3 ms·
After reading through the entire post, what I found lacking is a break down of what percentage of actual visitors & unique message senders were abusive. It app
by adventured 9y ago
After reading through the entire post, what I found lacking is a break down of what percentage of actual visitors & unique message senders were abusive.
It appears to be extraorinarily low. One in a thousand visitors or so. There's no question she was being abused because she was a woman. From what I can tell by their stats, that abuse appears to be coming from literally a handful of bad actors over a month's time, with many of the messages deriving from single persons.
So the question is: how do you deal with one bad person out of a thousand? I can't imagine there's any scenario, any technical implementation, that can eliminate abuse down much further than that no matter what you do. If you assume that even 1 in 100 men are going to be assholes, abusive, whatever, it's nearly impossible to stop them from behaving badly. You can take action after the fact, including in both anonymous and registered account situations. I also see no cultural scenario where you can pre-emptively stop that guy from being an asshole until he identifies himself through behavior.
One consideration would be pre-screening the chat for abusive language, abusive actors. That might catch a couple of the particularly bad examples. When a post is sent, you have to check it before it reaches the support agent. That's not difficult to do, except in the most heavily dodged scenarios of character replacement (s3x etc., some of which can still be pretty easily caught). Monsters will be monsters, there are at least a few options for dealing with them.
- valar_m 9y agoI suspect that you're deliberately missing the point, but I'll tell it to you anyway. It's not about the percentage of visitors that were abusive. It's the number of people who were abusive when they thought they were talking to a female vs. a male.
- bloak 9y agoWe don't know what they thought, obviously. We can only hypothesise about that. Psychology would be so much easier if we could directly observe other people's thoughts...