Thoughts on China’s Web Users: Part 1

I just spent the last couple of days in Beijing, working with the wonderful folks from Mozilla China, meeting with the major tech players, and covertly observing user behavior with other UI researchers in internet cafes. It was on of the most intense learning experiences I’ve been through in recent memory.

This is a brain dump, in no particular order, or some of the observations from the trip.

Slouch Browsing

The ineffable Mike Beltzner made roughly the same pilgrimage a month or two ago. A large portion of browsing in China seems to take the form of “slouch browsing”: one hand on the mouse, other hand dangling, slouched back in the chair. It’s a mouse driven world. Mike put it well, so I’m just going to quote him:

“Even the posture of the average Chinese internet user is different. In North America, people no longer ’surf’, but sit engaged with their computers, interacting with the internet & pages, typing, searching, creating. In China, users usually ‘urf’ one-handed. Sitting back, clicking, watching, hopping through a web of links, not searching or navigating on their own. It’s a bizarre difference.

I was told again and again (and certainly observed) that people don’t search as frequently in China as they do in the US and Europe. A lot of this stems, I believe, from the clumsiness of the input method (the IME), but I’ll come back to this in another blog post.

Entertainment Web

The web in China is not one of work. It’s about entertainment (which certainly explains the slouch browsing posture — although which begot which is something of a chicken-and-egg problem). Baidu, Google, and many others have MP3 searches (that include downloading!). Web TV — on demand movies and shows — is ubiquitous. I saw numerous people reading novels, late into the night (it’s only $1.75 to spend the night in an internet cafe), by the lonely irradiating glow of phosphorescent screens. And, of course, massively multiplayer games are all the rage. The internet cafes even have multi-terabytes of streaming movies and TV shows (providence unknown) stored on in-house servers accessible from garish and constantly moving web pages. Even with a Chinese Google UX researcher sitting by my side (thanks Jinghua!), it took us five minutes to figure out how to wade through the right incantation of clicks to make one play. There’s the juxtopostion of people refusing to wait for content, but yet are willing to put up with all sorts of interface travesties.

People first: One chat to rule them all

The chat conglomerate QQ is it’s own rather large universe in China, and it has a business model that puts all of our cute social networking sites (Facebook, I’m looking at you) to shame.

QQ started out as chat, but no includes it’s own web browser, online television, movies, music store, voip and video chat, screenshot/cast software, and the rest of the kitchen sink. It’s, of course, the browser that first caused my brows to furrow. Getting into the browser business using chat as the thin edge of the wedge seemed backwards. It isn’t, though. In some ways, we got it backwards.

Chatting is about people first — you communicate with those around you that you care about. As a surrogate for the shared experiences you can have by being physically collocated (i.e., watching TV together), QQ has been enabling shared virtual experiences. When chatting with someone, it makes sense you would want to listen to the same thing, browse to some of the same places, and comment on a movie while watching it together.

Whereas chat is fundamentally about people with information tacked on, the Web is fundamentally about information with people tacked on.

We should be thinking about how to make people more central to the Web (and Firefox in particular) as we move forward, especially in Asia. QQ has an enormous market penetrations both on computers and cell phones. And because it is centered around people and communication, a strong sense of identity and “brand” has grown up around it. For instance, QQ users are identified by a number instead of a more memorable login name; a low QQ number is socially worth more than a high QQ number which means people sell their low QQ numbers for not an insignificant amount of money.


QQ has a large number of contributors that aren’t in the company. For them, I was told a couple times, standard open source incentives aren’t enough to act as motivation. Working on an open source project just doesn’t have the same resume weight there as it does here. Open source isn’t considered “professional”. And the social reputation you gain from your peers doesn’t have enough sway. I’m not sure how much of this I believe, but it seems to be a common meme.

What QQ does to incentive its community is rather weird. The normal things they do is give contributors special VIP standing — special logins and special abilities. The abnormal thing they do is give those VIP folks the ability to access other people’s private data. Scary.

Privacy paradox

Privacy in China is one odd duck. From what I can tell, it is important, but only locally.

People are super concerned with Looking-over-the-shoulder privacy. Every time a computer is logged out of in an internet cafe, it restores the OS to a pristine state by (roughly) reinstalling. Digital ID theft — Game logins, QQ chat id, etc — is prevalent because the sale of pilfered digital goods isn’t really prosecuted. In fact, the “clear private data” button is so important that it is given the same visual primacy and placement as the back button in the QQ browse (yes, chat has its own browser).

On the other hand, people don’t care about their global privacy. What level of data is being kept by a service, and how it is being used, isn’t relevant in the choice of product. The huge amount of care we are putting into guaranteeing the security and privacy of data for Weave is mostly irrelevant for the Chinese market. In part, this is because the Chinese Government is pushing for a 1 to 1 correspondence between a user and their identity. In other words, stripping the layer of anonymity from the Internet. Thus, almost anything looks better than that in comparison.

The privacy paradox is that privacy maters locally but not globally. Out of sight, in this case, is out of mind.

Comments are closed.