Boundaries and information: Sidestepping restrictions through Internet conversations
First Monday

Boundaries and information: Sidestepping restrictions through Internet

Central Asian chat sites can transcend geographic boundaries and repression of information to provide access to information from around the globe. Chat and forums users gain information that is domestically censored or restricted in traditional media and share it with off–line domestic communities. Because youth is the dominant demographic participating in chat and forum sites, these online exchanges and communities are likely to have important longer–term implications for information seeking and expectations of new and traditional media. We use cross–country survey data, interview data, and participant observation of online chat and forum sites to explore the exchange of information, emergence of opinions, global connections, and off–line implications for communities within the region. This study examines how information and communication technologies take root in a technology–emergent society like Central Asia and has important implications for understanding attitudes toward and practices of adopting technology in other developing regions around the world.


Theoretical context
Methodology for examining chats and forums in Central Asia
Patterns of general Internet usage in Central Asia
Patterns of Central Asian chat and forum usage




The collapse of the Soviet Union and the birth of five newly independent Central Asian countries occurred simultaneously with the global revolution of information and communication technologies (ICTs) that introduced the Internet and mobile phones to the region [1]. Carrying over Soviet practices, government policies throughout the region repress and censor traditional media and Internet usage and limit political, civil, and religious freedoms. Nevertheless, the diffusion of ICTs to Central Asia and the growth of regionally focused chat and forum sites open new possibilities for local populations to exchange ideas, participate in public debates, and tap into worldwide sources of information.

We examine chats and forums that focus on Central Asia’s transition processes as a vibrant area for exploring the emergence and adoption of new ICTs in Central Asia. Citizens of these authoritarian countries have few options to freely voice opinions or gain access to domestic sources of unbiased news and information. Chat and forum sites allow participants, located both inside and outside the region, to engage with online counterparts to discuss a vast array of issues pertinent to the region’s post–Soviet political, economic, and social transformations — views that are often quashed in traditional media. Indeed, since we conducted our research in 2006–2008, the number of chat, forum, instant messaging, and social networking sites available to Central Asians, and the volume of young people participating in them, have grown exponentially. This analysis enables us to better understand the emerging Internet culture in Central Asia, the linkages between online and off–line communities, and the implications of new sources of information and technology in a restrictive media environment.

Much existing research on the use of chat and forum sites investigates users and applications in relatively developed, democratic, and technologically advanced societies. As a result, these studies miss the importance of how online interactions might allow participants to transcend government repression of information. While an increasing amount of attention is being paid to blogs in authoritarian countries (Chowdhury, 2008; Kelly and Etling, 2008; MacKinnon, 2008), we maintain that the dynamics of chat rooms and forums provide a lower barrier to entry for potential participants, which is especially important in digitally emergent regions. We argue that the online information sharing in chat and forum sites is critically related to real–world social interactions and social networks. Moreover, we argue that, because of the varied geographic locations of participants in a given chat or forum site, users are able to develop global connections that offer opportunities to circumvent the repression and censorship of information by local governments. Our argument builds on the two–stop flow theory of communication (Lazerfeld, et al., 1948; Katz and Lazerfeld, 1955) or the “dual link model” (Bandura, 1986). Chat and forum sites allow participants to access information that is otherwise not domestically available and share it with off–line communities. As a majority of Central Asia’s chat and forum participants are young people (under 30), their influence as “opinion leaders” in their local communities and their patterns of usage have important implications for future expectations of media and critical ramifications for online and off–line access to information.

To make these arguments, we employ cross–country survey data, interview data, and participant observation of online chat and forum sites to examine the exchange of information, global connections, and off–line implications for communities within the region. In addition to analyzing general trends of Internet use and information seeking behavior in Central Asia, we have gathered and analyzed data on the technical format, demographics, and social dynamics of chat rooms and forums in order to understand why and how people are using chats and forums in the region. This study has important implications for understanding the sources of information available and the attitudes of the region’s youth population, the dominant demographic of Internet users. Moreover, this study extends the analysis of how these technologies take root in a technology–emergent society like Central Asia, with important implications for understanding different attitudes toward and practices of adopting technology in other developing regions around the world.




Online and off–line communities are critical units of analysis for our study. In Central Asia, off–line communities typically consist of extended family, neighbors, and close personal relationships, such as classmates, and are less commonly associated with the impersonal ties that are used as markers of Western civil society or social capital (Putnam, 1993; Tocqueville, 1838). Despite the many modernization processes of the Soviet system, the reliance on close and personal social ties actually increased during the Soviet era as society became atomized and distrustful due to the terroristic policies of the Communist party and Soviet secret police (Jowitt, 1992). Neighborhood committees, traditional community institutions of self–help and dispute arbitration in several of the post–Soviet Central Asian countries, were co–opted by the state (Kamp, 2004). This co–optation discredited and caused the disintegration of traditional neighborhood networks. As a result, very personalized patterns of social networking and community building became vital for access to information and opportunities.

Throughout the Central Asian region, national governments control the information available in local news media and only report positive or non–controversial stories [2]. As a result, traditional social networks distribute information about local and international news, goods and services, and provide important alternatives to tightly regulated state information services. Moreover, decaying Soviet telecommunications infrastructure limits the access households have to information. For example, even in urban areas, some households still do not have phone lines, and entire villages may not have landline telephone access in rural areas. Although it varies by country, on average, only one–quarter of the population has access to landlines in Central Asia (International Telecommunication Union, 2005). In urban areas, the Internet is present, but it is expensive and inconvenient. A great deal of use occurs in public access sites such as Internet cafes. In a survey we conducted among 4,000 respondents throughout Central Asia in 2008, 40 percent of respondents who use the Internet indicated that they usually do so from Internet cafes. Due to limited access to ICTs, local residents rely heavily on family and friends for “real” news, even about critically important local events such as Kyrgyzstan’s Tulip Revolution in March 2005 and the Andijon uprising in Uzbekistan in May 2005. Chat and forum sites offer important new alternative sources of information and discourse on local, national, and international events and issues.

