Sleepless in Belgrade: A Virtual Community during War

In this paper the results of research on the role of a virtual community during wartime are presented. A virtual community within the Belgrade-based online system, SezamPro, was explored in the periods before, during and after the NATO bombing of Yugoslavia. This research found that the community had gone through significant changes during the War, including a) number of participants increased; b) users spent much more time online; and, c) reason for communicating changed dramatically. During war, online efforts focused on information gathering, social interaction and the expression of political opinions. In a period of crisis strong interpersonal relationships were established within the studied online group. Furthermore, in such state of affairs the Internet became an important source of information.

Contents

Introduction
Aims and Methods
Context
Results
Wartime Chat as a Source of Information
Wartime Chat and Political Opinions
Wartime Chat and Socialising
Conclusion

Introduction

Virtual communities are one of the frequently studied and debated topics within social informatics. They are interpreted as essentially a postmodern phenomenon (Reid, 1991), social creations of the postindustrial society (Fenerback and Thompson, 1995), the virtual substitute for the "real" place where people gather for conviviality (Rheingold, 1993) and the marketing trick of computer corporations (Critical Art Ensemble (CAE), 1995). Research on virtual communities has led to significant questions on the nature and role of computer-mediated social interaction, the characteristics of relationships developed online, and the impact of online social engagement on an individual's life "in reality". In regard to these issues, not only do the scientists' attitudes differ, but empirical studies demonstrate contrary results as well. For example, research at Carnegie Mellon University (Kraut et al., 1998) has shown that social relations are rarely established online, and that those established belong to the group of weak social ties. Contrary to these results, the research of Parks and Floyd (1995) demonstrated that personal relationships are often established via the Internet, and Rheingold (1993) described personal interrelationships developed within the WELL virtual community as strong and warm. Studies focused on the impact of online social interaction on traditional social life have also shown contradictory results. Again, efforts at Carnegie Mellon (Kraut et al., 1998) discovered that this influence was negative. Nevertheless, Katz and Aspden (1997) concluded that there was no evidence that Internet usage negatively affected traditional types of social engagement.

In order to examine these questions more thoroughly, it could be useful to analyze computer-mediated groups and online social interaction under certain conditions which affected all of the groups as a whole. An interesting research perspective can be achieved by analyzing these online phenomena in a state of war. An air campaign conducted from 24 March to 10 June 1999 by the North Atlantic Treaty Organization (NATO) against the Federal Republic of Yugoslavia (FRY) provided a possibility for such an examination.

The NATO bombing of Yugoslavia offered a great deal of evidence about high-tech achievements in military technology ("invisible" airplanes, "smart" missiles) and made us aware of a war in a digital age. Moreover, this conflict aroused questions significant from the perspective of social informatics, that is queries regarding:

The role of the Internet and computer-mediated communication (CMC) at wartime;

Motivation for using information and communication technologies (ICTs) in circumstances such as war; and,

The character of social implications, if any, of the usage of ICTs in a state of war.

These questions are actually about the meaning of "being digital" during a digital war. They compelled us to ask questions about the reactions of Internet users during such state of affairs. Furthermore, these are the questions regarding the role of computer-mediated social structures, such as virtual communities, in a state of war. In what way, if any, does the war affect computer-mediated structures? What kind of knowledge about these structures can we gain by analysing their functionality during a war?

The search for answers to some of these questions started in Yugoslavia soon after the War ended. A month after the bombing (July 17th 1999) the panel "www.agression - Towards the Ethnography of Internet communication" was organised in the Faculty of Philosophy at Belgrade University. It was aimed at starting a scientific query on the role of Internet and CMC during the War. The panel highlighted the extremely important role of the global network, both as means of information and means of communication, at a time of crisis [ 1].

The research presented here is a part of the project "Windows 99: Internet and War". This project was entitled after the metaphor created in Yugoslavia during the bombing. "Windows 99" was an ironical paraphrase of the famous Microsoft operating system and it referred to the "real" windows of Yugoslav flats at wartime. Covered with scotch tape, which was to preserve glass breaking and consecutive injuring, the windows were one of the most remarkable war symbols. "Windows 99" was the symbol itself, the mark of the war whose "digital participants" created this cyber metaphor, urging us to think of the notion of war in the postmodern age.

