At first blush this was easy to complete – go onto Taobao (淘宝网, akin to eBay+), and purchase an accurate vector graphic of a Chinese flag. The file, bought from a designer cost 150 RMB (€18, $24) – download a PDF version from here. A torrent site could have been easier, certainly cheaper – but we wanted to remain open to the designer/seller being able to adapt the design based on feedback from the crowd. The seller provided background information on the design of the flag – for example adhering to national regulation GB 12982-2004 of the National Institute of Standards for making a Chinese national flag.
As any consumer in China knows – what you buy and what you get is not necessarily the same thing – so we set up a quality control sub-task using a Chinese Wang Zhuan site AliPay (支付宝, akin to Amazon’s Mechanical Turk – Wang Zhuan literally translates to ‘earning online’).
For the quality control sub-task we asked 20 people to challenge the “correctness” of the flag graphic – if they found anything wrong they were rewarded a nominal sum of 4 RMB (€0.5, $0.6). A total of 100 RMB was allocated to the task that would include the 20% commission taken by the site.
Initially the crowd sourcing site Zhu Ba Jie rejected the task, because the task description included words considered sensitive. The task was rewritten and posted to crowdsourcing site San Da Ha replacing e.g. words like ‘national’ with ‘flag’ – and the task was accepted. (We also considered crowd sourcing site Renwumatou 任务码头, Task Harbour as a third option). After posting 20 people completed the task – with no-one challenging the flag’s technical accuracy, but a few people questioning its colour. This was not surprising – while a flag layout is tightly specified, colour is normally specified as a range depending on what kind of material it is printed on. As one of our crowd sourced responses pointed out that the way to verify the colour was to “cut your finger and compare with the colour of blood”.