New electronic media are especially important in a region characterized by a large ratio of young people to older generations. As is common around the globe, use of Internet and related technologies is strongly correlated with age in Central Asia. Younger people are more common adopters and regular users of these new technologies. In Central Asia, close to 50 percent of the region’s population is less than 30 years of age and almost 33 percent is 15 to 29 years of age [3]. Moreover, our survey data indicate that, in Central Asia, this young demographic is the most dominant group in ICT use, especially Internet and chat and forum usage. Youth exposure to these new technologies and media is likely to have important implications for future demand of communication and information sources. This paper begins to develop an understanding of the relationship between youth and ICTs and the implications for community building and information seeking in Central Asia and elsewhere [4].



Theoretical context

Web–based chats and forums are an increasingly important subject of research. Using common terminology, we define chat sites as resources that allow people to communicate online in real time. Most basically, they are virtual rooms where several people gather for communication on a variety of topics. This communication takes several possible forms: a user can observe ongoing discussions, a user is able to send messages addressed to everyone in the main room, a user can communicate one–on–one by addressing his/her messages directly to a specific person in the main room, or a user can start a “private” chat, which will take place in a separate room within the chat site. Usually, when entering a chat room, one registers with a personal login and password, and users recognize each other’s login names. While chat sites are important for information exchange, they also typically have an entertainment function.

Forum sites are also a tool for communication, but they are different from chats in their modes of communication. Forums are structured around specific, well–defined topics and visitors ask questions and share their views according to the topics. Usually, forums employ a form of communication where users communicate though simultaneous and/or asynchronous postings or with gaps in time between posts. Thus, unlike chats, forums are more often used for substantive debates because more time is allowed for participants to think about and justify their postings. They also have stricter rules of behavior for users, and administrators moderate the parameters of discussions.

Recent research on chats and forums focuses on important practical considerations such as applications and design implications in work place settings (O’Neil and Martin, 2003) or how chat site discussions push boundaries of cultural norms in specific societies (Al–Saggaf and Wecker, 2004). A body of literature also focuses on the development and maintenance of online communities through the creation of new rules and mechanisms to moderate social behavior in virtual spaces (Matsuda, et al., 2002), the creation of new identities, both individual and collective, through chat room language practices (Lam, 2004), and how those identities are supported through powerful community–like environments [5]. Finally, some studies have proposed that “cyberspace may act as a ‘salve’ … where there are barriers to the development of communities” (Reymers, 2002). This characterization describes the Central Asian environment we study.

Offering important insights into the generation of culture and norms within virtual societies, the above studies pay little attention to the links between participants in virtual and real–world networks, the importance of online communities as alternative sources of information, or the ability of virtual communities to circumvent repressive local media environments. Matsuda, et al. (2002), for example, use non–participant observations of online communities to argue that virtual places have many existing off–line social attributes and mechanisms, such as manners, etiquette, and mechanisms to discourage and/or sanction anti–social behaviors in virtual spaces. These virtual societies, they argue, also introduce new rules for online behavior that are rarely seen in real life. They are particularly interested in how these virtual societies form a culture that dictates user activities, but they do not explore the inverse relationship of how online dialogues influence the off–line networks that we examine.

Other studies offer important potential implications for understanding identity and community formation through chat and forum usage. Lam (2004) undertook a case study analysis of language practices and social relationships among immigrant Cantonese speakers living in the United States and elsewhere. Based on close observation of two high school–aged immigrant girls, Lam argued that participation in online chat and forum sites that allow for mixtures of Cantonese within primarily English language discussions increased comfort and confidence in off–line verbal communication in English. These mixed–language chat discussions also allowed participants to assume new self–identities based on language practices, which led to identification with new collective groupings. Mixed language use is an important aspect of sharing information in Central Asian chat and forum sites, where users commonly intersperse Russian, the national Turkic language of each country, and English [6]. Lam’s study, however, did not differentiate among the uses of chat and forum sites for information seeking, communication, or pure entertainment nor did it explore the online and off–line implications in repressive information environments.

Shoham (2004), by contrast, does explore the communication, information–seeking, and community–building aspects of chat room usage. In his words, “People enter magical chat–room doors to bond socially, to look for solutions to personal problems, or to satisfy a need for affiliation with a community.” [7] Through his ethnographic study of a chat room for Israelis in their 40s and 50s, Shoham argues that online communities consist of “real social members” and thus reflect real world social interrelations. He concludes that virtual spaces and communication tools such as forums and chats create virtual communities that sometimes turn into enduring off–line networks. Exploring the links between online and off–line communities is therefore a potentially productive way for generating “deeper and thicker understanding of modern–day community members.” [8] Shoham’s finding is important, but its scope is limited to the relatively liberal media environment in Israel, the older population observed, and the observations of the local social relations that develop. We extend this argument to our analysis of the correlations between online identity formation in a youth population and linkages between local and global social relations and information sources in repressive media environments. Widening of the scope of study has important implications for situations where the global sphere may offer opportunities for opposition or resistance to the hegemonic power exerted by the local government on its subjects.

Examining how online communities establish linkages between Central Asia nationals and the global diaspora also situates our study. We draw on Valverde’s (2002) analysis to explore identity formation and language use, local–global community building, and circumvention of government controls in chat and forum sites. Valverde explores affiliations between Vietnamese–Americans and Vietnamese nationals in virtual communities and argues that virtual communities are a safe environment in which Vietnamese–Americans retain or develop ties to the “homeland” through friendly postings, networking projects, and activities that have direct impact on social and political change in Vietnam. Valverde argues that, through the virtual space on the Internet, the Vietnamese–American community and their Vietnamese counterparts achieve relative freedom in expressing an array of political and personal points of view with a lesser degree of self–censorship than they would be pressured to adopt in their off–line societies. Through virtual communities, citizens in Vietnam and the overseas population have engaged in meaningful dialogue and developed safe spaces to discuss controversial topics. Finally, Valverde’s study takes seriously the attempts of the Vietnamese government to introduce ICTs in a manner that preserves its political dominance while at the same time the diaspora has successfully participated in Vietnamese nation building discussions. We extend this study to explore how chat and forum users might share information that is otherwise unavailable with their off–line social networks.