Aims and Methods

This paper presents the results of the research focused on one segment of CMC practice in Yugoslavia during the NATO bombing. The changes that appeared in the usage of a Belgrade-based online system, SezamPro, were analysed. As a user of SezamPro, I noticed that participation in the communication services of this system - the chat room and conference system - increased during the War. This discovery encouraged me to to find the causes for these changes after the War had ended. I wanted to find out what had motivated the SezamPro users to intensify their online activities at the time of a crisis, and in the services of communication.

I observed and participated in various online activities with SezamPro to collect relevant data for this study and also used an online questionnaire. In December 1999 the questionnaire was posted on the SezamPro Web site, and thus was available to all its users. The poll was anonymous.

In order to explore the topic properly and not to make an extensive questionnaire at the same time, I decided to focus the research on one of the two services mentioned. Since the research was aimed at analysing the changes in the usage of communication services, I chose to do the research on chat as explicitly communicational.

The questionnaire was designed in such a way to collect information on the users' chat practice before, during and after the War, so that the comparable data could be obtained. I identified the following key points for comparative analysis:

number of participants in a given chat;

frequency of chat visits;

The time spent in a given chat; and,

basic motives for participating in a chat.

The frequency of visits and the time spent in the chat were estimated according to an average value. Concerning the frequency of visits, on average there were three or more visits a week, with three or more hours a week spent in session. Basic motives for participating offered in the questionnaire were: fun, socializing, meeting someone, and getting informed (with a blank space offered for additional motives).

The questionnaire was filled out by 136 persons, of which two thirds (73%) were male and one third (27%) female. The majority of the respondents (68%) were between the age of 18 and 35. The age structure coincided with educational achievements - 50% graduated from high school (most of them were students) whereas 48% graduated from schools of higher education and faculties. Two thirds (70%) of the informants were residents of Belgrade and three quarters (80%) declared themselves to be of Serbian nationality.

Context

Computer-mediated communication analysed in this paper was placed within specific real and cyber contexts. For a better understanding of the circumstances under which the participants communicated and of the technical characteristics of the analyzed system, I now will provide a brief look at the SezamPro system and the NATO bombing of Yugoslavia.

Cyber Context

SezamPro [ 2] is the oldest online system in Yugoslavia. It was founded in 1989 as a BBS and it was very popular conference system in former Yugoslavia. In 1995, Sezam evolved to Sezam Pro, an online system. Today, SezamPro offers full Internet access, conference system, chat and public domain with various contents (further details can be found at http://www.sezampro.yu).

The conference system and the chat channel are available to members only. The conference system consists of 29 topics. SezamPro has one chat room, without a specified topic. No more than 100 users can chat in the room at the same time. SezamPro members can have only one identity, so identity switching is not possible. Users' identification is available to all other members. The exact number of SezamPro members is not available as the company's management considers this information confidential.

At wartime, there were no significant changes in the functionality of SezamPro. A generator solved shortages of electricity caused by the bombing, so the system was out of function for only three days during the course of War. There was one change in the use of SezamPro during the War; before the War, users could use up to two hours daily; during the War users had up to 2.5 hours available daily [ 3].

Social Context

The NATO bombing of Yugoslavia began on 24 March 1999. The air raids on the capital of Belgrade started on the same date. Belgrade is situated to the north of Yugoslavia and has a population of two million inhabitants.

The bombing was continuously conducted for 78 days. In that period, NATO conducted over 38,000 combat sorties, including 10,484 strike sorties [ 4].

This operation, as defined by American military terminology, was performed in accordance with the "National Military Strategy of Flexible and Selective Engagement" and it meant selective bombing of targets identified as strategic. These targets included infrastructure buildings, communication and public media (Dimitrijevic, 1999), so the attacks were mostly performed in inhabited areas. In Belgrade, the headquarters of Serbian State Television, the building of the Yugoslav Army Headquarters, the "Usce" business centre, heating plants, power stations, and many other buildings situated in the central, densely populated city areas were bombed. Belgraders heard air raid alarms 148 times and spent 774 hours under air raid danger. Similar situations occurred in other towns of Serbia. Despite NATO's technology, civilians and the civil infrastructure suffered the greatest damage. According to Yugoslav officials, 1,500 civilians were killed during the NATO bombing. The civil infrastructure was seriously damaged as well. There were 4,489 damaged residential buildings in Belgrade, with 72 of them completely destroyed [ 5].

