4 ms·

If we'd stop mixing up private communication with global-scope broadcast "platforms", that'd probably help. Not that it's the users' fault—it's that these plat

If we'd stop mixing up private communication with global-scope broadcast "platforms", that'd probably help. Not that it's the users' fault—it's that these platforms have structured themselves that way.

rglover 4y ago
This. With all of this AI/image recognition stuff, it'd be relatively easy to allow people to maintain a list of filters and when someone posts nudity, profanity, etc., just obscure it in their feed and present a "this is something you asked us to filter out <keyword>, proceed with caution" or just hide it entirely. Put the individual in control, and force them to create a list (if they wish) on next app open with a timestamp of if/when they declined to set something. As for the concern about disinformation, ignore the content and watch the action. If some guy in the hills of Arkansas wants to think "them dems are lizard people" but doesn't show any signs of taking action, who cares (rhetorical)? If they show signs of action, follow the usual path of legal recourse/escalate to the proper authority.
- yamtaddle 4y ago
  > This. With all of this AI/image recognition stuff, it'd be relatively easy to allow people to maintain a list of filters and when someone posts nudity, profanity, etc., just obscure it in their feed and present a "this is something you asked us to filter out <keyword>, proceed with caution" or just hide it entirely. Put the individual in control, and force them to create a list (if they wish) on next app open with a timestamp of if/when they declined to set something. Or even without AI tools. All this stuff is a complete non-issue on my extended-friend-group WhatsApps and my family text message threads. It was never a problem when I ran a private forum for people I knew, years and years ago. Mixing private communication with people you know with a global broadcast system seems to be what causes the problem to exist in the first place. Ordinary actual-humans-you-know communication channels don't need some 3rd-party censor, and for the most part don't even call for end-user-tunable censorship tools, beyond being able to leave & form groups or maybe set images to click/tap-to-view or something. Nothing sophisticated, certainly.

rglover 4y ago

This. With all of this AI/image recognition stuff, it'd be relatively easy to allow people to maintain a list of filters and when someone posts nudity, profanity, etc., just obscure it in their feed and present a "this is something you asked us to filter out <keyword>, proceed with caution" or just hide it entirely. Put the individual in control, and force them to create a list (if they wish) on next app open with a timestamp of if/when they declined to set something. As for the concern about disinformation, ignore the content and watch the action. If some guy in the hills of Arkansas wants to think "them dems are lizard people" but doesn't show any signs of taking action, who cares (rhetorical)? If they show signs of action, follow the usual path of legal recourse/escalate to the proper authority.

yamtaddle 4y ago
> This. With all of this AI/image recognition stuff, it'd be relatively easy to allow people to maintain a list of filters and when someone posts nudity, profanity, etc., just obscure it in their feed and present a "this is something you asked us to filter out <keyword>, proceed with caution" or just hide it entirely. Put the individual in control, and force them to create a list (if they wish) on next app open with a timestamp of if/when they declined to set something. Or even without AI tools. All this stuff is a complete non-issue on my extended-friend-group WhatsApps and my family text message threads. It was never a problem when I ran a private forum for people I knew, years and years ago. Mixing private communication with people you know with a global broadcast system seems to be what causes the problem to exist in the first place. Ordinary actual-humans-you-know communication channels don't need some 3rd-party censor, and for the most part don't even call for end-user-tunable censorship tools, beyond being able to leave & form groups or maybe set images to click/tap-to-view or something. Nothing sophisticated, certainly.