Although the body of scholarship discussed above quantifies and qualifies the development of online communities, there remains a lack of literature on the off–line effects of online communities. The two–step flow of communication (Lazerfeld, et al., 1948; Katz and Lazerfeld, 1955) bridges this divide. Researching media and voting practices, Lazerfeld, et al. discovered that many people acquire information and opinions from friends, family, and acquaintances rather than through more “authoritative” sources like traditional mass media. These informants or “opinion leaders” are regarded within their communities as well–informed about certain topics and they pass on information and opinions that may result in learning, decisions, choices, and opinion changes among their social networks. Our survey data suggests that, although overall Internet use is low, Central Asians recognize the value of information gained through Internet access. Thus members of society with access to online information are likely to be “opinion leaders” in their local communities. Our study addresses how participants in these online communities might be “opinion leaders” who share new sources from chat and forum sites with their off–line communities and circumvent the repressive local media environment.

Moreover, most existing studies on the use of chat and forum sites investigate users and applications in relatively developed and democratic societies [9]. As a result, these existing studies miss the context in which Central Asian chat and forum participants develop their online practices and communications and how online interactions might allow participants to transcend local constraints on information. Because the post–Soviet Central Asian region is characterized as low in associational behavior and social capital and restricted in information resources, understanding the nature of online community as offering new sources of information is an especially important contribution of our analysis. We, therefore, argue that online discussions are important alternatives to local media and other information sources, especially to the extent that they allow participants to transcend local constraints and link with participants in other geographic areas. These online interactions have important implications for information sharing in real–world social networks.



Methodology for examining chats and forums in Central Asia

This study employs survey data, interviews, and participant observations of chat and forum sites focusing on Central Asia. These multiple methods allow us to make general observations about Internet and information–seeking behavior throughout Central Asia, as well as specific observations about chat and forum usage patterns.

The survey is part of a multi–year, multi–phase project on patterns of ICT adoption and adaptation in Central Asia. Given the low rate of current Internet penetration in Central Asia, the survey also focuses on pre–existing patterns of information use, information seeking behavior, and levels of trust in various producers and sources of information. The survey was administered in Kazakhstan, Kyrgyzstan, Tajikistan, and Uzbekistan in 2006, 2007, and 2008 by a survey firm in Kazakhstan. Based on census information on age, gender, ethnicity, and geographic location released by the government of each country, a probability sample of 1,000 respondents aged 15 and older was surveyed in each country. The survey was administered in urban and rural areas from several regions of each country. The survey data is reported below on use of and attitudes toward the Internet and other information and communication technologies among Central Asian respondents. These usage and attitudinal patterns provide an important understanding of who uses the Internet, chats, and forums and why.

In addition, we employed participant observation of chat and forum sites focusing on Central Asia. We identified the chat and forum sites to observe through Internet searches, recommendations from local citizens, and consultation with local research assistants. We then joined three to five of the most popular chat and forum sites for each of the countries studied. We participated minimally in online discussions during three time periods: January to April 2006, March 2007, and in October to December 2007. For each observation, we logged onto a site for 30–60 minutes at various times of the day over the course of the observation period. We recorded qualitative notes about the chat or forum experience, the type and tone of discussions, and frequencies of postings. We made screen captures of the chat or forum site for each observation. We also collected data on the attributes of the sites including: stated purposes of the sites; target audiences; categories and numbers of actual users; patterns of use; substance of discussions concerning development of community, transboundary ties between local Central Asian chat and forum participants and their counterparts in other geographic locations, and use of chats and forums as alternative information sources or as a venue for discussions that would be repressed off–line; languages and scripts used; use and types of avatars; and, number of links to other sites. These factors influence the use and importance of chat and forum sites for local Central Asians and have critical implications for sources and reliability of information available to off–line communities.

Forum sites usually contain basic statistics on the number of users and have profiles of users, topics, and other information. As chat sites do not usually broadcast such data, we collected it manually by counting the number of users and making our own estimates about the types of users, topics, and other information. We also made notes of special features, if any, that the sites contain. Information on specific procedures and definitions is available in Appendix I. This data gives overview information on the technical set–up, demographics, and social dynamics of the chat rooms and forums.

Finally, we conducted a number of in–depth, semi–structured interviews with Internet users in Central Asia. These conversations provide greater context and qualitative understanding of attitudes about chats and forums and the information available through them.

The following sections present evidence on general Internet usage and patterns of chat and forum participation throughout Central Asia to support our argument that online discussions are important sources of alternative information in repressive information environments for both online and off–line communities. These findings are important for the Central Asian region, and they have critical implications for sources and reliability of information available in other technologically emerging regions of the world.



Patterns of general Internet usage in Central Asia

With an increase in the availability and use of these various Internet resources, it is important to understand who the users are, what motivates them to use these technologies, and what they do online. This section presents data about Internet use and information seeking behavior from our annual survey of four Central Asian countries between 2006 and 2008 [10].

While the absolute numbers from our survey indicate low levels of Internet penetration in Central Asia, local citizens are online in all of the regional countries and their use of the Internet has grown in recent years (see Figure 1). These are the human bridges between the online and off–line communities and the opinion leaders who can deliver information in the repressive countries.


Figure 1: Internet use in 2006 and 2008
Figure 1: Internet use in 2006 and 2008.
Note: N=1,000 per country per year.


