,
14 tweets,
11 min read
Read on Twitter
Twitter released several additional datasets this morning of accounts that were suspended on suspicion of being state-backed disinformation operations. Let's take a look at the additional 418 Russia-aligned accounts first.
cc: @ZellaQuixote
cc: @ZellaQuixote
There are 927536 total tweets in this dataset, with some activity as far back as 2010 but the majority of the tweets happening between early 2015 and November 6th, 2018 (US midterm election). 
What did these accounts tweet about? The "Trump", "Islam", and "Hillary Clinton" categories are the most prevalent. The #QAnon category has far more activity than in the previously-released IRA dataset (22179 tweets vs 2721).
On that note, #QAnon was one of the popular hashtags among these banned accounts. Also prominent was #ReleaseTheMemo (that memo was kind of snoozer in the end, wasn't it?), #PJNET, and various anti-Islam hashtags.
These accounts joined in on some interesting hashtags early. The first #ReleaseTheMemo tweet from a banned account is timestamped ~3 hours later than the first overall tweet, and their first #FollowTheWhiteRabbit (early #QAnon hashtag) tweet is 16 hours after the first overall.
Here are the media sources they linked. Breitbart and Fox were the most popular; a now-defunct Syria-related website, RT, wikilieaks, Reddit, and a smorgasboard of both genuine and fake news sites comprise the rest of the menagerie.
Were there bots among this group of banned accounts? At least six, which look very similar to one another based on schedule/services used. (If you're wondering why the accounts have names like @BHY5uExg0NHxpc02q5uVt4rX4aDCoozOqNHyVskycg=, the names were anonymized by Twitter.)
These 6 bots (active in 2015) also featured similar content. 99.7% of tweets were RTs; accounts RT'd were mostly a mix of banned Kremlin trolls, other banned accounts, since-deleted news feeds with strikingly similar names to those used by the IRA, and real news sources.
One of the accounts retweeted by the deleted bots, @Chris_1791 with 83K followers, was also popular among the set of banned Internet Research Agency trolls. It received 3410 RTs from 111 accounts, more than all but eight accounts that are still live.
At the present time (and when it was retweeted by the banned Kremlin trolls from 2015-2018), @Chris_1791 is an almost entirely automated account. The current automation tool of choice is dlvr(dot)it, which it uses to spam links to (mostly) right-wing media.
Up until early 2013, @Chris_1791 appears to have been largely human-operated. In the ensuing years, they've used a variety of automation services to create a right-wing news aggregation bot. We're interested in theories as to why the banned Kremlin trolls latched onto this one.
Going a bit further down the rabbithole, @Conservatexian is another largely automated pseudonymous account that received disproportionate attention from the banned Internet Research Agency accounts. @Conservatexian (like @Chris_1791) is largely a RWNJ news aggregation bot.
At the present time, the tweet schedules of @Conservatexian and @Chris_1791 differ. It is notable, however, that at the time the accounts were being amplified by the banned trolls the schedules were very similar (on top of the accounts linking a similar lineup of news sources.)
Was @Conservatexian always automated? To an extent, but until September 2016 a sizable proportion of the traffic was via the Twitter website (a small amount still is). Interestingly, this is the same time that Trump becomes a frequent topic of the account's tweets.
