6 ms·
>That's a tradeoff you can choose to make, but you need to enter into it with open eyes. No it's not. It's certainly not my choice. No one asked me if it's oka
by freejazz 6mo ago
>That's a tradeoff you can choose to make, but you need to enter into it with open eyes.
No it's not. It's certainly not my choice. No one asked me if it's okay for Facebook to distribute CSAM because you insist it would be worse if it didn't.
- scarmig 6mo agoI don't really care if you classify it as a choice or not. One set of actions results in more CSAM than others. Just because you don't like the implication of there being tradeoffs doesn't mean there aren't tradeoffs.
- freejazz 6mo agoYou classified it as a choice, not me.
- scarmig 6mo ago> or not
- freejazz 6mo agoSo you don't care that you're wrong? Not a surprise coming from someone handwaving away the mess Meta made.
- lostmsu 6mo agoWhat your opponent is saying is, "there are mutually exclusive A and B". A being widespread CSAM and B being somebody need to look at CSAM to remove it. Can you elaborate on what exactly is wrong there? Do you see the third alternative C and it's not the "whole choice"? Or are you saying A or B do not exist and therefore there's no choice? Please name C, or tell us why A or B don't exist (or aren't acceptable), or explain your view that doesn't fit into these options.
- ImPostingOnHN 6mo agoSome people are not okay with actively facilitating harm to people, even if inaction results in harm to other people. See: the trolley problem. This is totally okay, but the point made above is that >That's a tradeoff you can choose to make is not correct: It is a tradeoff that one specific person can choose to make, but not one that I or we can choose to make, because we don't control facebook. Mark Zuckerberg controls facebook. He alone can choose to make that tradeoff, or not, on behalf of society.
- lostmsu 6mo agoHe can't do it alone because Facebook is under US jurisdiction.
- ImPostingOnHN 6mo agoYes, facebook is under US jurisdiction. Yes, he alone can make the choice. Not you or I.
- bpt3 6mo agoIn what regard is it incorrect that a single, larger entity that is at least notionally committed to avoiding the existence of any specific type of content on their platform is more likely to successfully avoid the existence of that type of content on their platform than smaller entities with less resources? Now consider that some of those smaller entities might not be even notionally interested in avoiding the existence of that specific type of content on their platform, and are small enough for regulators to be unaware of its existence?
- freejazz 6mo agoTrying so hard not to say child porn.
- bpt3 6mo agoThe problem generalizes, so there's no need to focus on this specific type of content. Care to address the actual issue?
- freejazz 6mo agoBut its not a general problem, it's a problem specifically about child porn that we are discussing. The idea that there is no point in discussing the child porn problem on Facebook is exactly what I'm disputing.
- bpt3 6mo agoContent moderation is a general problem, whether you're talking about child porn, content intended for mature audiences, or memes about Winnie the Pooh. What I and others are trying to tell you is that your obsessive focus on Facebook as if they are the root cause of the problem is incorrect. There is no magic solution I'm aware of because each of them have some sort of tradeoff. The most extreme version of content moderation I can think of is that a human being examines and approves every single message of any kind before it is published, any image of a minor is banned because it's too hard to objectively define child porn (that still leaves the open question of how to determine if someone is a minor visually), and no accounts for anyone under the age of legal majority are allowed, as verified by a legal ID that is checked by a human being. Even in that case, kids will find some way to get an account or just use their parent's account, and the door is cracked open again. And the pedophiles will just go elsewhere, probably using a service with significantly less resources available to attack the problem, which is probably worse than the status quo. This doesn't even touch on the privacy concerns that most people would have with every message being reviewed. As I said before, I would welcome you to share the solution that you imply exists which addresses every issue above.