Such a broad stroke description of Internet users is illustrative, but finer grained demographic information helps to paint a more detailed picture of Internet use in Central Asia. Of our survey respondents, men are more likely to be Internet users, but, as Figure 2 shows, in Kazakhstan and Kyrgyzstan the gender breakdowns are very close. While the mean age of survey respondents who use the Internet is under 30, Internet and chat users tend to be young adults. Indeed, 41 percent of our survey respondents who use the Internet and 72 percent of respondents who use chat and forum sites were age 25 or younger.


Figure 2: Usage of Internet by gender
Figure 2: Usage of Internet by gender.
Note: N=474.


This youth demographic has important ramifications for online and off–line community in the future. This trend in age is both consistent with global Internet use and reflects the youth bubble in the Central Asia region. Not surprisingly, Internet and chat users in Central Asia are also primarily inhabitants of urban areas: Of respondents who said they use the Internet and chats, more than 70 percent reside in urban centers. Central Asians often have extended family living in towns and villages outside of the urban centers, so Internet users are able to share information beyond the main cities.

With this understanding of who Central Asia’s Internet users are, we now present information about why and how these individuals use the Internet. Despite limited connectivity, people in Central Asia actively learn about and discuss regional developments via the Internet. While traditional personal networks and nationally produced television are the most important and trusted sources of information, the Internet is viewed as a reliable source of information on regional and world events and educational and employment opportunities (see Figure 3).


Figure 3: Trustworthiness of media and social networks
Figure 3: Trustworthiness of media and social networks. Note: N=4,000.


More importantly, the Internet serves as an alternative source of information when traditional local sources are blocked, such as during the March 2005 Tulip Revolution in Kyrgyzstan and the May 2005 events in Andijon, Uzbekistan. During these events, government media outlets blocked reporting on the protests and government responses and information about the events were sparse or not available in the other Central Asian countries. Internet sites, however, were bustling with information, first–hand accounts, images, and video clips of the happenings, and local people with Internet access sought and shared this information with off–line communities [11]. As another example,, a regional Internet news provider, makes available news and forums about the Ferghana Valley, a region shared among the three Central Asian countries of Kyrgyzstan, Tajikistan, and Uzbekistan. This source of information is especially critical, as it can be difficult to obtain information about neighboring countries from traditional, local media sources, such as newspapers and television. In an innovative move, this news source started an initiative to compile a print digest of stories previously available only on the Internet for dissemination in local markets.

Indeed, 70 percent of Internet users surveyed responded that “access to otherwise unavailable information” was a primary reason for using the Internet to find news. Moreover, Internet users in Central Asia now clearly see the Internet as a way around the repressive authoritarian governments. The following interview response reflects a common view that, even should the government try to prevent access to foreign links within the country, “we will have access to international information. There is the Internet.” [12] As events in Burma in 2007, Nepal in 2008, and Iran and Urumqi in 2009 demonstrate, governments can and will take steps to shut down Internet traffic when they do not want citizens to provide or consume news on an international scale. The fact remains, however, that populations are becoming accustomed to the access to information provided on the Internet and they will adapt to attempts to close off those pathways.

In addition, 43 percent of the Internet users who responded to our survey indicated that they felt they had a greater understanding of politics as a result of their Internet use. Although this response was not overwhelmingly positive, it is interesting to note that in an environment of repressive authoritarian governments and restricted information, people see the Internet as a viable means to become better informed about the political system.

With a greater understanding of why Central Asians use the Internet, we now present what they actually do online, including chat and forum usage. The Internet allows users in Central Asia to communicate through e–mail, chats, forums, and Web cams with others in and outside the region. Much like in the United States, the Internet is becoming a replacement for traditional postal service and is more reliable than regular mail. In addition to communication, seeking information about employment opportunities and current events and research for work or school are commonly cited as Internet activities among Central Asians. The Internet is also an important form of entertainment, and downloading video and music are popular online activities.


Figure 4: Online activities in 2006 and 2008
Figure 4: Online activities in 2006 (N=376) and 2008 (N=474).


As Figure 4 shows, a little more than one–third of Internet users in our survey indicate taking part in chat room discussions. Thus chats and forums are important online activities. The next section details our findings from participant observation of Central Asia chat and forum sites and discusses how chat and forum sites are used in the region.



Patterns of Central Asian chat and forum usage

This section reports qualitative data on participant observations of a number of the most popular chat and forum sites focusing on Central Asia in each of the five Central Asian countries [13]. We classified data from chat sites according to general information about the site (domain, administrative and discussion languages, presence of moderator, rules of participation), geographic location of participants, discussions providing information that would be repressed off–line, dissemination of information and news that would not be domestically available, and demographic data and other characteristics of participants. These variables are important determinants of online community formation and have critical implications for sharing new information in real–world communities.

Although each country has unique characteristics, we discovered a number of common features among participants throughout the five countries. In particular, we found a wide geographic distribution of participants from inside and outside the region, supporting our argument that chat and forum participation enhances the sources of information available both to the participants and to their off–line communities. We also found that the content of the discussion, rather than the domain in which the site is located, is a major factor determining the popularity of the site. We find this especially interesting because we had hypothesized that individuals might find the use of chats and forums in a domain outside of the reach of a repressive government a safer environment in which to engage in public debate. Nevertheless, censorship and self–censorship practices within certain sites are easily observable. In addition, across all countries, Russian or English are the primary languages for site administration and substantive discussions. Rather than using individual national languages, the preference for Russian and English demonstrates the ties to a broader online community of participants from various geographic locales. Finally, we found that, even when using non–national language in the very global medium of the Internet, Central Asian chat and forum participants create and use avatars and emoticons, nicknames, images, and symbols that reflect their national or ethnic characteristics. In this way, online communities are imbued with local culture. These findings are discussed in more detail below.

Geographic location of participants

Several of the chat and forum sites we observed collect data on the geographic location of participants. In addition to participants within Central Asia, these sites indicate a large number of users residing outside of the regional countries. In a number of cases, as many as half of active chat or forum participants reside outside of Central Asia. As evidence, Figure 5 (below) is a screen shot of two sites’ graphic representations of their participants’ locations and shows the geographic distribution of participants on a forum site in Kyrgyzstan. Regardless of location, users communicate in a common language — usually Russian — and are interested in similar topics. Central Asian–focused forums thus attract groups of users from varied locations.


