3 ms·

> The point is, countering negative ideas with suppression does not work. This is the core of your argument, and if true, it’s unassailable. However, I don’t b

> The point is, countering negative ideas with suppression does not work. This is the core of your argument, and if true, it’s unassailable. However, I don’t believe it’s true. Do you have evidence that it is? I don’t mean well-crafted arguments by respected people, but actual evidence. For my part, I’ve seen evidence to the contrary. Elsewhere in this thread, there’s a study that cites the positive benefits of deplatforming on Reddit. I’ve also observed that online forums invariably turn into cesspits if they aren’t moderated. The larger the forum, the more aggressive that moderation has to be, to the point of banning and shadowbanning. HN does it. (For that matter, downvoting is another form of deplatforming, in that it literally pushes other people’s opinions out of sight.) So there’s two points of evidence that make me believe that deplatforming is effective: one is academic; and one is the personal observation, that I think we’ve all shared, that moderated fora work better than unmoderated fora. What evidence do you have? Please summarize rather than just posting links.

rhaksw 4y ago
> This is the core of your argument, and if true, it’s unassailable. However, I don’t believe it’s true. Do you have evidence that it is? I don’t mean well-crafted arguments by respected people, but actual evidence. > For my part, I’ve seen evidence to the contrary. Elsewhere in this thread, there’s a study that cites the positive benefits of deplatforming on Reddit. I'm familiar with that study and I'm pretty sure they acknowledged that they only measured activity of individuals who remained on Reddit. It's really hard to definitively say what happens when groups move to private forums where you can no longer track their conversations. Also, what people say in public vs. private can be different [1] [2], and I think periods when that is true (such as right now) should be cause for consideration about whether or not censorship contributes to that. I'd argue that state or widespread cultural censorship does contribute to self-censorship by chilling conversations, and that people's views don't change when you block them from view. It's like sending someone to prison vs. rehabilitation. You might still argue that the views can't spread as easily, and I would argue against that too. When groups are deplatformed, they find their own ways to communicate, either by joining private groups on services like Telegram or by creating their own platforms. Then they are outside your sphere of influence and harder to reach. We are pushing them in this direction and I think we will end up regarding that as a mistake. But it's not like this is the first time that's happened. You can join my talk next Wednesday, October 12th [3] to learn more about what I think with sources. It's the first one listed. This particular question is not the focus, but it is related. I'd also recommend listening to free speech defenders such as Ira Glasser, Nadine Strossen, Greg Lukianoff, and Jonathan Rauch, to name a few [4]. They've all spoken at length about this via many forms of media, books, podcasts, conferences, etc. Anyone interviewed on the So to Speak podcast is also great, particularly the earlier episodes where they bring in the older generations who've been defending free speech for their entire lives. The evidence is there, but every time there is a new technology it takes time to collect. You can instead consider a principled approach based on what you know works between individuals. Jonathan Rauch makes a great case for this in his books and speaking. > I’ve also observed that online forums invariably turn into cesspits if they aren’t moderated. I think this is true! Yet, the more we censor, the more concentrated the echo chambers become, both the ones you agree with and the ones with whom you disagree. So IMO we must stop regarding "ugly" forums as a bad thing to be avoided. We all have difficult conversations in the real world, and I think curating the online world to look nice and pretty only provides more evidence to the idea that our current "solution" to online disagreements isn't working and appears to be infecting the real world. > The larger the forum, the more aggressive that moderation has to be, to the point of banning and shadowbanning. HN does it. That is evidence that it's popular, not that it is a good idea. > (For that matter, downvoting is another form of deplatforming, in that it literally pushes other people’s opinions out of sight.) I agree, and I think that has the unfortunate side effect of hiding some useful rebuttals, as I mentioned elsewhere in this