8 ms·

> Not only has the filter now been removed, but the new entrants have been encouraged (through algorithms which promote divisive content) to produce and consume

> Not only has the filter now been removed, but the new entrants have been encouraged (through algorithms which promote divisive content) to produce and consume harmful content because it increases "engagement". I think this narrative has been blown out of proportion. The “algorithm” has become an easy stand-in to blame nebulous programmers for people choosing to view content they want to see. The “algorithm” isn’t literally a model trained to recognize divisive content and promote it to as many people as possible. It doesn’t have motives or intentions. It’s literally just a recommendation engine that suggests similar content to people. Platforms without algorithms are still full of divisive content which spreads virally. Some of the worst content on the internet won’t be found on Facebook or other moderated platforms at all. You have to work to get to it, and people do the work to get there. No algorithms to blame, just people. We really need to stop giving people a pass and stop piling our complaints on to the “algorithm”. People who make bad choices and consume bad content are doing so on their own volition, and we need to treat it as such if you want to make progress on the solution. Railing against faceless personifications of algorithms isn’t going to get us anywhere on solving these social problems.

Nextgrid 5y ago
> The “algorithm” isn’t literally a model trained to recognize divisive content and promote it to as many people as possible. It doesn’t have motives or intentions. It’s literally just a recommendation engine that suggests similar content to people. Mostly agreed, though I don't believe "similarity" is the only factor - "engagement" is also a factor, likely a big one considering the business model of the company. Yes, the algorithm doesn't have intentions or political bias (yet?), however it doesn't take a genius to infer that divisive content will generate more engagement than other, mundane content, and even more so when the algorithm has been deployed to production and there's real-world data to prove this. > Platforms without algorithms are still full of divisive content which spreads virally. Without algorithms the only way this content would spread is for people to explicitly re-share it to their friends - it will not pop up "organically" otherwise. You will need to explicitly attach your name to it if you were to share it, and be ready to defend it and bear the risk of being ostracized from your group. This also means that as long as you choose your Facebook "friends" carefully, you can curate your feed and keep craziness or other irrelevant content at bay. > Some of the worst content on the internet won’t be found on Facebook or other moderated platforms at all. You have to work to get to it, and people do the work to get there. No algorithms to blame, just people. And thankfully that worst content has (yet?) to make it to my non-technical friends' Facebook feeds, so it seems like the system is working as designed? Twisted people will keep finding & producing said content, but it's contained and will not make its way onto mainstream platforms without explicit efforts from those people and everyone else on the network to re-share it multiple times. > People who make bad choices and consume bad content are doing so on their own volition Disagreed. People went onto Facebook (or other social media) to keep in touch with their friends & family, but over time, the algorithm was locking them into an echo chamber full of whatever divisive, harmful, false or outright crazy content it could find as long as it was driving "engagement" numbers up. If you could travel back in time and ask some of the people that stormed the Capitol or are forever lost in the QAnon conspiracy whether they thought they would do so when they initially joined Facebook, I bet they would call you crazy. In fact, I bet a large chunk of those people are old enough to know the "old" Internet and yet didn't turn to extremism until the craziness was delivered to them in a nice, harmless-looking format.

Nextgrid 5y ago

> The “algorithm” isn’t literally a model trained to recognize divisive content and promote it to as many people as possible. It doesn’t have motives or intentions. It’s literally just a recommendation engine that suggests similar content to people. Mostly agreed, though I don't believe "similarity" is the only factor - "engagement" is also a factor, likely a big one considering the business model of the company. Yes, the algorithm doesn't have intentions or political bias (yet?), however it doesn't take a genius to infer that divisive content will generate more engagement than other, mundane content, and even more so when the algorithm has been deployed to production and there's real-world data to prove this. > Platforms without algorithms are still full of divisive content which spreads virally. Without algorithms the only way this content would spread is for people to explicitly re-share it to their friends - it will not pop up "organically" otherwise. You will need to explicitly attach your name to it if you were to share it, and be ready to defend it and bear the risk of being ostracized from your group. This also means that as long as you choose your Facebook "friends" carefully, you can curate your feed and keep craziness or other irrelevant content at bay. > Some of the worst content on the internet won’t be found on Facebook or other moderated platforms at all. You have to work to get to it, and people do the work to get there. No algorithms to blame, just people. And thankfully that worst content has (yet?) to make it to my non-technical friends' Facebook feeds, so it seems like the system is working as designed? Twisted people will keep finding & producing said content, but it's contained and will not make its way onto mainstream platforms without explicit efforts from those people and everyone else on the network to re-share it multiple times. > People who make bad choices and consume bad content are doing so on their own volition Disagreed. People went onto Facebook (or other social media) to keep in touch with their friends & family, but over time, the algorithm was locking them into an echo chamber full of whatever divisive, harmful, false or outright crazy content it could find as long as it was driving "engagement" numbers up. If you could travel back in time and ask some of the people that stormed the Capitol or are forever lost in the QAnon conspiracy whether they thought they would do so when they initially joined Facebook, I bet they would call you crazy. In fact, I bet a large chunk of those people are old enough to know the "old" Internet and yet didn't turn to extremism until the craziness was delivered to them in a nice, harmless-looking format.