6 ms·

Tolerance of wrong viewpoints is different than the active support algorithms give to discovering false content.

by wslack 5y ago

throwawaysea 5y ago
It’s not “active” support if the algorithm acts in a content neutral fashion, for example based on engagement metrics. In such a situation, changing the algorithm to artificially not let allegedly-false content be discovered is actively supporting the opposing viewpoints. Leaving the algorithm to act without artificial content-specific modification is not active support. Tolerance would be leaving the algorithms alone.
- jjeaff 5y ago
  I disagree. I think the algorithms are fundamentally immoral because they promote content that gets "engagement". Which includes and in many cases prioritizes content that people have engaged with because it causes a negative response. Rather than pushing good* content, it prioritizes lowest common denominator, reality tv, desperate pundit, fast food, self congratulatory, outrage porn garbage. *By good, I simply mean thoughtful, high quality, factual, educational, or otherwise uplifting content regardless of politics
  - ShroudedNight 5y ago
    I don't think "Good" is unambiguous enough to trust the platforms to promote it. How about simply "related"? Show people the content they've explicitly asked for. If people explicitly ask for outrageous content, then fine, but we needn't force feed it to society.
    - XorNot 5y ago
      Algorithms don't work like this though: content that feeds outrage disproportionately outranks content which doesn't. Algorithms don't discriminate "content" by it's actual content: they keyword match and look for clicks, and build a pretty perfect radicalization pathway more easily then they build a discourse [1]. You have probably experienced this: almost everyone has the experience of wanting to see some particular YouTube video, but opened it in an Incognito tab (or just avoided it) explicitly because they know the topic will prime the YouTube homepage to fill with nothing but things you don't want to see. [1] https://www.rollingstone.com/culture/culture-news/youtube-far-right-radicalization-study-877061/ https://www.rollingstone.com/culture/culture-news/youtube-fa...
- AlbertCory 5y ago
  Former Googler here (11 1/2 years, including in Ads) The idea that algorithms are "neutral" is laughable. There is a loosely organized group of activists out there who are aware of how these algorithms work and actively manipulate them. "Engagement metrics" are nothing more than these people pushing the buttons.
  - onetimeusename 5y ago
    I don't really understand why tech companies, like Google, go so far out of the way to maintain the image of being neutral. I agree they have a right to censor content they choose for whatever reason but what I don't understand is why they try appear to be neutral about their decisions. It feels like everyone is aware of what is going on, even other commenters who support Google censorship admit they approve of the bias. So why do tech companies cling to this line of being neutral when no one really seems to accept it and they themselves have no intention of being neutral? I feel like there wouldn't be any conflict about policies or complaints they have to deal with if they were more honest. Maybe it has to do with section 230. I don't know but I feel like we would be better off if consumers had more information.
    - mindslight 5y ago
      Every MITM-as-a-service starts off by being a neutral conduit to attract users, and then slowly adds restrictions to appease advertisers. But users never appreciate additional restrictions, and so Google (et al) have to keep marketing themselves as general hosts lest they lose even more mindshare.
- nobody_at_all 5y ago
  There is a difference between opposing viewpoints and factually incorrect information that is destructive. If your view is based on provable falsehoods, your view is worse than valueless. Tolerating these people is harmful to everyone, but that is not why Google is banning it. It is because it is harmful to Google's bottom line. At the end of the day, Google owns their servers and can say what is allowed and what is not.
  - peytn 5y ago
    Google is a legal construct. I can’t go have a coffee with Google. I can’t get a high five from Google. Google will do whatever our laws say it has to do in exchange for liability protections for its owners.
    - amanaplanacanal 5y ago
      If you can find a majority to agree you can change the laws. It seems unlikely though. I’m not even sure what you would change the law to be. Current reading of the US constitution says that Google has the same free speech rights that you do.
  - swayvil 5y ago
    >Google owns their servers and can say what is allowed and what is not. Well that's the problem in a nutshell. It ain't good for our society.
    - takeda 5y ago
      They are not there to improve society, they are there to make money, and if you think different you are a fool. Republicans fought to make companies people, and to be not regulated. And now are crying when it fired back, because forcing a company to publish or block something is actually stepping on its first amendment right. You won't really have such platform, unless it is done through a well established non-profit or through government (as long as government is Democratic and checks and balances work correctly).
    - 5y ago

Tolerance of wrong viewpoints is different than the active support algorithms give to discovering false content.

by wslack 5y ago

Tolerance of wrong viewpoints is different than the active support algorithms give to discovering false content.