Results

During wartime the number of SezamPro chat participants increased by 16%, the frequency of visits grew by 57%, and the number of users who spent three or more hours a week chatting increased by 84%. The basic motives for chatting considerably changed. Before the bombing, fun was the main stimulus (58%), while during the War acquiring information became the most important motive for participating in the chat (89%). When the War ended, SezamPro members indicated that they participated in the chat predominantly for socializing (46%).

What motivated SezamPro users to spend dramatic moments during the bombing online and to chat more often and for longer periods of time than they had before the War? In order to answer these questions, the functions of the analysed service during wartime had to be identified. The respondents' answers suggested that these functions could be recognised by analysing the chat as a source of information, as a place where people gathered and socialised, as well as the place where political opinions were expressed.

Wartime Chat as a Source of Information

In a state of crisis, information becomes crucially important. As Loges (1994) noticed:

"The desire to acquire information in the face of an acute threat, and the belief that the mass media are helpful sources of pertinent information, seems so self-explanatory that they need no further study" [ 6].

The Yugoslav anthropologist Slobodan Naumovic (1999) explained that:

"In a situation of aggression, the need for reliable information is related to obtaining the reliable news on time, including other pieces of information about the nature, causes and consequences of the conflict, particularly about its ongoing course. At wartime, news, commentaries, analysis, and other sorts of information are necessary to make decisions, overcome doubts and suspicions, face anxiety." [ 7].

However, wartime intensifies propaganda and censorship, so reliable information, as much as they are necessary, turn out to be nearly unobtainable.

During the war in Yugoslavia, the Internet became a particularly important source of information. Given the Internet's rich variety of sources, speed, and lack of political control and censorship, the Internet was a way to avoid war censorship and propaganda from both parties in conflict.

Most of the respondents (73%) secured information during the War via the Internet, whereas traditional electronic media were less represented (television 40%; radio 35%). A vast majority (96%) visited informational Web sites; CNN at http://www.cnn.com was visited (44%) most frequently followed by http://www.beograd.com (28%). Other Yugoslav Web sites were visited slightly more (19%) than alternative Western Web sites (17%). However, a majority of respondents (48%) reported that they consulted during the bombing both Yugoslav and Western Web sites simultaneously. Furthermore, the Internet was thought to be the most reliable medium (33%), with radio a distant second (8%) and television viewed as the least reliable (7%). It is important to note that the majority of respondents (35%) did not consider any media completely reliable. The only reliable information was empirically gathered, that is personally heard or seen.
Burnett (2000) suggested that:

"... one's 'information neighborhood' is not only made up of media sources, but also - and perhaps more importantly - by people, including family, friends, neighbors, co-workers, and a shifting network of acquaintances" [ 8].

Burnett also noticed that:

"The concept of an 'information neighborhood' as an environment within which practical information seeking and orienteering information seeking - as well as both directed and undirected browsing - can take place is particularly appropriate for virtual communities" [ 9].

The survey responses imply that the studied online group was an important "information neighborhood". For 89% of the respondents information gathering was the main motive for participating in the SezamPro chat during the War.

The role of SezamPro chat as a source of information could be noticed right at the very beginning of the War, during the night between 23-24 March 1999. That night was tense and dramatic for the inhabitants of Yugoslavia, since the former NATO Secretary General, Javier Solana, gave the order for the air raid. Expecting to hear air raid alarms and hoping that something would change at the last minute, Yugoslavs feverishly listened to news, collecting every piece of information, trying to foresee what might happen. The independent Belgrade Radio B92 was one the main sources of information. However, in the late hours, the broadcast of this radio station was suddenly interrupted. No one knew what happened. In an atmosphere of complete tension, this event was disturbing. Belgraders who were listening to the radio B92 that night started calling each other, trying to find out what had happened. Members of SezamPro started logging into the chat, hoping to get information concerning the event. Within seconds new chatters arrived, and their first question was: "Hey folks, what's happening with B92?" There were more and more visitors and the chat topic was unanimous - they were all talking about B92. After an hour the chat participants slowly started to go away, promising to forward any information on the subject. Soon after, the first pieces of information on closure of Radio B92 had arrived to the conference system [ 10]. The SezamPro "alternative information net" started to operate.

