5 Dec 2014

MHRM Inquiry To Twitter Management

By In the past few months I’ve seen some discussion on the exploitation of vulnerable aspects of Twitter’s rules by social justice warriors to silence political dissent by using bots to mass-block or mass-report so dissenters’ accounts would be suspended. On October 29, I emailed press@twitter.com with questions, explaining my intent behind the request. I wanted to be upfront about the fact that this was for an article to ensure that their reply would consist of information they wanted known. My email:


Recent trends in some groups’ tweets have included referencing the option included in the “block or report” menu to report another user for content that is “Generally offensive, disrespectful or in disagreement with my opinion.” Some users are touting that option as a means to silence political dissent. This sounded unlike a social site policy, and led me to look further into the reporting options.
Further investigation has only led to more questions, including questions about other options for reporting. The help bubble for that option says “Twitter does not screen content and does not remove potentially offensive content unless such content is in violation of the Twitter Rules and Terms of Service,” so I have read those.
Some of the twitter rules have wording which could be subject to broad interpretation and potential abuse. I’m writing to request clarification, as I am considering an article. My current focus is communication, harassment, censorship, and how twitter’s rules affect engagement between users in relation to political and other potentially controversial discussions. Any information I receive in response to these questions will contribute to that article.

For instance, listed under “targeted abuse” are
  • if the sole purpose of your account is to send abusive messages to others;
and
  • if the reported behavior is one-sided or includes threats
What factors determine if the sole purpose of an account is to send abusive messages to others?

Is disagreement with others’ opinion considered an abusive message, and if so does that mean engaging in debate by replying to publicly made statements of opinion with counterarguments is now considered abuse?
If not, what is the specific reason for including “in disagreement with my opinion” in the harassment reporting options?

I am assuming that tweets without @user mentions are not considered “targeted” because they do not involve intentional contact and users can block feed they do not want to see. Is this correct?

Does “targeted abuse” include @user mentions without abusive language, but which contain a dissenting counterargument to something the user has said? (eg. I disagree with @twitteruser’s stated opinion in LinkToPublication because reason #politicalorsocialconcept or @twitteruser I disagree with what you said.) If so, is that a blanket rule or does the behavior have to be repeated or continue after the user engaging in it has received some variation of “stop talking to me” before it is considered to be in violation?

Does the tone of a tweet change whether it is considered abusive, and if so what is the criteria for that? (“I disagree with what you said” vs a similar line containing expletives vs a line containing slurs or the suggestion that the user deserves to suffer violence because of his opinion.)

Under the Spam heading are the following criteria:
  • If you have followed and/or unfollowed large amounts of users in a short time period, particularly by automated means (aggressive following or follower churn);

When a political or social movement forms on twitter, users often follow each other so that posts about their specific pet issue will show up in their feed. Is it a goal of twitter to prevent this, or does following require automated means in order to be considered a violation?
  • If your updates consist mainly of links, and not personal updates;
Many users post links to news stories, videos, and memes related to hashtags created for issues and concepts that have meaning to them. Very often these are, to the user, personal updates because they are new information about the issue or concept. Will users whose personal updates usually include links to information be considered spammers?
  • If a large number of people are blocking you;
  • If a large number of spam complaints have been filed against you;
This looks like it could be used by ideological groups to silence popular or very vocal users whose updates contain statements of opinion to which they are opposed, by banding together to all block and/or report the same person, thereby putting him or her in violation. What is twitter’s administration doing to counter potential abuse of these rules?
  • If you post multiple unrelated updates to a topic using #, trending or popular topic, or promoted trend;
There are Twitter users currently interpreting this to mean that dissenting views to a hashtag’s political outlook and criticism of the belief or assertion behind a hashtag can be considered spam if it is posted to the hashtag. For example, posting an opposing viewpoint with a political hashtag can offend the readers of the hashtag, but it’s a discussion attempt. Is it considered spam even if it is not belligerent or threatening?
  • Randomly or aggressively following, favoriting or Retweeting Tweets;
This line is hard to interpret. Can you please give a more detailed explanation of what it means? How does twitter determine when following, favoriting, and retweeting are random or aggressive, and when doing those things is acceptable?

Finally, do you have a policy in place to deal with habitual, vexatious false flagging (users repeatedly reporting content which does not violate rules or terms of service in an attempt to silence speech they disapprove or people they dislike)? This is a behavior I’ve seen on other social networking sites where administrators have written and posted broadly interpretable policy without stipulating that false exploitation of it would also lead to discipline. What is Twitter doing to prevent users from banding together to engage in this type of harassment, and how do you differentiate between a user engaging in this and a user who spots or experiences and reports a lot of genuine violations?

If you provide me with a statement on the overall goal of Twitter’s rules and guidelines as they relate to discussion on the site, I will include that in the article.

If you have any questions for me, please let me know, and I’ll be happy to clarify anything that has not come across clearly. Thank you in advance for taking the time to address these questions. I hope to produce an article that will bring about better understanding of Twitter’s rules and their relationship to the many discussions among users of the site.

To date, I have not heard back. I cannot be sure if this is simply due to their staff for handling press requests being small and unable to keep up or if they are avoiding answering the questions. If it is the latter, it could be because now that they’ve realized there are holes in their policy, they want to fix them before replying. It could also be due to a desire to leave those holes in place without having to answer for them. The only way to know would be to hear back from Twitter, and that is largely out of my control. 
To try to get Twitter to answer my questions, I’ve tweeted @support.
The links go to the original posting of this on Breaking the Glasses, the Atheism+ blockbot list, and the github page for the GGautoblocker, which is up to nearly 10,000 accounts (including @WRCorg and at first, @KFC) blocked for being even remotely associated with #gamergate. According to the github talk, they can’t block Christina Hoff Sommers or Adam Baldwin without slowing down the list’s functioning. Apparently having a ton of followers makes a difference.
The GGautoblocker block list has its own personal Twitter account under its own name. If you want to find out if you’re on it, according to a link posted in that account, a list of blocked users, broken down into chunks of 500 accounts, is here.
I’m all for Twitter users having the choice to block in order to avoid unwanted contact and conflict. That’s everyone’s right. However, both bots appear to be in violation of the terms of service in that bots are forbidden, and these can easily be employed in the manner I’ve described in the letter to shut down accounts over political disagreement. The GGautoblocker is being actively used as a tool of attempted (and failed) intimidation, as users not associated with #gamergate who find themselves on it are being told by #gamergate’s opponents who they should or should not follow and who they should or should not allow to follow them.
If Twitter condones this use of blockbots, that is a clear demonstration that their rule system is not about preventing abuse but about facilitating a censorship effort. It’s a question that, if news@twitter will not answer, the @support staff should. The other question is: How many users are going to have to ask before Twitter’s staff will pay attention to the problem?


About Hannah Wallen

Hannah Wallen (aka Della Burton) became an MRA after weathering the direct impact of feminist advocacy on her family for years. She is the author of Breaking the Glasses, written from an anti-feminist perspective, with a focus on men's rights and sometimes social issues. She also writes and comments on reddit.com as oneiorosgrip.

Source

No comments:

Post a Comment