This is not what I had wanted to write about for my first post of 2014, but unfortunately it’s necessary—so much so, in fact, that I felt the obligation to crosspost both here and on my not-so-super-secret other blog in order to get this information out to as wide a readership as possible.
I’ve always had a bit of a love-hate relationship with Facebook. On the one hand, I like how easily it lets me stay in contact with family and friends across the country, people whom I would rarely see more than once or twice a year, if even that. On the other hand, I have the same privacy concerns that many other people have with respect to putting personal information, as well as pictures and videos of myself, family, and friends, onto Facebook. Now that I’ve become a (sort of) public figure (or, as I like to refer to myself, a micro-celebrity), I’ve thought that I should cull my friends list to just real friends with whom I have a connection (or at least have met in person or had private e-mail exchanges with) and set up a Facebook page for my public persona, to prevent people whom I don’t know or barely know from divebombing my wall with arguments. As I tell people, I don’t want obnoxious arguments on my Facebook wall; that’s what my blogs are for.
My personal social media preferences aside, Facebook does indeed have many shortcomings, but until something else comes along and steals the same cachet (which is already happening as teens flee Facebook to avoid their parents) and even after, Facebook will remain a major player in social media. That’s why its policies matter. They can matter a lot. I was reminded of this about a week ago when Dorit Reiss (who has of late been the new favored target of the antivaccine movement, likely because she is a lawyer and has been very effective thus far in her young online career opposing the antivaccine movement) published a post entitled Abusing the Algorithm: Using Facebook Reporting to Censor Debate. Because I also pay attention to some Facebook groups designed to counter the antivaccine movement I had already heard a little bit about the problem, but Reiss laid it out in stark detail. Basically, the merry band of antivaccinationists at the Australian Vaccination Network (soon to be renamed because its name is so obviously deceptive, given that it is the most prominent antivaccine group in Australia, that the NSW Department of Fair Trading ordered the anti-vaccine group to change its misleading name) has discovered a quirk in the algorithm Facebook uses to process harassment complaints against users and abused that quirk relentlessly to silence its opponents on Facebook.
Over the weekend of December 21-22, an unknown person or persons used a new tactic, directed mainly at members of the Australian organization “Stop the Australian Vaccination Network” (The Australian Vaccination Network – AVN – is, in spite of its name, an anti-vaccine organization – see also here; SAVN had been very effective in exposing their agenda and mobilizing against them). In an attempt to silence pro-vaccine voices on Facebook, they went back over old posts and reported for harassment any comment that mentioned one person’s name specifically. Under Facebook’s algorithm, apparently, mentioning someone’s name means that if the comment is reported it can be seen as violating community standards. Which is particularly ironic, since many commentators, when replying to questions or comments from an individual, would use that individual’s name out of courtesy.
Several of the people so reported received 12-hours bans. Some of them in succession.
The only common trait of these posts appears to be that they use the name of an AVN supporter, who then uses that use of her name as evidence of “harassment” to report the post. If you go back and look at the examples of comments that were used to trigger 12-hour Facebook bans, they are about as innocuous as one can imagine. For example:
Here’s another example:
Karen; a challenge for you: find a case where a judicial officer (judge, special master etc) has heard evidence for and against a link between vaccines and autism and then found in favour of a link. Give me a link to the judgment, not what what [sic] Mike Adams or others claim is in the judgment.
Antivaccinationists, being antivaccinationists, also can’t resist bragging about what they are doing, posting images of comments successfully removed and creating a Facebook page called “FB Time-Outs for Provaxers”:
One notes that this must either have been taken down or that it is a private, invitation-only group that doesn’t show up on Facebook searches. This is still going on. Indeed, I received an e-mail on New Year’s Eve from Joanne Benhamu, who confirms that this has been going on several weeks and further reports that the individuals responsible for this campaign have been on a reporting spree. She also reports that a couple of days ago every administrator for the “Stop the AVN” Facebook page, including Joanne Benhamu and Rachel Dunlop, has become the target of such bans. The effect, as described by Benhamu, has been to disable their ability to run the Stop the AVN Facebook page and prevent supporters from engaging in debate about the topic. She also gave examples of the frivolous nature of the complaints and the innocuous nature of the comments singled out for complaint. One comment that drew a 12-hour ban was: “Tess, are you disputing that your only source is a blog penned by a known liar?”. Another comment that has resulted in Dr. Rachael Dunlop being banned is: “Karen is a sock, or appears to be.”
I can’t emphasize enough how hypocritical this is.
Reasonable Hank has also pointed out the brain-melting hypocrisy of the AVN and documented several instances of the AVN crying “Censorship!” to raise money and quoting Meryl Dorey herself saying “Censorship is NEVER acceptable!” It’s a viewpoint that I actually have a great deal of affinity for. Remember, I’m the same person who has lambasted laws criminalizing Holocaust denial when arch-Holocaust denier David Irving ran afoul of such laws in 2006.