This role of the SezamPro chat lasted during the entire War. Night after night, the chat brought together a number of members of the system, up to one hundred of them, which was the highest number of participants the service could technically support. They chatted during the night, since that was the time when NATO attacks were predominantly performed. Chat sessions usually started just after the air raid alarms and together with the first explosions the number of participants increased. As soon as an explosion was heard, new chatters would join, asking: "What did they hit?" The Sezam¼s "alternative net of reporters" would then try to identify what had been hit. The participants from different parts of the city were giving eyewitness reports. An internal scale on the intensity of explosion was established, so the mark 10 meant that the window glass broke, 9 that the window glass shook, 8 that the explosion was very strong but with no obvious consequences and down to mark 1, which meant the situation was peaceful.

The following extract from the SezamPro chat session illustrates chatting at the moment of a NATO attack (that session began on 29 April 1999, at 00:48h):

> oops > it roars > wooow!!! > outch > where was the hit > uhhhhhhhhhhhhh > wow! > fuck > it roars above the very centre > what the fuck is this ?! > ADS [11] in the very centre > oh, fuck, there are explosions all over the city, crazy!!! > centre - 9/10 > 9/10 > Tas [12] 8/10 > the centre 10/10 > maybe the Pancevo bridge? > anyone seeing anything? > the windows are still shaking > the centre, near Vuk [13], it exploded like mad > Wow ... it means, towards the centre > 10/10 > honey, are you, O.K. ? > oh, fuck, why the fuck are they shooting planes above the city? > my chair flew away > we are having "fireworks" above the temple [14] > is it stronger in the centre than yesterday? > it is > stronger, you bet > aren't they attacking the Military HQ [15]? > Nemanja's Street [16]? > has anyone got the information where it exploded? > oh, fuck; people, chaos is outside!! > it sounded as if someone had pulled the trigger just below my window > studio b [17] says everyone should go to shelters > that's right > yes, it really is > what a horror, what a speedy chaos! > we're really beating them > the sky is clear tonight :( > no one seeing anything? > three flames of smoke somewhere near the partizan stadium [18] > ouch, the gravediggers [19] are in mourning > nooo, just don¼t touch my stadium! ;)

This segment illustrates a model that characterized almost every chat session on the SezamPro during the NATO attacks. After the explosion, there was an emotional outburst first ("wooow", "uhhhhh", "what the fuck") then reports on the situation in the participant's area followed ("the centre - 9/10", "near Vuk it exploded like mad") and finally the participants would try to find out what had been hit ("aren't they bombing the military HQ?", "Nemanja's Street?"). After the first explosion and shock, jokes would be cracked on the situation ("we're really beating them", "ouch, the gravediggers are in mourning"). Regarding the answers of the respondents (which will be quoted later in this paper) it is obvious that this humour was one of the basic elements that eased tensions in these difficult moments.

Air raids and explosions served as "triggers" for joining the chat and the majority of users saved their online time mostly for these moments (just to remind the reader, each participant were allotted 2.5 hours/day). When the chatters gathered after the air raid alarms, and there was no attack, the most common remarks would be "oh, I wasted my time, and nothing happened!" In these peaceful moments, the participants would slowly leave the chat, saying that they would come back "as soon as somewhere explodes". The owners of SezamPro, chat participants themselves, gave some extra time free of charge to the members, thus solving the problem of available online time.