thread [5]. > So there’s two points of evidence that make me believe that deplatforming is effective: one is academic; and one is the personal observation, that I think we’ve all shared, that moderated fora work better than unmoderated fora. What evidence do you have? Please summarize rather than just posting links. I'd say the jury's out on how to effectively moderate fora. We're just at the start of this era and it's been quite a bumpy ride. Moderation, at the very least, should be transparent to the authors of the content, and that's not the case across all of the major platforms, including this one. [1] https://youtu.be/-ByRjHwknbc?t=2892 https://youtu.be/-ByRjHwknbc?t=2892 [2] https://www.axios.com/2022/08/17/americans-voters-private-belief-poll-abortion-education-covid-19 https://www.axios.com/2022/08/17/americans-voters-private-be... [3] https://truthandtrustonline.com/pre-conference-workshops/ https://truthandtrustonline.com/pre-conference-workshops/ [4] https://www.reddit.com/r/reveddit/comments/wxfjvy/im_on_a_podcast_talking_about_shadow_moderation/in9fh8s/?context=3 https://www.reddit.com/r/reveddit/comments/wxfjvy/im_on_a_po... [5] https://news.ycombinator.com/item?id=33055630 https://news.ycombinator.com/item?id=33055630
- jdlshore 4y ago
  Thanks for your thoughtful reply. You didn't actually answer my request for evidence, though. I think is the study we've been discussing: https://seclab.bu.edu/people/gianluca/papers/deplatforming-websci2021.pdf https://seclab.bu.edu/people/gianluca/papers/deplatforming-w... This is my understanding of their analysis, based on a fairly shallow read: > Are accounts being created on an alternative platform after being suspended? A: Yes, 59% of Twitter users and 76% of Reddit users moved to Gab. > Do suspended users become more toxic if they move to another platform? A: Reddit users became more toxic on Gab. 60% of Twitter users became less toxic and 20% became much more toxic, although the most toxic posts contained hatred against Twitter and complaints that their free speech and rights had been denied. (Toxicity was determined using Google's Perspective API.) > Do suspended users become more active if they move to another platform? A: Yes. A manual inspection determines that at least some of that increased activity is complaints about being suspended. > Do suspended users gain more followers on the other platform? A: Although users tend to become more toxic and more active after they move to the alternative platform, their audience decreases. I think you could read this either way. Deplatforming is ineffective because it "radicalizes" those have been deplatformed. Or; deplatforming is effective because it reduces the spread of toxicity. Your post above is mainly focusing on the former; my post focused mainly on the latter. The jury's still out, as you said. Personally, I'll continue to lean in favor of moderation, if only for the selfish reason that unmoderated communities are nasty places, and I want to participate in communities that "bring me joy," to indulge in a Kondo-ism. I think we've shown pretty conclusively, though, that your argument "The point is, countering negative ideas with suppression does not work" is premature at best. I'll let you have the last word. Best wishes.
  - rhaksw 4y ago
    > You didn't actually answer my request for evidence, though I agree with the NYT that censorship is rooted in fear [1]. Evidence is aplenty of the benefits of open discourse. In real-world places where open discourse is encouraged, people and ideas thrive. Also, saying "you need to be protected from other people's words" is not a winning argument in the public sphere. People want to be trusted to make their own decisions about how to feel, not have the importance of speech dictated to them. It's really only a small minority who seek protection against certain viewpoints, and they too want to be able to express themselves. Unfortunately, censorship is also used against them, often with prejudice and without their knowledge [2]. History has shown how this has happened over and over, for example in "Don't Be a Sucker" (1947) [3]. If you choose to ignore it, that is your prerogative. History is evidence. > I think is the study we've been discussing Thanks for linking it, that's not the one I had in mind. To expand on my previous comment about how we should accept "ugly" forums, I think measuring toxicity is problematic. For one thing, it's a subjective measure. One man's trash is another's treasure. For example, here's an article from someone making a case in favor of Kiwi Farms [4]. But also, censorship can chill what people state publicly. I already shared Axios's write-up of a recent study that shows that these days, what people say publicly does not align