Logging on in China’s Internet Cafés

An HRIC Field Survey¹

China’s reputation as the second most “wired” country in the world conceals the vast access gulf that exists between the urban elite and less advantaged groups.

With more than 100 million users online,² China is now second only to the United States in the number of its people accessing the Internet. Given current growth rates, this number is expected to reach 300 million by 2008.³ The online population today is fueling Web sites, community forums, blogs, online games and more—in recent months, estimates for the number of bloggers have reached as high as five million,⁴ while online gamers have been placed at 20 million.⁵ However, the dominant media focus on the exponential growth in online netizens neglects the reality of China’s digital divide and its implications for equitable distribution of access to technology, information and an expanding virtual public space.

More than 25 percent of China’s online population depends on Internet cafés as an online access point.⁶ Outside of large urban areas, among China’s most vulnerable groups—adults and children of migrants, rural populations and ethnic minorities—low levels of home computer ownership mean that 80 percent of Internet users are able to access the Internet only through Internet cafés.⁷ Away from the more affluent coastal regions of Beijing, Shanghai, Tianjin and Guangdong,⁸ Internet cafés become the only resource available for many users.⁹ The six most underdeveloped provinces comprise less than one percent of China’s entire Internet population;¹⁰ in contrast to the wealthy, urban elite who have easy access to Internet-enabled home, office and even school computers, the economically disadvantaged have to rely almost exclusively on Internet cafés for online communication, information, and expression.

Internet café closures¹¹ and the stiff, complex maze of regulations that govern these cafés,¹² disproportionately impact population groups who are not among the wealthy, urban elite. For these groups, Internet cafés represent their most practical, and perhaps only, point of access to the Internet.¹³ In an effort to better understand the situation regarding access to online information that these groups are experiencing, and the actual environment of China’s Internet cafés, both physical and virtual, Human Rights in China (HRIC) conducted a limited preliminary field survey of thirty-five Internet cafés in eight cities across the central southern and eastern regions of China between July and August of 2005.

This field survey describes the availability and locations of cafés surveyed; software and hardware installed, including censorship and surveillance software and practices; and user demographics and ambiance inside the cafés. Although there was some disparity in technology, for the most part, the cafés encountered possessed excellent hardware and Internet connection speed regardless of socio-economic location. Unsurprisingly, online gaming and chatting comprised the most popular activities, but not all cafés were devoted to gaming or to the traditional younger male clientele. The survey found that, in general, the official rules and regulations that govern Internet cafés were enforced haphazardly and unequally—with the notable exception of an almost universal absence of minors in the cafés. However, the official oversight lacking in physical form was present virtually, as monitoring software was installed on all computers surveyed without exception.

Methodology and scope of survey

Internet cafés selected to be included as part of the field survey were located through the one or more of the following methods: 1. Using public transportation networks (primarily in Nanjing, Zhengzhou and Changsha); 2. Inquiring with local shopkeepers and vendors for locations; 3. Walking on city streets (primarily in the more compressed city center areas such as Xi’an, Guiyang and Nanchang); and 4. Targeting high probability sites such as transportation hubs, commercial centers or tourist areas.

Table 1: Overview of Internet cafés visited and surveyed.

Most cities contained many more cafés than the number actually surveyed. Cafés in close proximity to each other or those passed in the course of general transit around the city were generally not surveyed independently.

Once a café was located, the surveyor would request a computer, and unless directed to a specific machine, would select a terminal sufficiently far into the café to afford a better view of the actions of other patrons. If possible, a machine facing the door was chosen in order to allow additional observations of patrons entering and leaving the café. For each café, the surveyor spent approximately one hour browsing the Internet, checking specifically for access to the HRIC Web site, a Web site with several simple Flash games, Google and GMail Web sites, and various politically-oriented Web sites. This was followed by an exploration of the machine’s hardware and software and a check of bandwidth before the results were uploaded to a secured web location.

Although the Internet café experience of an obvious outsider is inevitably different from that of locals, it is still possible to observe and draw valid conclusions. Technical considerations do not change, but culturally, the reaction towards an outsider or foreigner is entirely different from that directed at a Chinese citizen. With this in mind, it was critical to pay more attention to the surroundings than the experience; for example, watching the front desk to see what was asked of people entering and leaving, noting the uses other clients were making of the computers, and observing the attention being paid towards them by café employees or official monitors.