Of course, government censorship is a different thing from the policies of a private company like Facebook. Facebook can do what it likes and set whatever policies it likes. It is Facebook’s policies that are problematic. As easily abused as Facebook’s policies appear to be, however, it is hypocritical in the extreme for Meryl Dorey to weep crocodile tears about “censorship” and then engage in the activities Hank documents, such as frivolous DMCA takedown complaints and intentionally marking unfavorable posts as spam. Reiss and Reasonable Hank provide compelling evidence that this is a deliberate strategy, most likely on the part of the Australian Vaccination Network and Meryl Dorey, designed to abuse the reporting algorithm used by Facebook in order to give pro-vaccine skeptical activists a “time out” from Facebook. During these 12-hour bans, they are unable to post or comment; they are, for all intents and purposes, locked out of Facebook. As Reasonable Hank explains, this is not a petty Facebook squabble. Remember that each of these reports means a 12 hour ban from Facebook for the user. It is also highly unlikely to be a coincidence that comments from months ago are found and reported at a time coinciding with the end of a user’s 12 hour ban. One user has reported multiple bans, one per day, so that he has not had continual access to his Facebook account for over a week. Just look at the list of comments that led to 12-hour bans based on frivolous reports of harassment and abuse. There is nothing abusive or threatening about any of them.
Again, the only thing these posts have in common is that they use the name of an AVN supporter, often only the first name. Apparently, yesterday Meryl Dorey herself finally commented, playing dumb in a transparent fashion:
After a disingenuous bit about how this is the first time she’s ever heard of this and a claim that she has yet to see any evidence that AVN-affiliated Facebook users have anything to do with this and a “nudge-nudge, wink-wink” urging that “we are better than this”, Dorey turns up the nudging and winking to the point where she probably has elbow bruises and corneal abrasions to conclude:
Oh, and honestly, pages and people who breach Facebook’s terms and conditions should be reported – don’t think that I am trying to stop you from doing that, I encourage you to! But please keep it to those types of pages and posts.
In other words, carry on what you’re doing, but please, just be a little less blatant about it and for heaven’s sake don’t publicly gloat about it when your abuse of Facebook’s reporting mechanism succeed!
Again, this is a deficiency of Facebook’s automated reporting mechanism. It’s a glitch that’s so transparently open to abuse that it’s a wonder to me that Facebook hasn’t fixed it already. As one of the commenters in Reasonable Hank’s post points out, it seems fairly certain that the name recognition is automated to the point that the person doing the reporting of abuse need only have an account with the same name or partial name as the name mentioned in the comment in order to report it for purposes of suppression, elaborating:
So if I happen to see some guy on a US politics site saying stuff I disagree with to some other guy named “Andy”, it seems I can get the abuser banned even though it’s not me he’s abusing. The target “Andy” would be none-the-wiser.
Gosh, if were a troll, I could just spend my days searching random FB sites for anyone who mentions “Andy”, and get the users banned for the hell of it.
The Facebook account most used to make these spurious complaints appears to belong to someone with the ‘nym Karen Little, who runs an antivaccine Facebook page in which she blatantly gloats about getting Australian skeptics banned, such as Peter Bowditch, and posts memes like this:
Unfortunately, Facebook has not proven itself to be particularly good at dealing with issues like this; its track record leaves something to be desired. However, with its more than one billion users, Facebook is still the 800 lb. gorilla of social media and is likely to remain so for the foreseeable future. It is a platform that can’t be ceded to the antivaccinationists without a fight. Moreover, if the Australian (Anti-) Vaccination Network succeeds in driving pro-science activists from Facebook with these tactics, other antiscience movements will be emboldened to do the same. As I’ve explained time and time again, attacking the person and trying to silence him is a feature, not a bug, of crank movements. As I discovered early on, they hate pseudonyms and would try to “out” me every way they could. I learned the reason, too, and that’s to target me personally in order to silence me. As I’ve experienced all-too-many times as various cranks have tried to harass me by complaining to my superiors at my job or even gone as far as to try to report me to my state medical board, which a misguided Stanislaw Burzynski supporter did last year. This is no different, except that it’s easier. My bosses know enough now to recognize crank complaints, and the state medical board rapidly decided that the complaint against me was bogus. Facebook has an automatic algorithm; human judgment appears to have very little to do with it.
So what can we do? First of all, this needs to be publicized. Whether it’s the AVN doing this (as seems very likely based on strong circumstantial evidence) or other groups of antivaccine warriors, this needs to be publicized. I don’t know whether complaining to Facebook will do any good or not, but it nonetheless has to be done at the Feedback link. It might also help to report this as a bug. The key is that we have to get a lot of reports; otherwise it’s unlikely that Facebook will notice. In the meantime, the only other suggestion I can think of is never to use the name of a known antivaccine activist, in particular the names above that were used to get people temporarily banned. It’s a matter of self-preservation.
Social media sites like Facebook are a very useful tool for community building and disseminating information. However, the can be abused, and that is what appears to be occurring here. There might come a time when Facebook ceases to be useful because its reporting algorithm is too easily abused. Here’s hoping that the management of Facebook can be made aware of that.