It’s not “active” support if the algorithm acts in a content neutral fashion, for example based on engagement metrics. In such a situation, changing the algorithm to artificially not let allegedly-false content be discovered is actively supporting the opposing viewpoints. Leaving the algorithm to act without artificial content-specific modification is not active support. Tolerance would be leaving the algorithms alone.

jjeaff 5y ago
I disagree. I think the algorithms are fundamentally immoral because they promote content that gets "engagement". Which includes and in many cases prioritizes content that people have engaged with because it causes a negative response. Rather than pushing good* content, it prioritizes lowest common denominator, reality tv, desperate pundit, fast food, self congratulatory, outrage porn garbage. *By good, I simply mean thoughtful, high quality, factual, educational, or otherwise uplifting content regardless of politics
- ShroudedNight 5y ago
  I don't think "Good" is unambiguous enough to trust the platforms to promote it. How about simply "related"? Show people the content they've explicitly asked for. If people explicitly ask for outrageous content, then fine, but we needn't force feed it to society.
  - XorNot 5y ago
    Algorithms don't work like this though: content that feeds outrage disproportionately outranks content which doesn't. Algorithms don't discriminate "content" by it's actual content: they keyword match and look for clicks, and build a pretty perfect radicalization pathway more easily then they build a discourse [1]. You have probably experienced this: almost everyone has the experience of wanting to see some particular YouTube video, but opened it in an Incognito tab (or just avoided it) explicitly because they know the topic will prime the YouTube homepage to fill with nothing but things you don't want to see. [1] https://www.rollingstone.com/culture/culture-news/youtube-far-right-radicalization-study-877061/ https://www.rollingstone.com/culture/culture-news/youtube-fa...
AlbertCory 5y ago
Former Googler here (11 1/2 years, including in Ads) The idea that algorithms are "neutral" is laughable. There is a loosely organized group of activists out there who are aware of how these algorithms work and actively manipulate them. "Engagement metrics" are nothing more than these people pushing the buttons.
- onetimeusename 5y ago
  I don't really understand why tech companies, like Google, go so far out of the way to maintain the image of being neutral. I agree they have a right to censor content they choose for whatever reason but what I don't understand is why they try appear to be neutral about their decisions. It feels like everyone is aware of what is going on, even other commenters who support Google censorship admit they approve of the bias. So why do tech companies cling to this line of being neutral when no one really seems to accept it and they themselves have no intention of being neutral? I feel like there wouldn't be any conflict about policies or complaints they have to deal with if they were more honest. Maybe it has to do with section 230. I don't know but I feel like we would be better off if consumers had more information.
  - mindslight 5y ago
    Every MITM-as-a-service starts off by being a neutral conduit to attract users, and then slowly adds restrictions to appease advertisers. But users never appreciate additional restrictions, and so Google (et al) have to keep marketing themselves as general hosts lest they lose even more mindshare.
nobody_at_all 5y ago
There is a difference between opposing viewpoints and factually incorrect information that is destructive. If your view is based on provable falsehoods, your view is worse than valueless. Tolerating these people is harmful to everyone, but that is not why Google is banning it. It is because it is harmful to Google's bottom line. At the end of the day, Google owns their servers and can say what is allowed and what is not.
- peytn 5y ago
  Google is a legal construct. I can’t go have a coffee with Google. I can’t get a high five from Google. Google will do whatever our laws say it has to do in exchange for liability protections for its owners.
  - amanaplanacanal 5y ago
    If you can find a majority to agree you can change the laws. It seems unlikely though. I’m not even sure what you would change the law to be. Current reading of the US constitution says that Google has the same free speech rights that you do.
- swayvil 5y ago
  >Google owns their servers and can say what is allowed and what is not. Well that's the problem in a nutshell. It ain't good for our society.
  - takeda 5y ago
    They are not there to improve society, they are there to make money, and if you think different you are a fool. Republicans fought to make companies people, and to be not regulated. And now are crying when it fired back, because forcing a company to publish or block something is actually stepping on its first amendment right. You won't really have such platform, unless it is done through a well established non-profit or through government (as long as government is Democratic and checks and balances work correctly).
  - 5y ago