Availability and locations

Most cities have Internet cafés widely available regardless of neighborhood. While the majority of cafés surveyed were located in mid- to high-end commercial neighborhoods, numerous cafés could also be found in low-end commercial and residential neighborhoods.

Table 2: Overview of the most popular games encountered during the survey.

The “genre,” type and quality of cafés tended to differ based on location; those in higher income neighborhoods possessed more modern hardware and software, as well as generally more comfortable surroundings than those located in less prosperous areas. Cafés found in low-end commercial and residential zones tended to fall into two main categories. Although highly similar in appearance, their target audience/patronage was widely diverse.

The first consisted of gaming cafés, frequented primarily by younger adults, with computers optimized for games (e.g., higher memory/CPU).The second consisted of cafés frequented primarily by older patrons. The computers available were often far inferior to those in the typical café used for web surfing and online chatting. These computers were more likely to have minimal gaming software installed, usually only the first person shooter Counterstrike¹⁴ (a staple fixture in every café), as well as a collection of simple puzzle-type games.

Software

Gaming

A primary use for Internet cafés, and the source of much of their clientele, is gaming—particularly online MMORPGs (Massively Multi-player Online Role-playing Games). With online gaming a staple activity in Internet cafés, China’s gaming industry has profited extraordinarily from the proliferation of the Internet, and its success has come to the notice of many market investors. Revenue from online games reached $298 million in 2004,¹⁵ and analysts expect the market to grow to $1.3 billion by 2009.¹⁶ Shanda Entertainment, China’s leading online game operator, controls 60 percent of the domestic market and recorded revenues of $132 million in 2004, a growth of 116 percent from the previous year.¹⁷ That same year, Shanda was rated the world’s highest valued online game operator,¹⁸ supplanting Taiwan gaming company NCSoft.

Not to be overshadowed by the online entertainment world, China’s home-grown search engine giant Baidu.com generated a media frenzy with its August 2005 listing on the Nasdaq stock market. The most widely-used search engine in China,¹⁹ Baidu.com ended its first day on Nasdaq with a 354 percent increase from its $27 IPO price.²⁰ Hailed as the next Google, Baidu.com’s story is only the latest and most prominent to emerge from China’s rapidly developing Internet landscape. The flush of excitement that companies such as Shanda and Baidu.com have generated only highlights the stark disparity illustrated by the very real and very rapidly increasing digital divide that China is facing today.

At the time of this survey, the most popular of the MMORPGs was World of Warcraft, which could be found on nearly every café computer. One month after its official release in China, Blizzard Entertainment issued a press release claiming more than 1.5 million Chinese subscribers, compared with two million subscribers in North America, Europe, Australia and Korea combined.²¹

World of Warcraft is closely tied into the general marketing culture targeted at youths. At the time of this survey, a nationwide promotional campaign was well underway, with the popular singing group S.H.E. featured prominently on advertising billboards for Coca-Cola, dressed in the costumes and armor of World of Warcraft characters. Promotional codes, given away under the caps of Coca-Cola bottles, could be redeemed for half an hour of playing time.

Unlike American MMORPGs, which generally use credit card payment systems, Chinese MMORPGs use pre-paid money cards to validate accounts.²² A money card with a value of 30 yuan purchases 60 hours of play time, which is credited to an account through the Internet. Registering a card to create an account requires providing the identification number of a government issued ID card. In addition, it is possible to connect only to game servers physically located inside China; servers from the United States, Korea and Japan were found to be inaccessible.

Other interesting games found in the cafés included the popular first person shooter Counterstrike, the fantasy MMORPG The Legend of Mir II, and various games developed in China. Some were markedly nationalistic; in one, set in Shanghai during the time of the Japanese invasion, players employ street warfare tactics to hunt down and eliminate legions of Japanese soldiers. Most cafés optimized for gaming had 20 to 30 different online role-playing games, with titles available from both foreign and domestic companies.

Instant messaging software

Without exception, every single Internet café computer encountered had installed at least one form of instant messaging software. The most popular, QQchat²³ (developed by Tencent), has spread to a level where it has even inspired stores to carry a variety of clothing bearing its logo and trademark. It was not uncommon to come across young people who did not have email addresses, but did have instant messaging accounts. Students use their cell phone’s text messaging functionality to establish times to meet online for chat sessions. Many cafés also had webcams available for video chats, which were observed to be used quite frequently.

