I noticed three main misleading arguments in the blog post:
* That the SSC survey tells us anything about the CW thread. Since per his own survey few of the SSC readers read the CW thread why should we expect that many of the CW regulars took the survey? More than twice as many SSC readers indicated that they were aware of the CW thread and choose to not read it than indicated that they read it, which you might take as evidence that the CW regulars are unlike the survey respondents as a whole in at least some ways. Even if you did establish that the readership was broad spectrum that wouldn't tell you anything about the absence of toxic garbage there. Most people are not comfortable engaging in a forum where harmful views 3-sigma outside of their overton window are taken seriously, even if they're only 1% of the posts in much the same way that most people would be uncomfortable accepting an optional blood transfusion where only 1% of the bags were contaminated by HIV and HEP-C.
[Here is the data on the survey respondents readership of the CW thread:
3175 (42.7%) "No, I don't want to read it", 2346 (31.5%) "No, I didn't know it existed", 1425 (19.1%) "Yes, I read it", 369 (4.9%) "Yes, I comment there", 118 (1.5%) no answer.
I don't want to assume bad faith, but I think it's really suspect that this wasn't mentioned in the post. I think it would be very surprising to the people arguing about the CW thread here on HN on the basis of scott's politics graphs.]
* That complaints are from Bay Area liberals that have never interacted with conservatives. As others in this thread have noted here the racist/xenophobic/misogynistic positions that have people shocked are views that wouldn't be presented in polite company pretty much anywhere in the US. Scott is unintentionally using a "motte and bailey" where he pretends that the complaint is merely conservative views, when instead what people get worked up about is stuff like forced sterilization.
* That the incident is an example of deplatforming or otherwise obviously related to trends that suppress public discussion. At least in so far as I am aware the well known examples of deplatforming looked nothing like the harassment Scott reported. I've seen no evidence that any well known deplatforming event was related to harassment by third parties. The CW thread itself was a move from the SSC subreddit to its own subreddit, not a deplatforming and has clearly radically increased its visibility.
And based on the commentary on HN, I think presence of these three fallacious arguments has significantly undermined public discourse on the subject. There are many thoughtful and interesting comments on HN-- expressing concerns about politics in the bay area, or about deplatforming-- which are essentially disconnected from SSC/CW thread specifically because these arguments served the rhetorical purpose of substituting the issue at hand with a largely unrelated one where the readers already had preformed strong opinions.
Scott would probably have done himself and the public a favor by making his post on this subject an adversarial collaboration.
* That the SSC survey tells us anything about the CW thread. Since per his own survey few of the SSC readers read the CW thread why should we expect that many of the CW regulars took the survey? More than twice as many SSC readers indicated that they were aware of the CW thread and choose to not read it than indicated that they read it, which you might take as evidence that the CW regulars are unlike the survey respondents as a whole in at least some ways. Even if you did establish that the readership was broad spectrum that wouldn't tell you anything about the absence of toxic garbage there. Most people are not comfortable engaging in a forum where harmful views 3-sigma outside of their overton window are taken seriously, even if they're only 1% of the posts in much the same way that most people would be uncomfortable accepting an optional blood transfusion where only 1% of the bags were contaminated by HIV and HEP-C.
[Here is the data on the survey respondents readership of the CW thread:
3175 (42.7%) "No, I don't want to read it", 2346 (31.5%) "No, I didn't know it existed", 1425 (19.1%) "Yes, I read it", 369 (4.9%) "Yes, I comment there", 118 (1.5%) no answer.
I don't want to assume bad faith, but I think it's really suspect that this wasn't mentioned in the post. I think it would be very surprising to the people arguing about the CW thread here on HN on the basis of scott's politics graphs.]
* That complaints are from Bay Area liberals that have never interacted with conservatives. As others in this thread have noted here the racist/xenophobic/misogynistic positions that have people shocked are views that wouldn't be presented in polite company pretty much anywhere in the US. Scott is unintentionally using a "motte and bailey" where he pretends that the complaint is merely conservative views, when instead what people get worked up about is stuff like forced sterilization.
* That the incident is an example of deplatforming or otherwise obviously related to trends that suppress public discussion. At least in so far as I am aware the well known examples of deplatforming looked nothing like the harassment Scott reported. I've seen no evidence that any well known deplatforming event was related to harassment by third parties. The CW thread itself was a move from the SSC subreddit to its own subreddit, not a deplatforming and has clearly radically increased its visibility.
And based on the commentary on HN, I think presence of these three fallacious arguments has significantly undermined public discourse on the subject. There are many thoughtful and interesting comments on HN-- expressing concerns about politics in the bay area, or about deplatforming-- which are essentially disconnected from SSC/CW thread specifically because these arguments served the rhetorical purpose of substituting the issue at hand with a largely unrelated one where the readers already had preformed strong opinions.
Scott would probably have done himself and the public a favor by making his post on this subject an adversarial collaboration.