with what they say privately [5]. > The jury's still out, as you said. Personally, I'll continue to lean in favor of moderation, if only for the selfish reason that unmoderated communities are nasty places, and I want to participate in communities that "bring me joy," to indulge in a Kondo-ism. I think we've shown pretty conclusively, though, that your argument "The point is, countering negative ideas with suppression does not work" is premature at best. FWIW, I think moderation is fine if the author is informed of actions taken against their content. That is not happening consistently on any of the platforms though, and hundreds of millions, perhaps billions, of users are impacted. Load 10 tabs of randomly selected active Reddit users on Reveddit [6]. Five or more will have had a comment removed within their first page of recent history. Almost none of these will have been notified, and all of their removed comments are shown to them as if they're not removed. I just did it and got 7. Reddit last reported 450 million monthly active users. And, Facebook moderators have a "Hide comment" button that does the same thing: > "Hiding the Facebook comment will keep it hidden from everyone except that person and their friends. They won’t know that the comment is hidden, so you can avoid potential fallout." [7] It's hard for me to believe that this has had no negative impact on discourse, particularly when our recent difficulties communicating across ideologies seem to align quite well with the introduction of social media. Things like this 1998 Firing Line episode [8] simply are not happening today. The depth of conversations these days is shallow and combative. > I'll let you have the last word. Best wishes. I will reject (graciously, I hope) your offer. I think continued discussion is the way forward. [1] https://www.nytimes.com/2022/09/10/opinion/schools-banned-books.html https://www.nytimes.com/2022/09/10/opinion/schools-banned-bo... [2] https://www.reveddit.com/about/faq/#need https://www.reveddit.com/about/faq/#need [3] https://www.youtube.com/watch?v=23X14HS4gLk https://www.youtube.com/watch?v=23X14HS4gLk [4] https://corinnacohn.substack.com/p/the-world-should-not-need-kiwi-farms https://corinnacohn.substack.com/p/the-world-should-not-need... [5] https://www.axios.com/2022/08/17/americans-voters-private-belief-poll-abortion-education-covid-19 https://www.axios.com/2022/08/17/americans-voters-private-be... [6] https://www.reveddit.com/random https://www.reveddit.com/random [7] https://www.agorapulse.com/blog/hide-comments-on-facebook/#options-for-social-media-managers https://www.agorapulse.com/blog/hide-comments-on-facebook/#o... [8] https://www.youtube.com/watch?v=2KQmPeM0Gmo https://www.youtube.com/watch?v=2KQmPeM0Gmo

rhaksw 4y ago

> This is the core of your argument, and if true, it’s unassailable. However, I don’t believe it’s true. Do you have evidence that it is? I don’t mean well-crafted arguments by respected people, but actual evidence. > For my part, I’ve seen evidence to the contrary. Elsewhere in this thread, there’s a study that cites the positive benefits of deplatforming on Reddit. I'm familiar with that study and I'm pretty sure they acknowledged that they only measured activity of individuals who remained on Reddit. It's really hard to definitively say what happens when groups move to private forums where you can no longer track their conversations. Also, what people say in public vs. private can be different [1] [2], and I think periods when that is true (such as right now) should be cause for consideration about whether or not censorship contributes to that. I'd argue that state or widespread cultural censorship does contribute to self-censorship by chilling conversations, and that people's views don't change when you block them from view. It's like sending someone to prison vs. rehabilitation. You might still argue that the views can't spread as easily, and I would argue against that too. When groups are deplatformed, they find their own ways to communicate, either by joining private groups on services like Telegram or by creating their own platforms. Then they are outside your sphere of influence and harder to reach. We are pushing them in this direction and I think we will end up regarding that as a mistake. But it's not like this is the first time that's happened. You can join my talk next Wednesday, October 12th [3] to learn more about what I think with sources. It's the first one listed. This particular question is not the focus, but it is related. I'd also recommend listening to free speech defenders such as Ira Glasser, Nadine Strossen, Greg Lukianoff, and Jonathan Rauch, to name a few [4]. They've all spoken at length about this via many forms of media, books, podcasts, conferences, etc. Anyone