Figure 5: Representation of geographic distribution of chat and forum users
Figure 5: Representation of geographic distribution of chat and forum users
Figure 5: Representation of geographic distribution of chat and forum users.
Note: In addition, a Kyrgyzstan forum ( uses ISP tracking to list participants’ countries as France, Kyrgyzstan, United States, Kazakhstan, Germany, Russia, Czech Republic, China, Turkey, and Belgium.


All of the chats and forum sites we observed were designed as generic conversation spaces that have evolved into conversations between people in a Central Asian country and citizens temporarily or permanently outside of that country. Despite the high cost of Internet access in Central Asia, chat discussions through the Internet are generally cheaper and more easily accessible than international telephone calls. In addition, throughout the region, and particularly in Uzbekistan, telephones are used more cautiously than the Internet for fear of government eavesdropping. At the same time, we find that the location of participants determines the type of content in which they are interested. Based on the topics discussed in the various chat sites and postings to forums, participants residing outside of Central Asia are generally searching for information about news and current events in a particular regional country. Conversely, people residing in a Central Asian country are often interested in social and political developments outside of their home country, information that is often difficult to find in Central Asia. Thus, local participants and their international counterparts exchange information, expanding the sources of information available in their domestic settings.

Discussion content

Contrary to our expectations, we discovered that hosting location or domain is not a factor determining use as measured by the number of visitors to a chat or forum site. Instead, the actual content of the site seems to play the significant role. In other words, chat and forum sites located both inside and outside a given country’s domain attract users at different rates based on the content at the sites, not the domain location. For example, the most popular Kyrgyzstan forum site,, is in the Kyrgyzstan domain, but the next most popular site is in the general domain ( Likewise, the most active Uzbekistan chat sites are located in the .uz or .ru domains (,, and, while the most popular Uzbekistan forum,, was created by immigrants from Uzbekistan and the majority of participants reside in Russia and United States. The most active forum in Kazakhstan ( is located outside the Kazakhstan domain. Despite these observations, the assumption that people feel more secure engaging in public debate in a chat or forum site outside of the local country domain might be valid on the basis of the observed self–censorship practices discussed below.

Forum sites are generally oriented toward specific audiences, and the conversations tend to cover specific topics: analytical discussions such as politics within a particular Central Asian country, social problems, news, humor and entertainment, search for people, professional topics, and nostalgia about the country of origin. Several of these topics are generally considered areas of discussion that would not be appropriate for public discussion within the traditional mass media in Central Asia. Most apparent, in this context, were analytic discussions about religion, politics and society, and news items. Almost all the forums that we observed included direct discussions about politics. In general, chat sites, however, did not focus on political or other controversial topics.

In one chat observation, for example, a user began an in–depth discussion of the tense relationship between the ruling regime and opposition parties in Uzbekistan. The user gave an overview of the persecution that opposition party activists and their children and relatives faced. The user then stated that (author’s translation from combined use of Russian and Uzbek languages; identifying user names and the site name have been removed):

<user1> “Opposition parties and democratic forces and anybody who was thinking differently than the ruling regime are kicked out of political life in the country and are represented to society as uneducated, terrorists, extremists and enemies of the people. Their attitudes about the processes happening in the country, they can only publish on the Internet, which is regularly blocked by the government.” (from site observation, 13 March 2006)

User1 dominated the chat with this line of discussion for more than five minutes. Then another user suggested a URL ( for a forum that might be more receptive to the topics. A few other chat participants stated agreement that they were tired of hearing from User1, but then another participant chimed in that “if you keep silent, you yourself are guilty [of political repression]”. User1 continued the political discussion for still a few more minutes and then User1 was not seen again for the rest of the observation. A GuestUser appeared in the site continuing the same line of reasoning. User1 may have simply stopped participating, or User1 may have been blocked by the site monitor and forced to create a new user id.

This example demonstrates how chat and forum sites allow participants to exchange information that is generally repressed off–line. Indeed, several of the observed forum sites include a note on the main page or in the administrative information that the site is intended as an area for “freedom of expression.” [14] At the same time, however, some sites, such as Diesel Forum in Kyrgyzstan (, explicitly state rules against the use of the forum for political or religious propaganda. Thus, the site rules limit the range of conversations, but, in most cases, those limitations do not explicitly prohibit the discussion of political topics. Yet, within particular chat and forum sites, we observed surprising levels of “self–censorship” in which other forum members warned a participant to be cautious about voicing opinions about politically sensitive topics.

For example, the following exchange demonstrates this type of group censorship (author’s translation from combined use of Russian and Uzbek languages; identifying user and site names and unrelated discussion from other participants have been removed):

[19:01:39] <user1> [expletive] I.Karimov [Uzbekistan’s president]
[19:01:51] <user2> User1, watch your words!
[19:02:00] <user1> but it’s true.
[19:02:15] <user3> User1, what are you doing?!
[19:02:39] <user4> User1, you better not write these kinds of things in the common forum.
[19:02:54] <user5> User2, are you afraid?
[19:03:02] <user4> the walls also have ears, you know!! Who knows … .

Self–censorship among participants in the forums was clearly identified, but it was harder to monitor in the chat sites because chats are generally non–archived, online conversations. In country forums where political themes are very sensitive, as in in Kazakhstan and in Uzbekistan, chat and forum participants use self–regulation to avoid discussions of political themes, such as elections, presidential candidates, government reforms, criticism of political system, government officials, etc. In these discussions, participants living inside the particular countries do not feel safe to discuss politics, even when the sites are hosted in domains outside of the local country, as users’ IP addresses can be tracked to a specific ISP within the country and a particular user. The fear of being tracked creates a level of uncertainty about the safety of engaging in open discussions on politically sensitive topics. Indeed, by doing our own IP tracing, we observed that people living outside the more repressive countries tend to engage in more open discussions on political topics. We also found, however, that discussion of social problems such as poverty and unemployment is less self–censored, even among participants residing in the more repressive regional countries.

In addition to self– or group–censorship, more explicit censorship from site managers also takes place. Generally site robots kick participants out of the chat discussions for a generic offense of “flooding” the chat. We could not categorically determine what triggered this type of censorship, but it seemed to be related to specific keywords such as the repeated use of the president’s name or for cutting and pasting articles about local political issues from online news sources.

Target and actual participants

Many of the chat and forum sites we observed state that they are intended for specific target audiences and track the demographic data of site participants. On both chat and forum sites, target users are generally young people living in Central Asian countries, young emigrants from the region, or people living temporarily outside of Central Asia. Based on self–reporting, participants in these online communities are typically in their late teens through late 20s. Generally a common interest in a specific topic or theme draws users to a site. These themes vary widely and include topics as diverse as ethnic minorities’ rights within a particular country to tips on studying or living abroad. As young people make connections to other participants around the globe and gain access to new sources of information, their world views and their relationships to their off–line communities might be expected to shift.

As we classified chat and forum sites, we found the greatest cross–country variation in terms of target audiences and volume of participants. We measured chat and forum site activity by the number of users logged onto the Web site at the times of observation and according to the total number of registered users. Based on registration information provided by users, Kazakhstan’s chat and forum sites have the most obviously diverse populations of users, and these sites are more specifically designed for certain target audiences. For example, while sites exist for general public audiences ( and, many other sites are designed for specialized groups such as bankers (, attorneys and legal professionals (, and library users (, among others. One possible explanation for the greater degree of specialization in Kazakhstan is the higher level of socioeconomic development in Kazakhstan, which gives the population more access and exposure to ICTs.

In measuring the volume of participants, we recorded numbers of registered users on forum sites and we observed the volume of unique participants during a given observation. Of course, there are non–registered users who cannot post comments or messages but who can observe discussions and read posted information. Likewise, one unique poster on a site might represent a group of individuals gathered around a single computer. By contrast, it is uncommon for chat sites to give the number of registered users. Instead, we counted the number of unique users present at given times of observation. We also returned to these sites one year later to assess the change in user volume and found dramatic increases in the numbers of users of both chat and forum sites throughout the region (Table 1).


Table 1: Chat or forum site with largest number of participants, 2006/2007.
Note: These figures are based on the number of users registered on a site at the time of observation in 2006 and 2007.
Chat or forum siteRegistered users (Kazakhstan)2,147/38,652
http:/ (Kyrgyzstan)11,922/10,829
http:/ (Tajikistan)2,398/3,065
http:/ (Turkmenistan)3,180/8,722
http:/ (Uzbekistan)16,000/31,325



Russian is the primary language for site administration and substantive discussion on the chat and forum sites. In a couple of cases, English or both Russian and English were used. The wide use of Russian language enables participants to expand their online communities beyond speakers of their national languages, even while retaining elements of their particular countries, ethnic groups, and/or languages. Indeed, despite the common use of Russian across all countries, the forum and chat sites differ in their use of English and local languages. In Kazakhstan–related sites, for example, English is more dominant than Kazakh. In Uzbekistan sites, however, Uzbek is more commonly the second language. In Tajikistan and Kyrgyzstan, site content is posted in the local language but mostly using Cyrillic, rather than Latin, script. Overall Cyrillic script was used most frequently among all the observed chat and forum sites [15]. Indeed, we found all but one of the sites have rules governing the use of Cyrillic script for both Russian and local language postings. For example, chat and forum sites in Kyrgyzstan and Kazakhstan have rules stating that, for the convenience of the author and readers, postings in Russian language should only use Cyrillic script. These sites even offer transliteration programs in case the Cyrillic script is not available on an individual’s keyboard. If the rule is violated and postings in Russian language are written using the Latin alphabet, the site moderator deletes the postings. Because of the frequent use of Russian and English, communities that might otherwise be linguistically, as well as geographically distant, find common language for communicating and sharing information in chats and forums. Language choice and usage serve to bridge different communities in the virtual space.

Local characteristics in avatars and emoticons

Despite the preference for non–national language, a variety of nationally or regionally inspired avatars and emoticons are noticeably used. This localization of the global medium varied in practice from simply using nicknames based on real or folklore heroes and celebrities with distinct national or regional characteristics, to complex graphic representations and uploaded photos with regional or national icons and images. Instead of drawing on global pop culture, these avatars and emoticons reflect the cultural features of the local countries. As shown below (see Table 2), even smiley faces take on local features. Thus, the avatars and emoticons used by Central Asian chat and forum participants reflect national and ethnic characteristics and, in this way, imbue online communities with local culture, even when the participants are thousands of miles away.


Table 2: National emoticons.
UzbekistanUzbekistan emoticon Uzbekistan emoticon
KazakhstanKazakhstan emoticon
TurkmenistanTurkmenistan emoticon
KyrgyzstanKyrgyzstan emoticon Kyrgyzstan emoticon Kyrgyzstan emoticon Kyrgyzstan emoticon Kyrgyzstan emoticon





Central Asia’s chats and forums have become important alternative sources of information in a region where governments repress access to local, regional, and international news and current events. Moreover, online communities consist of individuals who are part of real–world social networks and the information gained through chats and forums is shared in off–line communities. Because youth is the largest demographic participating in chat and forum sites, these online exchanges and communities are likely to have important long term implications for information seeking and expectations of new and traditional media.

The creation of virtual communities through chats and forums provides a new arena for public discourse in Central Asia. Because Central Asian chat and forum sites attract both regional populations and participants in other parts of the world, the information and viewpoints exchanged in these online discussions is varied and vast. Russian language, being the most commonly used language in Central Asian forums and chats, bridges participants across the region and also participants in more disparate locations. Because chat and forum sites are available in many countries of the world, the phenomena discussed in this paper have implications beyond Central Asia to the entire global sphere.

Finally, the exchange of information on chat and forum sites is developing in parallel to other types of new media, including blogs and instant messaging. Applying the findings from this study of chats and forums to understanding the adoption and adaptation of these new media will have important implications for online and off–line community building, access to and saturation of information, and expectations of future media and ICTs. End of article


About the authors

Erica Johnson is a PhD candidate in the University of Washington’s Political Science Department, studying political transitions and state–society relations in post–Soviet Central Asia.
Direct comments to ejj3 [at] u [dot] washington [dot] edu

Beth Kolko is associate professor of Human Centered Design & Engineering at the University of Washington. Dr. Kolko serves as principal investigator of the Central Asia + Information Communication Technologies (CAICT) project.

Odina Salikhbaeva was a research assistant on the CAICT project while pursuing her bachelor’s and master’s degrees at the University of Washington. She is now living and working in Tashkent, Uzbekistan.



The authors are grateful for the research assistance of Shoshana Billik and Anastasia Shpakova and thoughtful comments from Cynthia Putnam and Eric Gleave.



1. Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, and Uzbekistan are the newly independent states of Central Asia.

2. In fact, Reporters Sans Frontières has rated Central Asian governments among the most repressive in the world in press and Internet freedoms, even labeling several of the local governments “enemies of the Internet” (Reporters Sans Frontières, “Annual Report on Press Freedom, 2002–2007”).

3. International Crisis Group, 2003, p. 1.

4. While the relationship between the youth bulge and new ICTs has yet to be researched, a great deal of analysis focuses on the youth bubble and civil unrest and violence. See, for example, Jeremy Ginges, 2005. “Youth bulges, civic knowledge, and political upheaval,” Psychological Science, volume 16, number 8, pp. 659–660; and, Henrik Urdal, 2006. “A clash of generations? Youth bulges and political violence,” International Studies Quarterly, volume 50, number 3, pp. 607–630.

5. See, for example, Tepper, 1997; Kitalong and Kitalong, 2000; Valverde, 2002; Lam, 2004; Shoham, 2004.

6. The choice of a single or hybrid language is an important communication devise in blurring regional and global geographical boundaries online. In a related project study, Wei and Kolko (2005) explored how people in Uzbekistan adapted their linguistic behavior and language choices to the Internet and expressed themselves in culturally meaningful ways without being subsumed by a global agenda.

7. Shoham, 2004, p. 855.

8. Ibid., p. 856.

9. Emblematic of this approach are studies claiming that the Internet promotes open, democratic debate and allows for multiple perspectives (Nye, et al., 1997; Putnam, 2000; Rheingold, 1993). Other scholars, remind us, however, that the Internet offers opportunities for enhanced government control and surveillance (Yang, 2003; Kalathil and Boas, 2003).

10. Because of repressive government conditions, our survey does not include Turkmenistan. The International Telecommunication Union (ITU) estimates that about 0.7 percent of Turkmenistan’s population uses the Internet. With the death of Turkmenboshi, Turkmenistan’s president from December 1991 to December 2006, some of the government’s most repressive Internet policies have been relaxed and the number of Internet users in Turkmenistan could grow rapidly in the coming years.

11. Author’s interview, Bishkek, 23 July 2006.

12. Author’s interview, Almaty, 29 January 2008.

13. Although our survey was only conducted in four Central Asian countries, we were able to explore chat and forum sites for all five countries.

14. See, for example, the Forum of Immigrants of Uzbekistan,

15. As part of integration in the Soviet Union, each of the Central Asian republics was forced to write the national language in Cyrillic script. Since independence, the Central Asian national languages are increasingly written in Latin script.



Yeslam Al–Saggaf and John Weckert, 2004. “The effects of participation in online communities on individuals in Saudi Arabia,” ACM SIGCAS Computers and Society, volume 34, number 1 (June), p. 1.

Author’s interview with an Internet user, Almaty (29 January 2008).

Author’s interview with an Internet iser, Bishkek (23 July 2006).

Albert Bandura, 1986. Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, N.J.: Prentice–Hall.

Mridul Chowdhury, 2008. “The role of the Internet in Burma’s Saffron Revolution,” Berkman Center Research Publication, number 2008–08, at

Jeremy Ginges, 2005. “Youth bulges, civic knowledge, and political upheaval,” Psychological Science, volume 16, number 8, pp. 659–660.

International Crisis Group, 2003. “Youth in Central Asia: Losing the new generation,” Asia Report, number 66. Osh/Brussels: International Crisis Group, and at

International Telecommunication Union. 2005. “Europe and CIS’s telecommunication/ICT markets and trends,” at

Kenneth Jowitt, 1992. New world disorder: The Leninist extinction. Berkeley: University of California Press.

Shanthi Kalathil and Taylor C. Boas, 2003, Open networks, closed regimes: The impact of the Internet on authoritarian rule. Washington, D.C.: Carnegie Endowment for International Peace.

Marianne Kamp, 2004. “Mahalla Committees and social welfare in Uzbekistan,” In: P. Jones Luong (editor). The transformation of Central Asia: States and societies from Soviet rule to independence. Ithaca, N.Y.: Cornell University Press, pp. 29–58.

Elihu Katz and Paul Lazarsfeld, 1955, Personal influence: The part played by people in the flow of mass communications. Glencoe, Ill.: Free Press.

John Kelly and Bruce Etling, 2008. “Mapping Iran’s online public: Politics and culture in the Persian blogosphere,” Berkman Center Research Publication, number 2008–01, at

Karla Saari Kitalong and Tino Kitalong, 2000. “Complicating the tourist gaze: Literacy and the Internet as catalysts for articulating a post—colonial Palauan identity,” In: Cynthia L. Selfe and Gail A. Hawisher (editors). Global literacies and the World Wide Web. New York: Routledge, pp. 95–113.