The chat participants made efforts to pass information to one another. For example, those who had satellite dishes informed those who did not have them about foreign television reports. They also exchanged useful Web links and different pieces of information from "private" sources. This information could take the following form: "a relative from Canada told me that the bridges would be attacked tonight". This sort of information was common in Belgrade at the time. It is understandable why these rumours were spread; every new bit of information was considered useful, since news of any kind was hard to find. Being scared and uncertain, people tried to complete their view of the events and thus to prepare themselves for anything that might occur. However, rumors and misleading information had a negative effect, increasing panic without any reason. One rumor, for example, appeared on the SezamPro chat after NATO had struck the building of the Serbian Ministry of Internal Affairs, situated about 100 metres from the largest maternity hospital in Belgrade. One of the chatters spread "news from a confidential source" that all the babies born at the moment of attack were permanently deaf. This "information" caused real anxiety and unnecessary despair among participants. Subsequent medical analysis has shown that children born during the NATO bombing developed various types of physical and psychological disabilities, but universal deafness was not an immediate effect.

Many of the topics discussed in the chat were later elaborated in the conference system. For example, there were details about potential effects of depleted uranium in NATO projectiles and the ecological problems caused by the destruction of industrial facilities. SezamPro members, specialists in a variety of topics, distributed information according to their expertise and access to available facts. Such explanations were very useful, since they provided an example on how to act and protect oneself adequately. At the same time, these explanations identified problems objectively and provided some objective information, thus decreasing panic.

Wartime Chat and Political Opinions

Chat sessions were an important place where political attitudes could be expressed freely. In an officially proclaimed state of war, many citizens believed that expressing criticism of the regime or doubting officially proclaimed attitudes could result in repercussions. "Don't talk about it on the phone" - was a sentence often heard in Belgrade in those days and it indicated a high level of insecurity. However, the need to express political criticism remained.

"The critical consciousness of the society in the war and under the censorship could choose either 'dignified silence' or try as much as possible in media and other civilian kinds of resistance - to maintain the fire of critical discourse, necessary for a civilised existence of the society" [ 20].

The SezamPro chat was one of these "civil types of resistance". Virtual space functioned as an important substitution for a physical space - "problems" in the "real world" could be expressed in cyberspace.

In the chat, political issues were mostly discussed during peaceful intervals. People talked about the moves and statements of various politicans trying to predict further development of events. These debates often implied a severe criticism of the former Yugoslav president Slobodan Milosevic, as well as comments which the former regime would qualify as "treason". Political attitudes were mostly presented in the form of critical humour and ironical comments (as noted in aforementioned quote "we're really beating them").

The following example refers to the attitudes of chat participants towards the international politics of the former Yugoslav regime, and it illustrates some of the sarcasm:

> Lilic [21] is in Libya, asking Gadafy to take an active part in problem solving) > Brother Muammar, help! > Will they bring Gadafy¼s camels > We'll get the camels to attack the Apaches > new weapon: armoured camel > K - 117 A > nuclear camel > invisible camel > the camel spits and shoots down a f 117 > Stealth Camel > ;))) > ;)))

National media reporting was also criticised, particularly the Serbian State television, RTS. Here is an example of the participants' reaction to a certain RTS report:

> hey, yesterday news on RTS were really good: They ask people: "What do you think about Clinton?" (and then you can imagine what they think) and then "They accuse our media of being biased" and then an old fellow says: "What the hell the idiots mean by "being biased", we all think the same". : ))) > ; ))) > ; ))) > I am so sorry for not recording it, but who records the news anyway.

Such comments confirm the thesis that humour represents one of the most subversive ways of resisting violence. "Resistance by spirit" was the disclosure of contradictions, sometimes brought to absurdity, between reality and "official versions" of it. Humour was also a specific way to express anxiety and fear. In virtual communities, humour can also have an important role in the development of group identity and solidarity, as Nancy Baym (1995) noted:

"It is this transformative quality of humour which provides it with its social power to create and enhance participant solidarity and group identity. One way of doing it is by positioning all the group members as highly knowledgeable and competent readers of the genre" [ 22].

Baym also remarked that, according to John Morrell, sharing humour with others is a sociable gesture, and makes other people relax [ 23].

The fact that the SezamPro users expressed critical attitudes in the chat indicates that stronger interpersonal relations - and thus a higher level of trust among the participants - were established. These interrelationships could be the result of mutually experienced dramatic moments, such as the bombing. It also suggests that users perceived the chat as a place which was not controlled by the state (as compared to the telephone system, which was believed to be under state control and assumed surveillance) or, as the place in which remarks could be made without "real world" consequences.