interviewed on the So to Speak podcast is also great, particularly the earlier episodes where they bring in the older generations who've been defending free speech for their entire lives. The evidence is there, but every time there is a new technology it takes time to collect. You can instead consider a principled approach based on what you know works between individuals. Jonathan Rauch makes a great case for this in his books and speaking. > I’ve also observed that online forums invariably turn into cesspits if they aren’t moderated. I think this is true! Yet, the more we censor, the more concentrated the echo chambers become, both the ones you agree with and the ones with whom you disagree. So IMO we must stop regarding "ugly" forums as a bad thing to be avoided. We all have difficult conversations in the real world, and I think curating the online world to look nice and pretty only provides more evidence to the idea that our current "solution" to online disagreements isn't working and appears to be infecting the real world. > The larger the forum, the more aggressive that moderation has to be, to the point of banning and shadowbanning. HN does it. That is evidence that it's popular, not that it is a good idea. > (For that matter, downvoting is another form of deplatforming, in that it literally pushes other people’s opinions out of sight.) I agree, and I think that has the unfortunate side effect of hiding some useful rebuttals, as I mentioned elsewhere in this thread [5]. > So there’s two points of evidence that make me believe that deplatforming is effective: one is academic; and one is the personal observation, that I think we’ve all shared, that moderated fora work better than unmoderated fora. What evidence do you have? Please summarize rather than just posting links. I'd say the jury's out on how to effectively moderate fora. We're just at the start of this era and it's been quite a bumpy ride. Moderation, at the very least, should be transparent to the authors of the content, and that's not the case across all of the major platforms, including this one. [1] https://youtu.be/-ByRjHwknbc?t=2892 https://youtu.be/-ByRjHwknbc?t=2892 [2] https://www.axios.com/2022/08/17/americans-voters-private-belief-poll-abortion-education-covid-19 https://www.axios.com/2022/08/17/americans-voters-private-be... [3] https://truthandtrustonline.com/pre-conference-workshops/ https://truthandtrustonline.com/pre-conference-workshops/ [4] https://www.reddit.com/r/reveddit/comments/wxfjvy/im_on_a_podcast_talking_about_shadow_moderation/in9fh8s/?context=3 https://www.reddit.com/r/reveddit/comments/wxfjvy/im_on_a_po... [5] https://news.ycombinator.com/item?id=33055630 https://news.ycombinator.com/item?id=33055630

jdlshore 4y ago
Thanks for your thoughtful reply. You didn't actually answer my request for evidence, though. I think is the study we've been discussing: https://seclab.bu.edu/people/gianluca/papers/deplatforming-websci2021.pdf https://seclab.bu.edu/people/gianluca/papers/deplatforming-w... This is my understanding of their analysis, based on a fairly shallow read: > Are accounts being created on an alternative platform after being suspended? A: Yes, 59% of Twitter users and 76% of Reddit users moved to Gab. > Do suspended users become more toxic if they move to another platform? A: Reddit users became more toxic on Gab. 60% of Twitter users became less toxic and 20% became much more toxic, although the most toxic posts contained hatred against Twitter and complaints that their free speech and rights had been denied. (Toxicity was determined using Google's Perspective API.) > Do suspended users become more active if they move to another platform? A: Yes. A manual inspection determines that at least some of that increased activity is complaints about being suspended. > Do suspended users gain more followers on the other platform? A: Although users tend to become more toxic and more active after they move to the alternative platform, their audience decreases. I think you could read this either way. Deplatforming is ineffective because it "radicalizes" those have been deplatformed. Or; deplatforming is effective because it reduces the spread of toxicity. Your post above is mainly focusing on the former; my post focused mainly on the latter. The jury's still out, as you said. Personally, I'll continue to lean in favor of moderation, if only for the selfish reason that unmoderated communities are nasty places, and I want to participate in communities that "bring me joy," to indulge in a Kondo-ism. I think we've shown pretty conclusively, though, that your argument "The point is, countering negative ideas with suppression does not work" is premature at best. I'll let you have the last word. Best wishes.
