I think it is a LOT simpler than what you are suspecting. As someone in the comments has pointed out, mass reporting on certain types of articles or certain sources trigger automated review/removal of that content.
Normally, consensus reporting is a good sign that something is undesirable. In this case it has clearly been weaponized. This is a TUNING problem and it seems like social media companies have to find a way to balance on this
>This is a TUNING problem and it seems like social media companies have to find a way to balance on this
OK -- to play nice, let's say it's a tuning problem.
Why didn't this 'tuning problem' get fixed the first few times politically motivated messages got 'disappeared'?
At what point does the responsibility lie on those massaging the tuning methods?
my personal view is that social media is heavily gamed by the three-letter-agencies and all of their equivalents abroad. The existence of internal 'squads' within most of these groups that focus on social media networks reinforces that personal view -- if it wasn't an important battle ground then why form squadrons to seize and control it?
Why do we assume that it can be fixed? Even if there were no time constraint at all, and you could abolish all moderation tools to spend an hour carefully analyzing each individual post, I suspect you'd produce a lot of valid moderation complaints from people with different but reasonable assumptions. I don't think there exists any good strategy for moderating such controversial issues.
They could try to detect incipient political situations, similar to how they detect trending tweets, and then treat reports related to those tweets with greater than usual scrutiny.
I don't think anyone expects Twitter or other social media companies to get this perfect. The nature of the problem means there will always be edge cases, debatable calls, and algorithms automatically doing unfortunate things. I do think there is a reasonable expectation for these companies to both do more and make progress.
This. Facebook is a small startup and don't have moderators to check the validity of the report bombing at all. It's totally a glitchy algorithm. They're totally not waiting to use the «glitch» argument to counter it the moment there is a public outcry.
Normally, consensus reporting is a good sign that something is undesirable. In this case it has clearly been weaponized. This is a TUNING problem and it seems like social media companies have to find a way to balance on this