3M Russian troll tweets
github.com
github.com
Edit: It appears they have informed some users: https://twitter.com/jmorse_/status/954556158047543296
Edit2: Here's a chart of all timestamps by hour+minute: https://imgur.com/a/xiNuagy
Edit3: Here's a list of all accounts tab separated with number of tweets: https://pastebin.com/raw/Em4NeHN8
You probably won't be able to view it on github.com because the file is 700+ MB.
curl https://media.githubusercontent.com/media/fivethirtyeight/russian-troll-tweets/master/IRAhandle_tweets.csv > tweets.csvIs this to spread divisiveness? For example, all the 10_GOP tweets are, of course, very pro-trump:
> cute 10 yo girl had a Trump dress on. Hannity asked her why she liked him. She said because he's doing great things for the country!
Whereas 4MYSQUAD, seems mostly Black Lives Matter related tweets, some are strongly against Trump/the right:
> Do conservatives give a fuck about #policebrutality?
USFreedomArmy is far-right:
> No wonder liberals cannot think logically. The U.S. FREEDOM ARMY awaits your enlistment
And user AAnitRacist is left again:
> Megyn Kelly tears into Trump over attacks on Trump U judge's ethnicity
----
EDIT: I see now the csv has RightTroll & LeftTroll as column values, which is nifty. The accompanying article states:
> Left Trolls often adopt the personae of Black Lives Matter activists, typically expressing support for Bernie Sanders and derision for Hillary Clinton, along with “clearly trying to divide the Democratic Party and lower voter turnout.”
Naturally, then, the best way to prevent that from happening is simply to sow discord. Get the people angry at each other. Hence, you see Russia playing all sides.
How? We can't actually invade Russia. There also doesn't seem to be much interest from anyone in heavily sanctioning Russia.
Which is super convenient for the Russians, because previous sanctions have in fact seemed to hurt them, and the Magnitsky Act also seemed to have an effect
People get very zoned in on very small and rare occurrences. I see this all the time in software support; a customer will report some bug, and claim that it is happening constantly; you dig a little more, and they continue to insist that it is happening all the time; when you get to the bedrock facts in logs and traces, you find that it happened twice out of tens of thousands of interactions over the last few months.