Wan Shun Eva Lam, 2004. “Second language socialization in a bilingual chat room: Global and local considerations,” Language Learning & Technology, volume 8, number 3 (September), pp. 44–65, and at

Paul F. Lazarsfeld, Bernard Berelson, and Hazel Gaudet, 1948. The people’s choice: How the voter makes up his mind in a Presidential campaign. Second edition. New York: Columbia University Press.

Rebecca MacKinnon, 2008. “Flatter world and thicker walls? Blogs, censorship and civic discourse in China,” Public Choice, volume 134, numbers 1–2, pp. 31–46.

Kouichi Matsuda, Takahiro Miyake, and Hirofumi Kawai, 2002. “Culture formation and its issues in personal agent–oriented virtual society,” Proceedings of the 4th International Conference on Collaborative Virtual Environments (30 September–2 October; Bonn, Germany), pp. 17–24.

Joseph S. Nye, Jr., Philip D. Zelikow, and David C. King (editors), 1997. Why people don’t trust government. Cambridge, Mass.: Harvard University Press.

J. O’Neil and D. Martin, 2003. “Text chat in action,” In: K. Schmidt and M. Pendergast (editors). GROUP ’03: Proceedings of the 2003 International ACM SIGGROUP conference on Supporting Group Work, pp. 40–49.

Robert Putnam, 2000. Bowling alone: The collapse and revival of American community. New York: Simon and Schuster.

Robert Putnam, 1993. Making democracy work: Civic traditions in modern Italy. Princeton, N.J.: Princeton University Press.

Reporters Sans Frontières, 2007. “2002–2007 Annual Report on Press Freedom,” at

Kurt Reymers, 2002. “Identity and the Internet: A symbolic interactionist perspective on computer mediated social networks,” paper presented at the Association of Internet Researchers Third Annual Conference, Maastricht, the Netherlands (October), at

Howard Rheingold, 1993. The virtual community: Homesteading on the electronic frontier. Reading, Mass.: Addison–Wesley.

Aviv Shoham, 2004. “Flow experiences and image making: An online chat–room ethnography,” Psychology and Marketing, volume 21, number 10, pp. 855–882.

Michele Tepper, 1997. “Usenet communities and the cultural politics of information,” In: David Porter (editor). Internet culture. New York: Routledge, pp. 39–54.

Alexis de Tocqueville, 1838. Democracy in America. Translated by Henry Reeve, with an original preface and notes by John C. Spencer. New York: G. Dearborn & Co.

Henrik Urdal, 2006. “A clash of generations? Youth bulges and political violence,” International Studies Quarterly, volume 50, number 3, pp. 607–630.

C.K.L. Valverde, 2002. “Making transnational Viet Nam: Vietnamese American community–Viet Nam linkages through money, music and modems,” unpublished doctoral dissertation, University of California, Berkeley.

Carolyn Y. Wei and Beth E. Kolko, 2005. “Resistance to globalization: Language and Internet diffusion patterns in Uzbekistan,” New Review of Hypermedia and Multimedia, volume 11, number 2 (December), pp. 205–220.

Guobin Yang, 2003. “The co–evolution of the Internet and civil society in China,” Asia Survey, volume 43, number 3, pp. 405–422.


Appendix I: Definitions and data collection procedures for chat and forum observations

We used the following definitions and procedures as the basis of our participant observations of the chat and forum sites:

Stated purpose of the site was determined through the site name, administrative pages of the site, and other signals for the intended topics of discussion.

Target audience that a particular Web site appeals to was determined according to the administrative pages and content of the site, advertisements, and topics of discussion.

Categories of users included information on the gender, age, and location of actual users of the site. This information was gathered from personal profiles available at the sites.

Number of users was captured either from statistical data compiled by a particular site or by manually counting the users at the specific time and date of observation.

Content of discussion/topics discussed in chats and forums were identified as news, analytical discussions, entertainment, search, Q&A, language forums, and clubs. This category also allowed us to explore the discussion of community development and use of chats as alternative sources of information.

Geographic location of the participants was based on personal profiles and content of discussion. It measured the location of Central Asian chat and forum participants around the globe.

Chat and forum activity was measured by the volume of unique posts to the chat or forum site.

Administrative language corresponded to the administrative part of the site.

Primary v. other language in chat/forum was counted by word. These tallies in chat sites were based on the number of words in each language captured during the time of visit to the chat. The “primary language” corresponded to 60 percent or more of the language used, and other languages were used in 40 percent or fewer chat and forum discussions.

Use of Cyrillic v. Latin script was determined by counting the frequency of words in Cyrillic versus Latin script.

Avatars and emoticons of the chat and forum users were observed and defined as: nicknames based on local real or folklore heroes and celebrities, and local, national, or ethnic images, symbols in simple graphic representation (smiley face v. elaborate representation), complex graphic representation, and uploaded photos.


Appendix II: Screenshots of regional chat and forum sites


Figure 6: Kazakhstan:
Figure 6: Kazakhstan —



Figure 7: Kyrgyzstan forum:
Figure 7: Kyrgyzstan forum —–



Figure 8: Uzbekistan forum:
Figure 8: Uzbekistan forum —



Figure 9: Tajikistan forum:
Figure 9: Tajikistan forum —



Figure 10: Tajikistan:
Figure 10: Tajikistan —



Figure 11: Turkmenistan forum:
Figure 11: Turkmenistan forum —



Editorial history

Paper received 15 May 2009; accepted 10 July 2009.

Creative Commons License
This paper is licensed under a Creative Commons Attribution–Noncommercial–No Derivative Works 3.0 United States License.

Boundaries and information: Sidestepping restrictions through Internet conversations
by Erica Johnson, Beth Kolko, and Odina Salikhbaeva.
First Monday, Volume 14, Number 8 - 3 August 2009

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2019. ISSN 1396-0466.