Finally, it should be emphasized that these critical remarks did not imply a "pro-intervention" attitude, since an unjustified analogy was often made between "anti-Milosevic" and "pro-NATO" attitudes. Anti-regime attitudes did not imply justification of NATO actions. The best example is the chat session on the night when the business centre "Usce" was struck. This building was a base of the radio and TV station owned by Slobodan Milosevic's daughter. As soon as the chatters heard the news that this building had been struck, their reactions were of deep anxiety that some of the employees were killed. The political attitude did not overcome the human one. An identical situation occurred when the building of RTS was struck. Critical remarks for the way this station was informing the public were put aside when sixteen employees were killed in the NATO attack.

Wartime Chat and Socialising

Previously examined results showed that the SezamPro war-time chat functioned as a significant vehicle to gather information and to express political opinions. However, the chat acted as a place for social interaction.

During the War, particularly at night during the NATO attacks, social interaction was mostly reduced to small groups, predominantly the family members and neighbours. Both the NATO bombing and actions of the Yugoslav Army ADS, as well as darkening of the city, made it dangerous to leave home, except in the direction of shelters. In these circumstances, SezamPro's cyber space enabled what was physically impossible, gathering people facing the same problems, with the same need to express and share their feelings. For one of the respondents:

"the feeling that we are all in same trouble and a relief because of the possibility to get in touch"

was the most important characteristic of wartime chat. Another respondent explained the significance of people gathering online from his personal perspective:

"I live in a strategically important area [and] it was problematic to go out at night, so most of my social life was directed to the chat."

The respondents' answers also indicated that the exchange of information about the ongoing air raids in Belgrade was basically an act of social interaction as well:

"Commenting second after second what was going on, was a way to share fear with someone ... ."

For another informant:

" ... the sense of togetherness and the fact that no matter how much we tried, at the end of chat we knew nothing more about the hit places than we did before the chat."

was the most vivid remembrance of chatting.

In fact, it cannot be determined to which extent the chat participation provided a more precise and complete insight on the course and results of the NATO attacks. In certain chat sessions, the participants had really managed to identify the target before an official statement had been provided. In most cases, however, this kind of mutual informing included only assumptions and suppositions. Moreover, the scale for establishing the intensity of an explosion actually expressed the intensity of participants' impressions after the explosion and was more "an instrument for measuring the fear", than a mean to identify the exact location of an attack. Therefore, the real function of wartime chatting was to satisfy the need for human contact.

"Indeed ... the exchange of information is, in any human situation, fundamentally a social interaction rather than a mere instance of goal-oriented information retrieval or interaction with an information system" [ 24].

The socialising role of chat could also be seen in the change of participation motives. Before the war, fun was what most of the participants were looking for in the chat. After the war, socialising became the major stimulus. This fact suggests that during the bombing social relationships were developed within the studied group, and they persisted into the post-war period.

The informants' responds also show that wartime chatting had an important role in improving their psychological state. Half of the informants stated that it eased their anxiety:

"There were always some new pieces of information, I always knew what was happening, the presence of so many people, even in the chat, eased my psychological state."
"The time spent in front of the computer contributed to my overcoming certain things, with fun and a touch of fear. It helped me accept the aggression as reality, and not run away from it."
"It (chatting) contributed to releasing of tension, I was really glad to get in touch with someone else, who like me had escaped drafts and remained normal."

Furthermore, even two thirds of informants believed that they contributed to easing of other chat participants' tension. Here are some of their explanations about how they did, or did not, ease anxiety of their fellow chatters:

"Saying that it was not so terrible, and that it would be better."
"Talking about nicer things."
"At least with my presence."
"We understood that fears and nervousness were not individual things, but a part of some greater misfortune."
"By amusing others."
"By giving information on time."
"I don't think I could ease anyone's tension, bombs were bombs."
"The tension was too strong to be released by one person."