User software

In addition to a wide variety of games and chatting software, the most common software installed on Internet café computers were Microsoft programs, including Microsoft Internet Explorer, Windows Media Player and Microsoft Word. In general, the systems did not appear designed to protect against user interference. This survey found that the user could install new controls onto the machines (for example, Macromedia Flash Player, Java and other simple application software) without running into permissions issues. The exception to this was in Guiyang, where every single café surveyed used a terminal software program to substantially limit user actions. The terminals allowed users to run programs, but denied access to the Control Panel and Task Manager, and did not allow users to manually run commands or use the right mouse button functionality on the Desktop.

Monitoring software

Two primary monitoring software packages were in visible use—PubWin²⁴ (by far the most popular) and WebWatcher. Both were designed specifically for Internet cafés. In addition to monitoring, keyword filtration and general censorship, the software also provided café specific functions, such as allowing the central console to enable and disable access, and monitoring how long each computer had been in use.

Hardware

Most computers in the Internet cafés had excellent hardware consistent with the desire to support the advanced system requirements of modern games. CPU speeds in excess of 2 Gigahertz (GHz) were common, as were computers with 512 Megabytes of memory.

Generally speaking, the machines did not have exterior access bays (for example, CD/DVD-rom drives, floppy drives), although most did have one or two USB ports open and available to connect cameras or flash drives. Flash drives, while uncommon, were seen in several cafés. Although all the computers had a variety of virus and Trojan horse programs running (suggesting generally poor anti-virus security), they generally showed no evidence of previous users. This would suggest that hard drives were wiped and replaced on a regular basis.

Ambience

Patrons

Although the Internet café clientele tilted heavily towards young male students, a significant number of women and older professionals, as well as manual workers, could be found. Rather than gaming, the older patrons were observed to be more often watching movies (which were readily available from a number of online sources) or surfing the Internet.

Minors (youths under 18) were rarely seen in the Internet cafés surveyed,²⁵ even in Qingdao and other cities, where cafés were often located in close proximity to middle schools.²⁶ In the case of Qingdao, where at least one café was located directly across the street from a middle school, the lack of patronage by students suggested active deterrents discouraging their presence.

In general, cafés were found to focus on either a younger gaming audience or an older audience. Most of the adults using Internet cafés appeared to be in their late 20’s to mid 40’s, with very few older people in evidence.

Web or café

Although the primary purpose of the cafés was to provide computer access, many cafés demonstrated an effort to live up to their derived nomenclature. Most cafés sold cold drinks from a refrigerated cooler in the front; others also had snack foods or dried noodles available for purchase. A sizable minority of cafés provided patrons with free tea or boiled water, which was periodically refilled for them throughout their visit.

Other cafés (primarily in Changsha) went so far as to offer food service. However, apart from those in Changsha, the cafés generally employed minimalist efforts that were largely ignored by the café patrons, who focused on the primary purpose of computer and Internet access.

Layout

The cafés were generally set up as one large room of tables lined with computers.²⁷ A wide area of space was often provided behind the machines, where people could (and did) congregate, with several people watching one screen. In many cafés, small, enclosed booths could be found along the edges of the main room, with four to six computers generally occupied by groups who had come in together. In addition, several cafes had separate small rooms closed off from general view.

In poorer areas, cafés had clearly been built into pre-existing spaces with minimal modifications. In Xi’an, the presence of kitchen plumbing indicated that one café had previously been a series of small apartments.

On the other extreme was a café in Changsha with no central room, but rather a series of small spaces each containing four computers and sofas, with an ambiance similar to that of an expensive restaurant.

The décor in the Internet cafés seemed to be roughly correlated to price, with cafés in more affluent neighborhoods tending to have a more comfortable ambiance. However, prices seemed to be far more closely tied to the status of the neighborhood than to the quality of the café; in poor neighborhoods, the more elegant cafés were often priced similarly to neighbors with fewer amenities.

