate, without Twitter. For you, it's a place to chill and chat. For me and many like me, it's a lifeline in hostile territory.


Twitter was never a place "to chill and chat" to me. I was an early adopter as well and used it as a more personal RSS reader. I followed mostly software developers and tech outlets. I honestly enjoyed the mix of 140-character tech discussion and the occasional insights into other devs' work and personal life. At some point however, my personal feed was full of drama, politics, gender, bullshit, nonsense, … It was worse than my Facebook feed. and the platform gave me 0 control over what I see from whom. The best I could do was to follow, hope for the best, unfollow, rinse and repeat. I quit a long time ago, and I don't look back.


When did Twitter turn into hostile territory? Or are you saying that the world outside of Twitter is hostile and Twitter is some way of dealing with that in the present day?

I remember when people started being harassed on Twitter, and it wasn't close to the launch, rather it was when it was sometime after it became massively popular. If we can try to remember we will see that it was different back then when we were early adopters.


Outside world. You should get to know your LGBTQ and other marginalized friends better if you aren't aware of how bad things are. Twitter may have been the first place they found anyone else like them. Even accepting environments (rare) can be difficult to cope with if you have no one to talk to who really understands what it's like.


> Spammers, [...], spam,

Could easily be addressed by Twitter but for some reason, much like their ignoring of harassment, they choose not to.


What's the best metric Twitter can use to say they're doing well? It's number of users. So fake users improve their rankings, and their "performance".

Today I posted a tweet, I received 12 "likes". 9 were from profiles created this month, with no profile details, no picture, and content which was 100% retweets. Fake users.


Yeah, and it's the same for Reddit. They don't care for bots reposting content they stole from legitimate users, because it makes the site look good.


And when you consider the other best metric for the company is activity/tweets per second/minute/hour/day, you can probably figure out the other half of why they don't crack down trolling, bullying or harassment in general. Being an overly confrontational jerk gets people replying to you, which makes the stats look good.


why does twitter have to be the internet police and crack down on everyone saying something "naughty"? how about, if you feel like you're being harassed, you block the person who's bothering you?


I think it's partly a problem of scale.

When you become a target of abuse - or mass criticism or what have you - you're dealing with hundreds or thousands or higher orders of tweets to deal with.

Often more than enough to drown out the positive, because detractors are so often more vocal than supporters.

That can be very flustering on a site with a small linear feed. Individually blocking people doesn't scale.

Block lists kind of help but they're a brute force solution and we can all see the problems and edge cases from that solution.

It's cynical, but perhaps there is a market opportunity there.

It just seems strange that Twitter hasn't come up with something significant in that vein.


Thats the intention of bullying to silence your voice ...


If there's hundreds or thousands of tweets that bother you then just log off. In real life, if there are thousands of people mad at you, do you get to send police to their door to silence them? No, you just ignore them and move on. I mean if they're threatening personal harm that's different. But if they're just calling you names than who cares? Didn't we learn how to deal with this in high school?


> But if they're just calling you names than who cares?

People are very social and, for better or worse, social media is a direct extension of the social lives of many people. "Turning it off" can be the same as cutting out an important part of their life.


In high school, persistent and large-scale bullying is generally dealt with by the school before, as you suggested, the student on the receiving end takes the much more drastic step of moving schools.


You mentioned regularly deleting old tweets, so I'll share a useful tool I built:

For deleting and archiving both posted and liked tweets: https://github.com/Datamine/Archive-Tweets

The archive feature downloads the JSON for a tweet, including any images.


Odd I posted this comment 3 days ago, but it says 2 hours... and the thread is back on the front page.


It says 3 days on your profile, but the link and discussion are new. Might be a bug.


Discussions are sometimes re-bumped by the mods if they feel it is an interesting topic and didn't get enough discussion the first time around. dang describes it here: https://news.ycombinator.com/item?id=11662380


Thank you, learnt something new today!


There's a service called TweetDelete that will automate this. Given the nature of Twitter as a ephemeral communication platform, there's very little value to having old tweets anyways.

I have mine set to purge anything older than a month.


Easy answer: Stop using Twitter. Stop caring about Twitter. We have the power to kill this nonsense in two seconds -- we only need the will to do so and if we don't we're to blame.

I stopped using Twitter and I'm happier.


First Facebook, then LinkedIn, and recently Twitter are gone from my life, and good riddance! Now if anyone wants to figure out what's happening in my life they know how to reach me, and we can both have the pleasure of actual social interaction.


I unfollowed a bunch of accounts that were too political, disabled retweets on a couple others, muted a couple words (trump is a good thing to mute), found some interesting people to follow and now Twitter consistently brings me interesting links and ideas. Oh and the occasional pun.


It says Twitter doesn't allow automated likes, which is what bots usually do.

https://support.twitter.com/articles/76915

D. Automated actions you take on Tweets or accounts

Automated likes: You may not like Tweets in an automated manner.


How difficult is it to detect Twitter bots? I would have thought there'd be lots of obvious signals (who the bot is friends with, what they post, how often they post, how old their account is).


It must be relatively easy to find bots, but I imagine Twitter has a decent size ecosystem of neutral / positive bots for various QoL tasks that would be collateral damage as intent is much harder to determine.


It feels as though Twitter has a vested interest in making its network seem as active as possible, irrespective of quality of interaction. In a way, Twitter's userbase appears like a captive audience - they've no desire to leave, no matter how bad it gets. Oddly masochistic in a way.


You could argue that twitter bots don't do much damage. If you don't want to follow a bot, just don't follow them. They should be easily recognizable.

People who did the auto-follow thing were screwed though.


Bots do damage search and hashtags and other discovery mechanisms beyond just your following list.


No, that's not how Twitter bots work. They damage the overall ecosystem in both active and passive ways. Of course you can choose to not follow someone/something, but they can still harm your experience as a user.


Someone should create a pattern catalog for bots to enable easier discussion about them.


Odd place for me to ask this but does anyone know what's with the ring of twitter user wth the usernames ending in gives,gibs, givs as in johnGivs, jasonGives as an example. They all say they will release account "methods" for Spotify, Apple, PlayStation , steam etc for retweets. I came across the pattern a few weekends ago noticing they say to each other they keep their accounts on ghost script and they were all mad it was down one night. Is this just a ring of stolen accounts or stolen account cards?


So...someone could write their own twitter bots that trigger these russian troll bots with their trigger keywords. Then the russian troll bots will waste their time trolling the wrong accounts and make themselves look silly at the same time.


I made my Twitter account in 2007. Those were the days. Now, the platform is the mostly used for propaganda, perception building, marketing and hate mongering.

Also, there are no friends or foes on Twitter, only interests.


What a coincidence, I just saw one today in a highly political thread. Just goes to show I should stay away from political stuff on twitter.


Where's the evidence that these are bots?


Regardless of motives, I think it's blindingly obvious that getting 500 retweets within nine minutes for "wonder whether the use of the @DFRLab handle and talking about a bot attack is enough to trigger one" involves someone or some algorithm triggering a Twitter botnet.

It's not like the original tweet was so entertaining that everybody seeing it would want to share or the original tweeter had more than a few hundred followers





Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: