This file contains the text of "An Emerging Form of Human Communication: Computer Networking," a paper presented by Howard Rosenbaum and Gregory B. Newby at the Annual Meeting of the American Society for Information Science (Toronto, November 1990). The proceedings, in which this paper appears, was published by Learned Information in Medford, New Jersey. Feel free to share this information with your colleagues or quote these materials, provided proper citation is included (note that Learned Information owns the official copyright to the printed version). The authors can be contacted at: /* Gregory B. Newby, Assistant Professor in the Graduate School of Library */ /* and Information Science, University of Illinois at Urbana-Champaign */ /* 501 E. Daniel Street, Champaign IL, 61820-6212. E: gbnewby@uiuc.edu */ /* V: 217-244-7365 F: 217-244-3302 W: http://alexia.lis.uiuc.edu/~gbnewby/ */ -------- An Emerging Form of Human Communication: Computer Networking by: Howard Rosenbaum Gregory B. Newby School of Information Studies Syracuse University A Paper Presented at the Annual Meeting of the American Society for Information Science. Toronto, Ontario November, 1990. 0.0 ABSTRACT Computer networking is an emerging form of communication which is having major societal and cultural impacts. We first focus on BITNET and Internet, which are parts of a worldwide computer network for researchers, academicians, and information professionals. We discuss some of the services and resources that are available on the network, describe ways that these services can be accessed and used, and suggest several directions for research that we believe will be will be significant in understanding the impacts that computer mediated communication will have on social interaction, organizational structure and culture. Throughout, computer networking is treated as an emerging form of human communication. Next, there will be a discussion of directions in which research has been moving, and we will conclude with some suggestions for future research. This review of the research literature will demonstrate that, while there are certainly provocative findings, there is a need for a new research perspective to investigate CMC as an emerging form of communication. This perspective, informed by theories and methods from the social sciences, will attempt to understand the impacts of CMC by focusing on the perspectives and experiences of network users. 1.0 INTRODUCTION 1.1 Plan of the Paper A person can sit at a terminal in her New York office and, with a few keystrokes, can communicate with a colleague halfway around the world. She can send a file containing the draft of an article on which she is collaborating for her colleague's comments. Next, she can send messages to committee members, arranging a meeting for the end of the following week and indicating that they should only respond if they have schedule conflicts. Then she can use her terminal to access a database located on the West Coast and retrieve a copy of some statistical software that she wants to use to analyze data that her colleague has sent to her. Finally, when her work is finished, she can check in on BITNET to monitor an online conference where people from around the world are carrying on a discussion about possible ways that electronic links can be established with researchers in the Soviet Union. There will be a number of messages in her electronic mailbox that she can read and then save or discard. If there is a comment she wishes to add to the discussion, she can compose her response and easily post it to the discussion list, where it will be distributed to all the members on the list. All of these activities can be accomplished by accessing BITNET, an international computer network managed by EDUCOM which links universities and research institutions around the world for the cost of the telecommunications connections and the requirement that each member bring a new member institution online. As people use BITNET and other networks, and integrate networking into their work routines, they are participating in an emerging form of communication called computer mediated communication (CMC). As people become more proficient in this form of communication, they will change the ways in which they conduct research, write articles, and communicate with colleagues and other network participants. This work begins with a discussion of the fundamentals of computer networking, and an introduction to BITNET and Internet. Then, we will examine some of the services typically available on networks. These include: ¥ personal or private messaging, commonly called "electronic mail" ¥ asynchronous file transfer ¥ file and mail servers ¥ mail lists or interest groups, commonly called "computer conferencing" ¥ interactive or synchronous communication ¥ synchronous file transfer ¥ remote sign-on. Each of these network functions will be discussed in practical terms, and their significance as new forms of human communication will be developed. Next, there will be a discussion of directions in which research has been moving, and we will conclude with some suggestions for future research. Some of the areas which will be covered include: ¥ the emergence of social norms governing CMC ¥ the use of communications theories to make sense of this emerging form of communication ¥ the effects of CMC in organizational settings, including academia, industry, and government ¥ the social psychology of CMC ¥ the education and training of CMC users ¥ communication issues raised by the ways in which CMC is changing organizational structures and processes. This review of the research literature will demonstrate that, while there are certainly provocative findings, there is a need for a new research perspective to investigate CMC as an emerging form of communication. This perspective, informed by theories and methods from the social sciences, will attempt to understand the impacts of CMC by focusing on the perspectives and experiences of network users. 