Ask HN: Why are Twitter and Facebook so poor at moderating hate speech?
In response to this: https://www.washingtonpost.com/news/morning-mix/wp/2016/11/11/get-some-of-them-to-kill-themselves-popular-neo-nazi-site-urges-readers-to-troll-liberals-into-suicide/?tid=hybrid_collaborative_3_na
I want to understand why social media companies don't seem to look as though they are taking responsibility for moderating extremists and coordinated hate groups trolling on their platforms. This problem has been escalating exponentially for years, especially over the US election cycle, and is a critical issue. Why does Silicon Valley feel as though it must not take a position on this?
Twitter, Facebook in my belief are no longer politically neutral organizations when they allow vindictive trolling and misinformation to proliferate and serve the agenda of alt-right hate groups. Do the people working at these companies ever ask themselves these questions or are they purposely oblivious to what's going on?
16 comments
[ 6.9 ms ] story [ 61.0 ms ] threadFor the second, you've merely provided (what I assume are) non-exhaustive examples.
Even there it is very hard to prove things. Was a particular message inspired by a particular post? People get bullied into suicide without there being a concerted effort and conspiracy, etc.
It is a tough issue and please don't frame it as left vs right because one thing history shows is that a law meant for one cause will be used for another. For instance, the "religious freedom" laws that boomed in the 1990s started as an attempt to legalize the use of peyote by the Native American Church. The association with anti-gay causes is entiely new. In reality, the biggest benefactor of those laws has probably been the Church of Scientology.
Is Black Lives Matter a "hate group?" What about Blue Lives Matter? What about an anti-BML group? Who decides? And once they're a hate group, according to you everything they publish is automatically considered "hate speech" and therefore banned. Seems overly broad.
Honestly yelling "it is just obvious" is very naive and unworkable. People have tried for years to come up with black and white definitions of these things and they've all failed. It often boils down to "when I see it, I know it" but that varies per individual.
From what I've observed, every publication has the Southern Poverty Law Center as the arbiter — and then you end up with things like this http://www.theexmuslim.com/2016/10/27/southern_poverty_law_c...
That said, I'm only inferring the nature of the "threats" from your description, I don't make a habit of looking at politicians' twitter feeds.
Lets look at one side: Your black so your automatically less
Lets look at the other side: Your a white male so you privileged
One of these is hate speech, the other is a protest/complaint about the current state of things. Functionally both are the same.
For years the courts have been cautious about dealing with language and the first amendment. If you sit down to think about WHY it is because it is a slippery slope.
Twitter and Facebook already do take positions: they actively kick people and groups off their services. Literally happens every day.
Prior to kicking people off, they actively "shadow ban" unsavory elements of their user base to prevent their ideas/hate speech from spreading.
Twitter even does this kind of shadow banning on a per-tweet basis. For instance, I follow a journalist named Sharyl Attkisson (@SharylAttkisson). She would always retweet Trump and Clinton in pairs, but I almost never got the Trump retweets (which gave me the distinct impression she was pro-Clinton). I only found out about the per-tweet suppression Twitter employed when she started tweeting about it.
Twitter even goes so far as to put up big ass warnings now when you click on web links to content that violates Twitter's (progressive-friendly, anti-hate) terms of service! The one I saw most recently was for links to https://voxday.blogspot.com/, which is (apparently) some kind of white nationalist site. Presumably, some people who try and follow such a link in their feed will not actually read it, thinking it's some kind of malware site.
What more do you expect them to do? From my perspective, they're already doing a lot.