The 2020 study published in Computers in Human Behavior analyzed the top 100 subreddits — the most influential communities on the entire platform. Their finding? 15% of these subreddits contained content likely posted by bots or corporate trolls specifically designed to promote companies or organizations.
“15% of all subreddits contained corporate bot content” is very different than “15% of all content”
This also doesn’t really give a whole picture. How much of this content actually trends? There’s always some corporate sludge at 0 points if you sort by New, which is how the site is supposed to work. And even some stuff that gets brute forced through is “Hey fellow kids” level obvious and gets trolled or removed by mods.
And while Reddit right now is a soulless husk, all of these things need to be studied on Lemmy as well. Right now there’s probably not much because we’re just not a big target, but as it grows it’s certain that corporate shills and propaganda farms will start to target us. Like I’m not opposed to the general idea of the research, but it really needs to be more specific and in-depth to be helpful
I’m not saying there aren’t subtle ones that get through, I’m just saying that there’s nothing in the report distinguishing between the obvious shill and a well disguised advert. Granted, if an advert is disguised well enough, it’s hard to track in a study like this, but there are plenty that aren’t obvious until you start poking around posting history or metadata
I still remember the time of the great Miele shilling by the vacuum repair guy. Every thread about vacuums had the same conversation about how only this single company is the best company of all. It worked, my brother bought one of those vacuums.
“15% of all subreddits contained corporate bot content” is very different than “15% of all content”
This also doesn’t really give a whole picture. How much of this content actually trends? There’s always some corporate sludge at 0 points if you sort by New, which is how the site is supposed to work. And even some stuff that gets brute forced through is “Hey fellow kids” level obvious and gets trolled or removed by mods.
And while Reddit right now is a soulless husk, all of these things need to be studied on Lemmy as well. Right now there’s probably not much because we’re just not a big target, but as it grows it’s certain that corporate shills and propaganda farms will start to target us. Like I’m not opposed to the general idea of the research, but it really needs to be more specific and in-depth to be helpful
Or you’re experiencing the toupee phenomenon. you only notice the obvious ones.
Note their methodology for this study, afaict, also would entirely miss subtle stuff.
Either the point about frequency is valid, or this is a weak headline, no?
I’m not saying there aren’t subtle ones that get through, I’m just saying that there’s nothing in the report distinguishing between the obvious shill and a well disguised advert. Granted, if an advert is disguised well enough, it’s hard to track in a study like this, but there are plenty that aren’t obvious until you start poking around posting history or metadata
Not 15% of all subreddits, 15% of the top 100 that they studied.
But they don’t link the study, or give any way to find it. I’m very suspicious now.
I still remember the time of the great Miele shilling by the vacuum repair guy. Every thread about vacuums had the same conversation about how only this single company is the best company of all. It worked, my brother bought one of those vacuums.
Apparently the first link to a ‘pew study’ is wrong (it goes to pew, but doesn’t mention reddit much). See here