Using Internet cafés

Pricing²⁸

The cost of using computers in the Internet cafés was generally based on an hourly charge, with prices rounded up or down to the nearest whole or half denomination. Most cafés required users to submit a deposit of 5 to 20 yuan prior to use, with the remainder refunded at the time the patron left the café. Other methods included pre-paying for a set amount of time (with no refund of any remaining balance), or simply paying post-usage. Prices ranged from 1.5 yuan to 4 yuan per hour in direct proportion to the class of neighborhood.

Although not as well-equipped as those in more wealthy neighborhoods, the quality of computers was not a significant factor for cafés in poor neighborhoods. Those locations often still possessed excellent bandwidth and modern hardware at a lower price than a café in a high-end commercial neighborhood, which might have poorer bandwidth and computers at a higher cost. The time of day was also a variable in pricing, with the peak hours of 4:00–10:00 p.m. often costing one yuan to two yuan more per hour than less popular times.

Registration policies

Despite official regulations requiring registration with a government issued ID,²⁹ registration policies at Internet cafés seemed to vary widely by region. In the north and the east, registration requirements were minimal, while in the southern and western regions they were enforced far more strictly.

Cafés that did not ask for identification often still had a registration book at the front desk, in which staff members were seen to write apparently random identification numbers and names during their free time.

Website Accessibility

HRIC’s Web site³⁰ was inaccessible from any of the Internet cafés surveyed, which was expected given the results of previous studies.³¹ The Web site for China Labour Bulletin,³² a Hong Kong-based organization that promotes independent and democratic unions in mainland China, was similarly unreachable. Web sites for The New York Times, CNN and the BBC were also tried at each café. These English-language news Web sites were almost always reachable, with the few failed attempts plausibly attributable to genuine technical issues. Prior to August 1, access to Google and Gmail was inconsistent at best, with only around 60 percent of attempts resolving appropriately. However, during the month of August up through the end of this survey, access improved considerably, with only 10 percent of attempts failing.

Interestingly, these limited results ran contrary to recent reports³³ that found access to the BBC’s Web site consistently blocked from within China. The irregularity with which Web sites are blocked, both within the span of this short survey and in relation to other reports and comprehensive studies, lends additional evidence to the theory that China’s Internet censorship is fluid and constantly evolving.

Closed cafés

As previously noted, violations of several of the regulations governing Internet cafés—particularly with regards to the presence of minors and registration—seemed to be common, apparently leaving cafés vulnerable to being shut down. In Xi’an, Chengdu and Guiyang, cafés in the immediate vicinity of the train station were closed and shuttered, while in Kunming only four active cafés were found.Three additional cafés had clearly recently been in business, but were no longer open. Given the generally busy nature of the remaining establishments, it seemed unlikely that the cafés had closed for purely economic reasons.

Addressing the challenges of access and censorship

A comprehensive understanding of the Internet café environment in China will require the investigation of significantly more Internet cafés in additional locations, such as Xinjiang and Tibet. However, this initial research effort was valuable in gaining a more informed appreciation of the operating environment of Internet cafés, as well as a first-hand awareness of the practical issues facing Internet café users in China.

Although this brief survey seemed to suggest that many of the official regulations governing Internet cafés are selectively enforced and inconsistently implemented, their existence still places all Internet cafés in a precarious situation and creates an environment that encourages self-censorship and fear. Without knowing when or where monitoring is going on, the only viably safe option for the Internet café user is to assume that all actions are being watched and recorded.

Internet cafés are an integral component in bridging the digital divide by offering access to the Internet and all its resources to economically disadvantaged groups who do not have access through their homes, work or school. When Internet cafés are forced to adhere to regulations that deter and limit their usage, China’s most vulnerable groups are restricted from the full scope of information and interaction that is so easily and readily available to the wired elite.

The PRC has employed increasingly sophisticated systems of information control, including legal regulations, social and policy controls, censorship and surveillance technology.³⁴ As documented in a recent study by Open Net Initiative, less than ten percent of the search results for terms such as “sex,” “pornography” and “nude” were blocked, but more than 60 percent of Chinese-language sites with information on opposition political parties were inaccessible.³⁵ Impeding access to Internet cafés will effectively contribute to undermining and silencing the online voices of China’s underprivileged citizens.

More research is needed to develop solutions to work towards addressing the related challenges of equal and fair access to information technologies and an uncensored and diverse virtual public space that will contribute to the strengthening of the growing civil society in China.