Ensuring the right kind of content is posted on your community is an ongoing challenge for community managers. There exists a fine line between good moderation and censorship.
Are there any lessons that can be learnt from the worlds biggest user of social networks, and indeed the biggest censor of content? Some recent research by Rice University has delved into the complex world of Chinese online censorship.
The research looked in particular at the speed with which censorship occurs on the Weibo social network. They found that the site uses a combination of keyword matching software and human censors to both monitor and delete controversial posts from the site.
What’s more, if an individual develops a habit for posting controversial content, they quickly become monitored more closely. The research revealed that most deletions occur within five minutes of them being posted, which is crucial to stop other people sharing the message and things spiralling out of control.
For those not familiar with Weibo, they’re a microblogging site akin to Twitter, complete with 140 character messages and hashtags. Around 300 million people currently use the site, which compares with the 175 million or so registered Twitter users.
The research began by tracking 25 users who had already been flagged for posting content on the Weibo banned list. They then broadened their search by tracking another 3,000 users who had retweeted one of the original 25′s posts more than five times.
With this sample in place, the 3,025 users were monitored for a period of time to determine both how often they had posts deleted, and how quickly the content was zapped.
The researchers found that over 4,500 posts were deleted from this group each day. The researchers found that the majority of posts were deleted within an hour of being published, with 90% of deletions occurring within 1 day.
The analysis also revealed a sophisticated mechanism to remove all reposts of deleted posts, often within five minutes of the original post’s deletion. Deletion times were found to be significantly shorter for a subset of users who tended to post deleted content most often, an indication that the site actively monitors the activity of some users.
“Roughly 12 percent of the total posts from our sensitive users were eventually deleted,” the research team says. “We have enough of these posts to be able to run topical analysis algorithms that let us extract the main subjects that Weibo’s censors seemed concerned with on any given day.”
Whilst few networks exhibit the kind of censorship seen on Weibo, the research is useful in providing an understanding of a topic that is likely to grow in importance on social networks wherever they may be.
“There has been considerable debate in the US recently about extending copyright law enforcement to include various kinds of filtering online. China already has laws in place for companies within China to filter online content.” says Jed Crandall, assistant professor of computer science at University of New Mexico.
If censorship does become more commonplace, then sites like Weibo will increasingly be looked at for inspiration, as they battle the dual concerns of keeping ad revenue high whilst also complying with local laws.