1.2 The Problem The basic argument of this paper is that CMC is an emerging form of communication which is changing the ways in which people in academic, governmental, and industrial settings are interacting with one another. The example with which this paper begins does not represent a level of sophistication with computer networking which is only available to a privileged elite. It does describe a set of activities which any computer literate person with access to a network such as BITNET can quickly learn and begin using. Acquiring the abilities to exploit the power of computer networking and to integrate CMC into one's work causes a range of fascinating changes in the individual and the organization within which the individual works. At the very least, the acquisition of computer networking skills is a form of empowerment which increases the individual's abilities to exchange and manipulate information. This occurs because CMC, which does share some characteristics with other, more traditional forms of communication, also has novel characteristics which have yet to be taken into account theoretically and methodologically. The vast potential that CMC has to fundamentally alter communicative behaviors and social interactions is just beginning to reveal itself to academic researchers. This is important, because much of the research on CMC has focused on the design, implementation and engineering aspects of computer networking and has virtually ignored its status as a new and different form of communication. There is a need for a different direction in research, one which focuses on the psychological, social, and cultural impacts of computer networking on people, organizations, and society. This paper has two main goals. First, it will provide information about the fundamentals of computer networking so that researchers, academicians, and information professionals can begin to explore CMC on their own. Towards this end, there is an extensive discussion of networking services that are available on BITNET, a network to which most academic institutions have access, and other networks, such as Internet. Second, we review some of the literature which has investigated CMC in a variety of settings, in order to draw out some of the major findings concerned with CMC as an emerging form of social interaction. We will conclude with some suggestions about directions for future research. 2.0. Fundamentals of Computer Networking 2.1 The technical side A computer network consists of nodes, and connections among the nodes (links). For BITNET [1], most of the nodes are mainframe computers, and most of the links are 9600 baud dedicated phone lines[2] between one node and the node(s) adjacent to it. Many organizations support networks for local use: LAN's (Local Area Networks). LAN's often operate at speeds much greater than 9600 baud [3]. The major national computer network aside from BITNET, Internet, supports single-user microcomputers and other small systems, in addition to mainframes [4,5]. Human and computer networks are the focus of a entire field of study [6]. Network theorists recognize three fundamental parts to any network: nodes, links, and structure. The structure of a network determines the path which any signal must take between nodes. In a computer network, nodes are often referred to as hosts, and signals are called messages. Because BITNET has a fixed structure, with only two or even one link terminating at each host, there is usually only one path for a message to take between any two hosts. This means that whenever a host along the way is offline or very busy, messages can be delayed. BITNET has plans for upgrade to a more open structure, allowing for more direct paths and faster transmission than the current structure [7]. As human communication requires sets of rules which allow us to understand one another, computer networks require rules in order for them to transmit messages. These rules are stated in the form of protocols. The protocol used by BITNET follow RFC822 [8], and other protocols to embed each message in a package that allows each host to identify where the package is to be sent. These protocols are important, because different types of computers use different methods for representing and processing information (for instance, the internal representation for the letter "A" on IBM mainframes is not the same as on VAX mainframes). Since different networks use different protocols, it is necessary to have hosts between the networks which understand both protocols, and are able to translate between them. These hosts are called gateways. Gateway machines act as central switching stations between BITNET and Internet, and between North American Internet hosts using the TCP/IP protocol, and European hosts using the x400/x25 protocol. Because of this ability to translate between networks which use different protocols, it is a simple matter to communicate electronically with users on other networks all over the world. 2.2 Functions of computer networks (Table 1 here: Network activities in brief) Despite the wide range of physical configurations of computer networks, there are only a few different types of functions for which they are used. These capabilities are separated into two broad categories, asynchronous and synchronous communication. Asynchronous communication is non-interactive. For practical purposes, asynchronous communication occurs when messages are held before they are passed on. Synchronous communication refers to interactive processes, when messages are transmitted quickly from place to place. On computer networks such as BITNET, where messages are necessarily passed on from one host to the next, the difference between asynchronous and synchronous communication is mainly one of the priority given to each form of network traffic. A given network may not have the ability for all the types of network activities described below. New technologies are adding to the list of computer network activities -- voice mail is an example of a relative newcomer to the computer networking scene. For each of the network activities below, the discussion starts with a general introduction, gives an evaluation of the qualities and characteristics of the activity, and then concludes with an evaluation of the activity as a form of human communication. 