There was a glorious few minutes the other day when many people in my Twitter feed were high-fiving each other for being on an SJW autoblock list. It functions by automatically blocking a list of "key" SJWs and all of their followers. Its Twitter account, @sjwautoblocker (which has been suspended), boasted listing over 120,000 "known SJWs."
For those who don't spend all their time on Twitter and Tumblr: "SJW" is a pejorative term that stands for "social justice warriors," which often includes anyone marginally leftist and who is vocal about social justice issues. SJW is a term frequently used by vehement MRAs (mens rights activists), conservative-leaning folks, and self proclaimed "value-free" people to mock and trivialize progressives. There seems to be a sense that SJWs aren't "free thinkers" (not necessarily true) and are only concerned with censoring (also not really true). Many people, including myself, have reclaimed it since we're not ashamed to advocate for pesky things like less harassment and hate speech, and general human rights.
So discovering that I was on such a list made me a little gleeful. But a little digging showed me that the SJW autoblock list is far from amusing.
My assumption, and that of most of us, was that the list was created out of the ongoing #gamergate fiasco. This is only partially true. The SJW autoblocker is a fork of software engineer Randi Harper's ggautoblocker (which means the original source code has been modified into new software by a different developer), which automatically blocks people tweeting about #gamergate. According to this Tumblr post, Harper's autoblocker was originally written to reduce the amount of harassment she was receiving from #gamergate supporters (she works at games company Kixeye). "In the original version, the list of sources was very short (7 names), and following 2 or more of those accounts was highly predictive of being somebody she didn't want in her mentions."
Harper took to Reddit and Twitter to warn the new SJW autoblocker was created by someone stalking her not explicitly identified as a #gamergate supporter. This person, according to Heroku software designer Tim Chevalier, is currently being investigated for harassing women.
Suddenly it doesn't feel so good to be included.
Harassment has always been a cornerstone of #gamergate, which I discussed back in August, ranging from crude commentary and revenge-porn style comics to rape and death threats. (I don't want to link to these, but you can easily find them.) We already knew the Internet isn't safe for women, because the world isn't safe for women. The Internet does, however, make threats, stalking and harassment all too easy to dish out, and too difficult to prosecute.
In Catherine Buni and Soraya Chemaly's piece in The Atlantic on safety and social media, they highlight two studies: The first found that 20 per cent of misogynist tweets were of a threatening nature. Another Demos study reported that even though female journalists, male politicians and male celebrities face the most online abuse, women are more likely to be targeted because of their gender; and men do the majority of the harassing.
As the two authors note: "For women of color, or members of the LGBT community, the harassment is amplified." Indeed, in this context, some #gamergaters have allegedly been lurking the dating profiles of LGBT developers, searching for controversial information to be used against them.
In the midst of all this, Twitter announced yesterday that the company is "enhancing" functionality around harassment reporting and blocking. The blog post states: "we're improving the reporting process to make it much more mobile-friendly, require less initial information, and, overall, make it simpler to flag Tweets and accounts for review. These enhancements similarly improve the reporting process for those who observe abuse but aren't receiving it directly."
These changes are just rolling out so it's difficult to comment on their effectiveness, but these seem to be positive changes. One major criticism of the reporting feature was that users couldn't do so if the harassment wasn't directly happening to them, so this will improve users' ability to flag accounts for review.
Will it really be a "safer Twitter?" Maybe. Improved reporting and blocking functionality are good, important changes. But what we actually need is so much more than this. Simply removing tweets from one's mentions (which is a manual process, usually) doesn't make them disappear, and suspending accounts don't address the harmful behaviour in a substantial way. While Harper's ggautoblocker has reduced the likelihood of some abusive messages showing in her mentions (and those who subscribe to it), it hasn't stopped people from showing up at her workplace and tweeting about it.
A lot of harassment happens online, but it can also extend into the physical realm. Women who are vocal online know this all too well. Anita Sarkeesian cancelled a talk at Utah State University after the school received an e-mail threatening a "Montreal-Massacre style attack." Both she and games developer Brianna Wu left their homes due to threats, and I'd bet many more people have had to do the same.
The substantial change that we need – to end to harassment in general – goes far beyond automatic or easier blocking, and even further beyond Twitter or other networks. For now, I guess we'll take what we can get.