Onlinecensorship.org launches second report, "Censorship in Context" (PDF)
by Jillian C. York | Nov 16 2016
Onlinecensorship.org is pleased to share our second research report, "Censorship in Context: Insights from Crowdsourced Data on Social Media Censorship." The report draws on data gathered directly from users from April to November 2016, and covers six social media platforms: Facebook, Flickr, Google+, Instagram, Twitter, and YouTube. In the report, we also look at the increasing number of media stories surrounding this issue.
Here are some of the highlights:
The vast majority of the reports (82%) related to English language content. Spanish was the second most frequently reported language at 3%, followed by German (2%) and Bengali (2%).
Many users (44.7%) reported trying to appeal the takedown of their content across platforms.
We saw an increase in complaints of politically-motivated censorship, much of it pertaining to the United States election.
This report covers 230 submissions, 38% of which originated from outside the United States.
We collected 294 reports of content takedowns in the media during the past year. Of those reports, 74% pertain to Facebook. 47% across platforms relate to nudity-related takedowns.
Our first report, released in March 2016, centered on trends in censorship on social media platforms. In this paper, we focus on providing insight from our user reports to develop a concrete accountability framework for social media content moderation.
Since our launch one year ago this week, Onlinecensorship.org has sought to provide an objective, data-driven voice in the conversation around commercial content moderation. Our small team—which includes staff members of EFF and Visualizing Impact as well as independent researchers—has continued to intensify our focus: We have released two reports; enabled users to report to us in three languages; expanded our outreach to include weekly media tracking, and have provided insight to a number of prominent publications.
Aug 23 2017
We're thrilled to announce that we'll be working on a new project to create a consensus-based priority list of the information users and researchers need to ...