Finally, it is interesting to compare the characteristics of virtual community established within the SezamPro with wartime communities that existed in physical space. These "real" groups were mostly organised within shelters and improvised hiding places. The following quote briefly describe one of these groups:

"Life in shelters and hiding-places, in people's own and neighbours' cellars, with a radio or a TV set, does not offer much. Tired and worried faces, a game or two and endless talks on the topic: "What do they want?" and "What do we want? ... ." [ 25]

This fragment gives a picture similar to the one from the SezamPro community. Information gathering, socialising and talks on current topics - these were the basic elements of both the "real" and the SezamPro-based communities. To paraphrase Howard Rheingold (1993), at wartime, people in virtual communities do just about everything people do in real life.

"This has to do with 1) normal human needs, 2) considerably intensified by war, 3) which could be satisfied in some other way, but 4) the realisation of which was achieved through the Internet communication which enabled important and even crucial advances" [ 26].

Conclusion

In a state of crisis, a computer-mediated service became an important communication channel, enabling social interaction. Virtual space was not a bridge between geographically dispersed individuals, but an alternative gathering place in circumstances which disabled physical contact. Under these circumstances, strong social relationships were established online and the sense of togetherness was developed within the studied group. With the studied group, several kinds of behavior emerged:

Faced with "information hunger", the chat participants were not just "information gatherers", but active "information suppliers";

Although the users were under pressure themselves, they were trying to behave socially responsible and thus contribute to easing of other participants' anxieties; and,

Risky political attitudes were expressed within the community, freely.

The analysed chat channel was a specific "virtual shelter" that participants used in order to protect their critical attitude and to maintain an emotional balance. This way these "digital people" fulfilled some of their basic needs in cyberspace as others did in the real world.

Furthermore, this study points to important role of the Internet as a source of information in a crisis. This role emerged from the interactive nature of the medium, where it became possible for users to reciprocally inform each other in real time during specific events. This role of an Internet-based medium may have developed as a consequence of a noticeable mistrust of information from traditional media.

Finally, it should be emphasized that studying the use of the Internet during the NATO bombing provides only some answers to some significant questions in humanistic informatics. It cannot answer all of them, nor can it give a general conclusion. A variety of social, political, economic and similar "non-cyber" elements make this case unique. The use of the Internet in a crisis affecting a society with different cyber and real contexts would probably be different. If the topic was not so morbid, one could say that more empirical data from similar occurrences would be needed if we were to develop some generalities about the behavior of "digital persons in a digital war". However, all one can hope for is that the "Yugoslav case" will remain the solitary example.

About the Author

Smiljana Antonijevic is a graduate student in the Department of Social Anthropology in the School of Philosophy at the University of Belgrade.
E-mail: rosa@sezampro.yu

Acknowledgements

This paper was originally presented at at "Internet Research 1.0: The State of the Interdiscipline, The First Conference of the Association of Internet Researchers", 14-17 September 2000, at the University of Kansas in Lawrence, Kansas.

Notes

1. Five studies were presented to the panel. Conducted during the War, they resulted from the specific, threefold role of its authors - the researchers, Internet users and war participants. The studies were focused on the role of the Internet as a source of information (Naumovic), the Internet as an "informational tool" (Milenkovic), as well as on the character of wartime CMC within professional online communities (Vucinic, Milinkovic) and among "ordinary" users (Antonijevic). For more information see Naumovic (1999).
2. "Sezam" is Serbian for Sesame, while "Pro" stands for provider.

3. This system of temporal restrictions to online access changed at the end of 1999. Currently, there are no daily limits but instead monthly portions.
4. Amnesty International, 2000, paragraph 1.

5. Tripkovic, 1999, p. 130.

6. Loges, 1994, p. 5.

7. Naumovic, 1999, pp. 258-259.

8. Burnett, 2000, paragraph 9.

9. Op. cit., paragraph 10.

10. After several hours it was found out that the police had stopped and banned broadcasting of this radio station and the editor-in-chief of B92 had been arrested.
11. Air defense system.

12. Tasmajdan, a park in Belgrade downtown.

13. Vuk Karadzic monument, situated in downtown Belgrade.

14. The church of Saint Sava, situated in downtown Belgrade.

15. The Yugoslav Army Headquarters, the building situated in the Belgrade downtown.

16. The street in the Belgrade downtown.

17. The television station of the Belgrade city government.

18. The stadium of Belgrade football team "Partizan", situated a bit out of the city center.

19. "Gravediggers" - a nickname of the "Partizan" football club fans.

