China's soft censors clamp down on micro blogs
'Weibos' riddled with 'harmful' rumours
New research by a US university has shone a little more light on the murky world of Chinese web censorship to tell us that, yes, home grown social media is policed pretty damn effectively in the People’s Republic.
In what they claim is the first study of its kind into “soft censorship” – that is, deletion of individual messages rather than the wholesale blocking of sites - researchers at Carnegie Mellon University's School of Computer Science analysed millions of Chinese microblog, or weibo, posts to uncover which terms drew the ire of the censors.
The team collected some 57 million messages posted on the popular Sina Weibo platform from June to Septmber 2011 using a developer API. A few months later it then checked a random subset of these messages and then another subset containing known politically sensitive terms, to see which had been deleted.
According to the team, a combination of automated technology and manual labour ensures that politically sensitive terms are deleted. Some, like outlawed spiritual movement Falun Gong or human rights activists Ai Weiwei and Liu Xiaobo, are likely to be flagged straightaway while others are dependent on circumstances and context.
For example, the term 'Lianghui' is normally a legitimate reference to the meeting of the National People's Congress and Chinese People's Political Consultative Conference, but in February 2011 became used as a code word for planned protest and thus was censored, the researchers explained.
On another occasion normally acceptable references to Communist Party hero Jiang Zemin were censored after rumours of his death circulated in early July 2011.
At the height of the rumours, the researchers compared mentions of his name on Sina Weibo – one in every 5,666 – with the Chinese language version of Twitter – one in 75 tweets – to highlight that messages were indeed being deleted.
They added that the authorities focus particularly on areas of known political unrest such as Tibet, where half of all messages generated locally were deleted during the research period.
The report continues that not all censored content is political, with the authorities also cracking down on web rumours of contaminated salt following the Japanese nuclear incident at Fukushima.
If anything, the research is notable for proving that state censorship of the web in China is not yet completely 100 per cent effective. Although the Great Firewall does a pretty good job of blocking sites on the blacklist, some weibo messages are still slipping under the censor's radar, although it will surely not be long before that is righted.
The government's latest ploy to discourage the posting of any controversial content is to mandate that users sign up to weibo accounts with their real names - something Sina thinks could lose it 40 per cent of its punters.
Under Hu Jintao’s leadership the country has seen a definite online crack down on free speech, with regular purges of web sites deemed to be hosting fraudulent, pornographic or "harmful" content.
This came most notably in November 2011 when first China’s major tech companies were
bullied into encouraged to remove any content deemed harmful to the state, and then journalists were given strict reporting guidelines designed to discourage them from reporting on stories circulating on social media.
Twitter is of course well placed to enter the Chinese market should the government deem it fit after launching functionality which allows tweets to be blocked at a country level if they don’t conform with local laws. ®
Sponsored: Benefits from the lessons learned in HPC