Use each of these three services every time you want to check if a Twitter user is a troll or a bot.
Read the instructions, it’s less complicated than building a Kinder egg toy from the 80s.
Check with Botometer, made by university types with big noggins.
If the account you are checking scores more than 2 (it’s in the yellow, orange, or red) it’s probably not your close friend “Roger off of the interwebs” and fails the test.
Check it again with Bot Or Not, made by a data nerd who loves probability, probably.
If the account you are checking scores more than 0.6 the person behind it probably doesn’t drink in your local and fails the test.
Check it again with Bot Sentinel, made in America (but actually great rather than great again.)
If the account you are checking scores more than 25% (that’s 26% and over) you probably can’t buy British made goods from it direct.
This is challenging, because you will need to count between one and three, so we’ve made a handy rhyme*:
If it failed just one, it’s all good fun.
If it failed two, it’s talking poo.
If it failed three…it’s a troll, you get me?
Seriously, if an account fails one check it might just be an idiot, which is likely given Britain. If it fails two checks it’s almost definitely deliberately stirring and pushing disinformation. If it fails three, it’s probably based a lot farther away than Dagenham, if you catch our drift.
*We never said the rhyme was any good.
Bot – a fully automated account
Cyborg – a partly automated account which relies on human input part of the time and scheduled activity the rest of the time
Troll – a human account operating to tasking and reporting back on progress
A benign bot can be described as one which is “irritating or revolting but not aimed at influencing public discourse or voter behaviour”
A malign bot can be described as one which is “aimed at influencing public discourse by amplifying false, inaccurate, or dangerous information, and shows signs of tasking to these ends”