I first came across Weibo last month, while tutoring a Chinese graduate student at the writing center. In a paper, she’d conflated some of Facebook’s functions with twitter, by referring to Weibo as the “Chinese Facebook.” Weibo was a hybrid of both Facebook and Twitter, which had caused the confusion. It showed up in other papers by graduate students–I go to Boston U’s J-School where a lot of international students are Chinese — but it never made me think twice. I didn’t even google it, probably because I don’t expect to ever use Weibo. Then, yesterday, I saw this.
What first grabbed me was not Chinese Internet censorship, which is hardly news, but the fact that Propublica had actually taken the trouble to format things properly….with an actual background. In previous posts I’ve trashed Propublica for its formatting, but here was an example that proved me wrong. (Though, to be fair, this was uploaded five days ago.) Here’s the header:
ProPublica has been collecting images that have been deleted by censors from Sina Weibo, “China’s Twitter,” since May. We gathered a team of people proficient in Mandarin to read and interpret 527 deleted images collected during a two-week window this summer. The images provide a window into the Chinese elite’s self-image and its fears, as well as a lens through which to understand China’s vast system of censorship. If you work as a Weibo censor and are willing to speak to ProPublica about your experiences, please contact at firstname.lastname@example.org
Right after that, in parentheses, a PGP KEY is provided. This is followed by a ‘Related Story’ and ‘Methodology’. The links have some redundant information that is already on the project page. This suggests that Propublica does not expect the project page itself to be clicked on that much. So the the other pages, which are more in line with the usual propublica style, link to the project page. It’s really there to lure regular propublica users to the project page. That being said, there is novel information scattered in these extra pages. The methodology, specially, deserves a page by itself. Here’s an example of the thoroughness:
Since the goal was not to collect a representative sample of all Sina Weibo posts, but to collect as many censored posts as possible, researchers assembled a subset of users who had previously posted content that was later censored. They picked 50 users from accounts found on WeiboScope, a University of Hong Kong project that archives deleted content on Weibo. They then added another 50 users, finding members of the first cohort’s Weibo circle who posted similar content.
All 100 users have a minimum of 2,000 Weibo followers. Researchers sought out users who said they were journalists or lawyers. Members of these professions, in the judgment of our researchers, are more likely to exhibit behaviors that would cause them to be censored, such as vocalizing social criticism and posting messages about human rights violations online
The images on the project page are pretty evocative. But there are some for which cultural context is required to appreciate them. For example, one contrasts Obama bumping fists with a janitor, with a Chinese official surrounded by a group of women. The caption reads “”One photo comments on civil society versus people’s society.” And I have no idea what this means. Perhaps I should ask one of my Chinese students.
I’ve always wondered how internet censorship in China works, but never bothered to find out. Here, I learned bits about “automatic filtering technology.” And how people can get around it. This would be pretty difficult to do in English.
We assume that Western democracy allows free speech etc, so this could never happen. Except, I’ve had a couple of posts magically ‘erased’ from facebook within minutes of my posting them on somebody’s wall. In both cases, it was a good friend’s wall and I checked with him. He hadn’t even seen it. I won’t broach what the posts were about here, but I think this kind of filtering technology is way more effective than the Chinese one.
In conclusion, considering that Propublica says it has been “watching the watchers” for five months, I still want to know why it only published data for this two week window.It says a lot of posts during this time had political content, but I would want to know what was significant about this period: what was going on in China at the time….
- How to Get Censored on China’s Twitter (acenewsservices.com)
- Sina Weibo making more money than ever, ad revenue up 125% (techinasia.com)
- These Images Were Censored From China’s Sina Weibo (animalnewyork.com)