4 ms·

How are these bad actors coordinating hashtags in a way that can't be detected? If they really have some sort of secret channel that no one else can infiltrate

How are these bad actors coordinating hashtags in a way that can't be detected? If they really have some sort of secret channel that no one else can infiltrate, why not just use that to distribute stuff directly instead of running an evil cabal of hashtags? Surely, even given the evil cabal of hashtags, you could find a lot with simple heuristics? When you ban a tag, track which new new hash tags suddenly explode in popularity. Pull a list of hashtags with an unusually high number of links/images to sites that don't normally get linked? Track the people who follow these hashtags, and see what hashtags they go to next If you actually want to fight the problem, it doesn't require a huge amount of resources - the problem is entirely that Meta has no actual desire to fight the problem (after all, it costs money and doesn't produce any benefit to them)

subjectsigma 3y ago
I think it’s a bit harder than you realize. > How are these bad actors coordinating hashtags in a way that can't be detected? They manage to do it in China despite the massive censorship there. https://sg.news.yahoo.com/chinese-vpn-providers-hide-literature-disguise-names-031019230.html https://sg.news.yahoo.com/chinese-vpn-providers-hide-literat... > it doesn't require a huge amount of resources Are you suggesting that the Chinese government just isn’t trying hard enough, and if they really wanted to they could stop all dissidents in China for forever?
- handoflixue 3y ago
  I never said it was perfect. I'm saying that if you invest resources, you can accomplish a lot - even if it isn't perfect. China would be a perfect example of my point: they've succeeded in vastly reducing the amount of censored content available, and the average person has no real desire to work around that. I really don't get how you think your example disagrees with me - do you think China hasn't accomplished anything there?
  - subjectsigma 3y ago
    Well, they set out to eliminate VPN usage and prevent their citizens from criticizing the government. There’s still VPNs and people still criticize the government. I don’t think the “average person” is using a VPN but I would be shocked if subtle ways to dodge censorship weren’t widely known. It’s also a treadmill - the problem isn’t fundamentally solved, every time a new hashtag is squashed another one pops up. They will just keep manually identifying and banning them for forever. Flipping the question around, I’m not really sure how Facebook’s current systems don’t fit your definition of “it’s not perfect but it works”?
    - handoflixue 3y ago
      > I’m not really sure how Facebook’s current systems don’t fit your definition of “it’s not perfect but it works”? I think there's a lot of low-hanging fruit we could easily address with a small increase in spending, and I think it's fair to shame companies for not doing that. I also think there's a lot of mid-height fruit that could plausibly be addressed with a lot of spending. I think it's fair to praise companies that ARE doing that, but it's hardly a source of shame. And of course, there's plenty of fruit we're never going to deal with, no matter how much we spend. I think Facebook is under-spending and missing out on low hanging fruit
ozim 3y ago
Very naive way of thinking that there must be a secret channel or cabal. Think more about how memes spread in the internet - there is no “meme administration” that picks what is funny and what not. You still have to make difference between trends of normal hashtags and evil hashtags how do you do that? Also how do you even think there would be explosion when bad actors are aware they have to hide. Main point is simple heuristics won’t work and you don’t have to be evil mastermind to circumvent them.
- handoflixue 3y ago
  Okay, but why can't the moderators monitor this spread the same way the users do? > how do you do that? I already listed a few examples. If it was unclear, the other part of the process is "have a human check the hashtag", sorry > Also how do you even think there would be explosion when bad actors are aware they have to hide. You'd be really amazed how much low-hanging fruit there is here. If someone is posting publicly on social media AND using hashtags, they're not exactly in the top half of OpSec
  - ozim 3y ago
    To be fair I was arguing assumptions "all can be fixed automatically" and that "all can be fixed, all bad guys can be caught". I do agree that there is a lot that can be caught with a bit more effort and that Meta does not really want to spend resources because it doesn't earn them money. But also I argue that there still will be a lot of bad stuff and there is not enough money/resources to stop all.
    - handoflixue 3y ago
      I think we pretty much agree there - we're never going to wipe it all out, but I do think our current spending level is missing out on a lot of low-hanging fruit that could easily be eliminated if management cared to

subjectsigma 3y ago

I think it’s a bit harder than you realize. > How are these bad actors coordinating hashtags in a way that can't be detected? They manage to do it in China despite the massive censorship there. https://sg.news.yahoo.com/chinese-vpn-providers-hide-literature-disguise-names-031019230.html https://sg.news.yahoo.com/chinese-vpn-providers-hide-literat... > it doesn't require a huge amount of resources Are you suggesting that the Chinese government just isn’t trying hard enough, and if they really wanted to they could stop all dissidents in China for forever?

handoflixue 3y ago
I never said it was perfect. I'm saying that if you invest resources, you can accomplish a lot - even if it isn't perfect. China would be a perfect example of my point: they've succeeded in vastly reducing the amount of censored content available, and the average person has no real desire to work around that. I really don't get how you think your example disagrees with me - do you think China hasn't accomplished anything there?
- subjectsigma 3y ago
  Well, they set out to eliminate VPN usage and prevent their citizens from criticizing the government. There’s still VPNs and people still criticize the government. I don’t think the “average person” is using a VPN but I would be shocked if subtle ways to dodge censorship weren’t widely known. It’s also a treadmill - the problem isn’t fundamentally solved, every time a new hashtag is squashed another one pops up. They will just keep manually identifying and banning them for forever. Flipping the question around, I’m not really sure how Facebook’s current systems don’t fit your definition of “it’s not perfect but it works”?
  - handoflixue 3y ago
    > I’m not really sure how Facebook’s current systems don’t fit your definition of “it’s not perfect but it works”? I think there's a lot of low-hanging fruit we could easily address with a small increase in spending, and I think it's fair to shame companies for not doing that. I also think there's a lot of mid-height fruit that could plausibly be addressed with a lot of spending. I think it's fair to praise companies that ARE doing that, but it's hardly a source of shame. And of course, there's plenty of fruit we're never going to deal with, no matter how much we spend. I think Facebook is under-spending and missing out on low hanging fruit

ozim 3y ago

Very naive way of thinking that there must be a secret channel or cabal. Think more about how memes spread in the internet - there is no “meme administration” that picks what is funny and what not. You still have to make difference between trends of normal hashtags and evil hashtags how do you do that? Also how do you even think there would be explosion when bad actors are aware they have to hide. Main point is simple heuristics won’t work and you don’t have to be evil mastermind to circumvent them.

handoflixue 3y ago
Okay, but why can't the moderators monitor this spread the same way the users do? > how do you do that? I already listed a few examples. If it was unclear, the other part of the process is "have a human check the hashtag", sorry > Also how do you even think there would be explosion when bad actors are aware they have to hide. You'd be really amazed how much low-hanging fruit there is here. If someone is posting publicly on social media AND using hashtags, they're not exactly in the top half of OpSec
- ozim 3y ago
  To be fair I was arguing assumptions "all can be fixed automatically" and that "all can be fixed, all bad guys can be caught". I do agree that there is a lot that can be caught with a bit more effort and that Meta does not really want to spend resources because it doesn't earn them money. But also I argue that there still will be a lot of bad stuff and there is not enough money/resources to stop all.
  - handoflixue 3y ago
    I think we pretty much agree there - we're never going to wipe it all out, but I do think our current spending level is missing out on a lot of low-hanging fruit that could easily be eliminated if management cared to