Share...
12 Jan 2015 12:27

Twitter takes aim at trolls—and promises more

Glenn Fleishman explains Twitter's blocking system and its freshly-fortified abuse-reporting tools.

By Glenn Fleishman

Twitter is home to all kinds of speech, some unpopular and some unpleasant, and it's often been faulted for an apparent indifference to harassment campaigns. Today, that changed: the company streamlined its abuse-reporting process and promised more to come.

Today's changes relate to blocking, which in Twitter limits another user's ability to interact with you. A blocked user's @-mentions disappear from your timeline, and your tweets cannot be favorited nor retweeted by people you block. The blockee is also automatically unfollowed if they were following your account. (The secret Twitter hip-check move is tapping Block and Unblock, which knocks someone off your stream without notifying them.)

Twitter has added a few refinements, being rolled out over "weeks" to all users. A Report for Spam option dates back to at least 2011, and appears in third-party software. Its web site and first-party client apps began offering a more granular File a Report in August 2013 with its iOS app.

But the report option is tweaky. The first step isn't bad, selecting a kind of behavior to report, such as "this tweet is annoying" or "this account may be compromised." Select "this user is abusive," and a set of options appear, including impersonation or private information. Tap Next in the current version, and you're required to fill out a form. The form is offputting, but also irritating to deal tap details into on a mobile device.

The revision adds the option to report a "sensitive image," and then dramatically improves the next steps, providing items from which to select, and offer an optional text field for more details at the end. This will be much easier to tap through on a cell phone, and will likely result in a vastly higher level of reporting as a result. Twitter also called out in its blog that this new system makes it easier for people to report abuse they witness occurring to others; its previous system dissuaded people form such reports.

Twitter has a nice sense of humor. As Ian Miles Cheong, a target of GamerGate, noted, "Twitter Support’s video features an alligator that gets blocked. It’s a gator. Get it?"

Twitter also will add a Blocked Accounts view on its account page, which should make it easier to remove people one has accidentally blocked or to review a block list if you're lucky enough to have only a few entries.

Further, Twitter is restoring a piece of behavior that seems to have disappeared a year ago when it briefly turned block into mute and then back again. While logged into an account in a browser or Twitter app that's been blocked by someone else, that blocked user cannot view the profile or tweets of the user who blocked him or her. (Twitter's mute invisibly filters messages from a muted party without them having any idea that's happening. Third-party client mute options are even more extensive.)

Blocking access to a profile for a public account is a minor deterrent. Someone can use a private browsing mode or another client or browser to see the tweets being made. But for casual or drive-by harassment, which is clearly the majority of abuse, it breaks the chain.

For committed stalkers or campaigns of abuse, however, it has no real effect. One hopes this is the first step coming out of a three-week partnership with Women, Action, & the Media (WAM), in which that outside group collected information about Twitter harassment, passed it to Twitter's Safety team, and responded to the person who filed a report about the outcome. The experiment ended just a few days ago. WAM says a report is coming. (I filed two reports during that period: one account was suspended; the other involved a deleted tweet.)

No word yet on whether collaborative blocking, which I wrote about at Boing Boing in August, will be added. This tactic allows people to share lists of those they bump and restores some symmetry to asymmetric campaigns of intimidation or abuse.

I currently use Randi Harper's Good Game Auto Blocker list via Block Together. Block Together is a Twitter web app that can examine and modify your block list. Randi's list cross-references Twitter followers of several GamerGate figures who are widely retweeted and who encourage dogpiling, to produce a list of over 9,000 Twitter accounts. (The Good Game system has a whitelist appeals process for those who follow the accounts but don't engage in abusive behavior.) Randi's list is a key reason I've been able to continue to use Twitter, as it prevents relentless ideological sealions from crowding my mentions.

In today's announcement, Twitter said that "the first of several behind-the-scenes improvements to the tools and processes that help us review reported Tweets and accounts" will enable faster response times. Of course, no one sensible wants Twitter to implement easy autobanning, because such tools are easily turned by trolls, griefers, and dogmatic groups against those who are abiding by the rules of a service. Collusion begets abuse of systems that lack good oversight. (Such behavior dates back at least thousands of years; for instance, the ostracism of Hyperbolus of Athens in 416 b.c.e.)

And having an unpopular point of view or being a jerk shouldn't result in a Twitter suspension. The ability to make one's Twitter timeline and experience one's own is what's at issue alongside the abuse reporting: to block those one doesn't want to hear and prevent their repeated assault, while still allowing them to rant as much as they wish to, so long as it doesn't violate the rules that Twitter sets. Most of the replies to me from the GamerGate community—distinguished from the more vile ones routinely received by women and people of color—are speech I don't care to hear, but I wouldn't report as a violation.

I hope Twitter can use the reputation information they collect—faves, connections, longevity on the service, previous abuse reports, and the relationship among accounts that simultaneously lodge complaints—to sift through legitimate and illegitimate violations.

Twitter said it's not done, and hints at what's to come, including "new enforcement procedures for abusive accounts." As I wrote recently for Boing Boing in "Serial offenders plague Twitter," quite a bit of the worst abuse comes from people who aren't deterred from registering accounts in series as each previous account is shut down by Twitter. Sarah Jeong described at The Verge some secret tools Twitter was testing. Perhaps these minor tweaks in blocking and reporting are just the tip of the iceberg of what's to come.

Published 2:19 pm Tue, Dec 2, 2014

, , , ,

About the Author

Glenn is a veteran technology journalist based in Seattle who regularly contributes to the Economist, Macworld, Boing Boing, Six Colors, TidBITS, and other publications. He's been blogging since 2001 at his Glog. In October 2012, Glenn won Jeopardy! twice. He tweets @glennf.

Continue the discussion at bbs.boingboing.net

14 replies