Users raise concerns as Facebook discreetly censors a few posts

Several users' posts concealed behind an obscure message saying 'attachment unavailable'

On a January morning, Ishan Bhamri woke up to his Facebook newsfeed populated with veiled posts from his friends. Veiled, because these posts carried an obscure message: 'Attachment Unavailable'; in place of what might have been a status update, image or something shared from another Facebook page. 

Bhamri, a New Delhi-based user, wondered whether he was the only one who couldn’t see the posts.

He later wrote on Facebook: “Have you noticed something weird going on FB today? A lot of the stuff people are sharing is coming as 'attachment unavailable' on my feed and I can't see it. So I used a VPN to check if it was being controlled by the Indian service providers. Answer was pretty apparent.”

Bhamri could see the posts after he activated a Virtual Private Network, or VPN service, that masked the location of his device to a country or region other than India. “And there you go, censoring content, eh?” he wrote.

Another user, Sandeep Kaur, who stays in British Columbia, Canada, was perplexed when some of her posts were concealed from the newsfeeds on her Indian friends. “It’s crazy they can’t see the posts. Why would they (Facebook) censor when all I shared was memes and random articles,” Kaur told this reporter. One post Kaur shared as a screenshot that she claimed her friends in India couldn’t see (and still can’t). The post was a science joke from a Facebook page called Physics-Astronomy.com.

A similar situation was met by Riya Trehan, a design student in Melbourne, Australia, whose posts – a meme from one Hilarious Texts Facebook page, another from a page called Postize and even a link from BuzzFeed Australia Facebook page – appeared as “attachment unavailable” on some Indian users’ newsfeed. Trehan pointed out that she did not receive any communication from Facebook that her posts were removed, or whether they violated any guidelines in the first place.

(Nut-graph here) The above examples point to an indeterminate form of censorship Facebook is employing at a time when the world’s largest social media company is under global scrutiny over the user-generated content that flows on its platform. Facebook may have put in place restrictions that limit certain content in certain geographies, or, access to content by certain users-- and is rolling out these changes discreetly.

Facebook did not comment on this story till press time.

The manner in which Facebook algorithms work is opaque, said Karthik Balakrishnan, a trustee at Internet Freedom Foundation, an Internet rights group.

There are two ways Facebook removes content: pre- and post-. If a user flags a post as inappropriate, the platform reviews it and may or may not take it down. It also does it preemptively where its own algorithm labels the post potentially inappropriate and sends it to a review team. 

“It is possible that Facebook is flagging the user and not the content. Perhaps, a particular user can't see the content because he or she is more susceptible to sharing sensational content so Facebook doesn't serve it to this user,” said Balakrishnan.

Users pointed out that what is worrisome is that many don't discover if their posts are censored to some users. A feedback informing the said users that their content were removed because it violated guidelines was missing in these cases. Users on Reddit, a platform for the tech and software community, also pointed to seeing a lot more of “attachment unavailable” messages on their newsfeeds.

“It’s difficult to make out whether it’s deliberate intervention or a programmatical error,” said Abhishek Baxi, a New Delhi-based digital consultant. “Perhaps it is some content marked as offensive by users or their own algorithm in an incorrect manner. But (content) being visible in one country and not in the other points to deliberate intervention by Facebook to block off some content.”

After confessing that the Facebook platform was manipulated by Russian actors to influence US presidential elections in 2016, the firm has doubled down on efforts to sterilise the platform from misinformation, hate speech and other forms of hazardous content.

Facebook was also the centre of controversy for allegedly abating hate speech by groups inflicting violence on Rohingya Muslims in Myanmar. Now with upcoming general elections the US and India, the task cut out for the Mark Zuckerberg-led company is to scrutinise content without going so far as curbing free speech.