- rhaksw 4y ago
  > You didn't actually answer my request for evidence, though I agree with the NYT that censorship is rooted in fear [1]. Evidence is aplenty of the benefits of open discourse. In real-world places where open discourse is encouraged, people and ideas thrive. Also, saying "you need to be protected from other people's words" is not a winning argument in the public sphere. People want to be trusted to make their own decisions about how to feel, not have the importance of speech dictated to them. It's really only a small minority who seek protection against certain viewpoints, and they too want to be able to express themselves. Unfortunately, censorship is also used against them, often with prejudice and without their knowledge [2]. History has shown how this has happened over and over, for example in "Don't Be a Sucker" (1947) [3]. If you choose to ignore it, that is your prerogative. History is evidence. > I think is the study we've been discussing Thanks for linking it, that's not the one I had in mind. To expand on my previous comment about how we should accept "ugly" forums, I think measuring toxicity is problematic. For one thing, it's a subjective measure. One man's trash is another's treasure. For example, here's an article from someone making a case in favor of Kiwi Farms [4]. But also, censorship can chill what people state publicly. I already shared Axios's write-up of a recent study that shows that these days, what people say publicly does not align with what they say privately [5]. > The jury's still out, as you said. Personally, I'll continue to lean in favor of moderation, if only for the selfish reason that unmoderated communities are nasty places, and I want to participate in communities that "bring me joy," to indulge in a Kondo-ism. I think we've shown pretty conclusively, though, that your argument "The point is, countering negative ideas with suppression does not work" is premature at best. FWIW, I think moderation is fine if the author is informed of actions taken against their content. That is not happening consistently on any of the platforms though, and hundreds of millions, perhaps billions, of users are impacted. Load 10 tabs of randomly selected active Reddit users on Reveddit [6]. Five or more will have had a comment removed within their first page of recent history. Almost none of these will have been notified, and all of their removed comments are shown to them as if they're not removed. I just did it and got 7. Reddit last reported 450 million monthly active users. And, Facebook moderators have a "Hide comment" button that does the same thing: > "Hiding the Facebook comment will keep it hidden from everyone except that person and their friends. They won’t know that the comment is hidden, so you can avoid potential fallout." [7] It's hard for me to believe that this has had no negative impact on discourse, particularly when our recent difficulties communicating across ideologies seem to align quite well with the introduction of social media. Things like this 1998 Firing Line episode [8] simply are not happening today. The depth of conversations these days is shallow and combative. > I'll let you have the last word. Best wishes. I will reject (graciously, I hope) your offer. I think continued discussion is the way forward. [1] https://www.nytimes.com/2022/09/10/opinion/schools-banned-books.html https://www.nytimes.com/2022/09/10/opinion/schools-banned-bo... [2] https://www.reveddit.com/about/faq/#need https://www.reveddit.com/about/faq/#need [3] https://www.youtube.com/watch?v=23X14HS4gLk https://www.youtube.com/watch?v=23X14HS4gLk [4] https://corinnacohn.substack.com/p/the-world-should-not-need-kiwi-farms https://corinnacohn.substack.com/p/the-world-should-not-need... [5] https://www.axios.com/2022/08/17/americans-voters-private-belief-poll-abortion-education-covid-19 https://www.axios.com/2022/08/17/americans-voters-private-be... [6] https://www.reveddit.com/random https://www.reveddit.com/random [7] https://www.agorapulse.com/blog/hide-comments-on-facebook/#options-for-social-media-managers https://www.agorapulse.com/blog/hide-comments-on-facebook/#o... [8] https://www.youtube.com/watch?v=2KQmPeM0Gmo https://www.youtube.com/watch?v=2KQmPeM0Gmo