20. Pancic, 1999, p. 210, emphasize in original.

21. Former Vice President of the Yugoslav Government.

22. Baym, 1995, paragraph 78.

23. Op. cit., paragraph 90.

24. Burnett, 2000, paragraph 9.

25. Gligorijevic, 1999, p. 107.

26. Naumovic, 1999, p. 258.

References

Amnesty International, 2000. "NATO/Federal Republic of Yugoslavia "Collateral Damage" or Unlawful Killings? Violations of the Laws of War by NATO during Operation Allied Force," at http://www.amnesty.org/ailib/intcam/kosovo/index.html, accessed 28 December 2001.
N. Baym, 1995. "The Performance of Humor in Computer-Mediated Communication," Journal of Computer-Mediated Communication, volume 1, issue 2, at http://www.ascusc.org/jcmc/vol1/issue2/baym.html, accessed 28 December 2001.
G. Burnett, 2000. "Information Exchange in Virtual Communities: A Typology," Information Research, volume 5, number 4 (July), at http://www.shef.ac.uk/~is/publications/infres/paper82.html, accessed 28 December 2001.
Critical Art Ensemble (CAE), 1995. "Utopian Promises - Net Realities," at http://www.critical-art.net/, accessed 28 December 2001.

B. Dimitrijevic, 1999. "Rat NATO-a protiv Srbije 1999: Vojni aspekti i posledice," Nova srpska politicka misao, Srbija i NATO, volume 1, pp. 161-174.

J. Fenerback and B. Thompson, 1995. "Virtual Communities: Abort, Retry, Failure?" at http://www.well.com/user/hlr/texts/VCcivil.html, accessed 28 December 2001.
D. Gligorijevic, 1999. "No i dan," In: D. Tripkovic. Beograd pod bombama. Beograd: Skupstina grada Beograda, Sekretarijat za informisanje, p. 107.

J. Katz and P. Aspden, 1997. "Cyberspace and Social Community Development: Internet Use and Its Community Integration Correlates," at http://www.nicoladoering.de/Hogrefe/katz2.htm, accessed 28 December 2001.
R. Kraut, V. Lundmark, M. Petterson, S. Kiesler, and W. Scherlis, 1998. "Internet Paradox: A Social Technology That Reduces Social Involvement and Psychological Well-Being?" American Psychologist, volume 53, number 9, at http://www.apa.org/journals/amp/amp5391017.html, accessed 28 December 2001.
W. Loges, 1994. "Canaries in the Coal Mine: Perceptions of Threat and Media System Dependency Relations," Communication Research, volume 21, number 1 (February), pp. 5-23.

S. Naumovic, 1999. "Netwars: Internet i agresija na Jugoslaviju," Nova srpska politicka misao, Srbija i NATO, volume 1, pp. 253-276.

T. Pancic, 1999. "Tisina, ratujemo! NATO intervencija, intelektualci i javnost," Nova srpska politicka misao, Srbija i NATO, volume 1, pp. 207-212.

M.R. Parks and C. Floyd, 1995. "Making Friends in Cyberspace," Journal of Computer Mediated Communication, volume 1, number 4, at http://www.ascusc.org/jcmc/vol1/issue4/parks.html, accessed 28 December 2001.
E.M. Reid, 1991. "Electropolis: Communication and Community On Internet Relay Chat," adapted from an Honours thesis written at the University of Melbourne (Australia), at http://www.irchelp.org/irchelp/misc/electropolis.html, accessed 28 December 2001.
H. Rheingold, 1993. The Virtual Community: Homesteading on the Electronic Frontier. New York: HarperPerennial, and at http://www.rheingold.com/vc/book/, accessed 28 December 2001.

D. Tripkovic, 1999. Beograd pod bombama. Beograd: Skupstina grada Beograda, Sekretarijat za informisanje.

Editorial history

Paper received 17 December 2001; accepted 27 December 2001.

Copyright ©2002, First Monday

Sleepless in Belgrade: A Virtual Community during War by Smiljana Antonijevic
First Monday, volume 7, number 1 (January 2002),
URL: http://firstmonday.org/issues/issue7_1/anton/index.html