2.2.1 Asynchronous network activities 2.2.1.1 Electronic mail or email Description: Email is the function most often used by most network users. Email is a specialized form of file transfer (below). Email is also the only one of the forms of electronic communication listed here which can be used to access virtually all of the computer networks linked by gateway hosts. A user creates an email message the same as she would edit a normal file on her host computer. The computer's email program solicits the electronic address of the intended recipient, and allows for the insertion of a subject line and other optional fields. The computer then sends the message along to the destination as specified in the address field, which may include sending the message to a gateway for further processing. Although the mechanics of email differ from computer to computer, mail received takes the general form of a file with a header specifying the return address of the sender, the subject line, and other information which often includes a record of hosts through which the message passed on the way from its source. This header is followed by the text of the message, as typed by the sender. Qualities and characteristics: Email is an efficient, cheap, and fast alternative to the regular postal service or the telephone. Depending on the time of day, network load, and how "far" (in terms of network hops) a message has to travel, email takes anywhere from a few seconds to overnight to reach its destination. Longer delays are sometimes produced when hosts on the network are unavailable, or links between hosts are down. Email as human communication: Email is a strong example of an emerging form of human communication. Limitations of email include, on the human side, misunderstandings which result from users who do not grasp the rules for this form of communication and how they differ from other media. On the technical side are difficulties associated with getting the physical message to the intended recipient: difficulties in figuring out the appropriate return address to a sender; the lack of a comprehensive directory of electronic addresses; and problems with knowing whether a message did not get through, or whether the recipient simply has not looked in his electronic mail box. The uses for email overlap but do not include the uses for regular postal service mail, the telephone, the telegram, the facsimile (fax), the interoffice memo, and other forms of communication. Similarly, the rules [9] of email overlap with other forms of communication. Email can be used for many of the same purposes as other media, but does not necessarily have the same effects. Email may be sent to individuals or groups, who may or may not know each other, and who may be subordinates, supervisors, or equals. As such, email can take the form of an interoffice memo, or a proposal for research, or a casual introduction of a friend or an idea, or an intimate discussion between close friends... Despite the flexibility of email for supporting different forms of communication, email is limited to the capabilities of the computer that produces it, and the network that carries it. Mainframe text editors are not the same as word processing programs -- they do not allow for the inclusion of graphics, underlines, boldface, colors, scientific equations, or almost anything other than simple text. The sender is limited by his creativity to produce the special effects which text alone cannot supply [10]. An anomaly for email is the contrast between the large number of purposes to which it may be put, and its narrowness as a channel for communication. Advances in networking will eventually bring about the necessary protocols for the transmission of graphics, bold and underline, equations, and colors. Until then, there is a very real danger of email being misunderstood, because the channel is insufficient to transmit the desired message. Much of the negative affect produced by email, known as "flaming," results from misunderstandings that probably would not have occurred in other communication contexts. The fast transmission of email and the ability to quickly compose and send responses and comments combine to produce an environment where tempers may flare, and people may say things without taking time for reflection. Experienced networkers know that a message that offended may be reinterpreted in a more favorable light on another day, and they know the wisdom of holding their message for a few hours of days to ensure that it says what they want it to say. The correction mechanisms brought into play when the rules of email are broken are not well established. Institutions which allow their students or employees access to computer networks are currently wrestling with the problem of supplying adequate education to their users, in the absence of a clear curriculum for doing so. BITNET has some of the most clearly stated rules for acceptable conduct, but those who violate the rules are seldom punished. The socialization of new users of computer networking is necessarily slow and disorganized, because the social forces affecting networking interactions are in a state of flux and redefinition. The future of email will see the introduction of protocols for the transmission of richer messages. It will see standardization of policies across networks. Most importantly, the social norms associated with email in all its different forms will solidify, and eventually be as firm a part of society as are the rules for the telephone. 2.2.1.2 Asynchronous file transfer Description: Asynchronous file transfer is the general class of network activity that provides for email. Again, the mechanics for sending and receiving files differ from computer to computer. Unlike email, file transfer is limited to networks which use the same protocol. This is because the standard header which accompanies email is not included in the file -- only the file itself and the electronic addresses of the sender and recipient are present. Qualities: Asynchronous file transfer is frequently used to circumscribe the limitations of email. Email is usually limited to text only (no special format files, such as executable program images). The width of email text is limited to 80 columns, and, at some hosts, there may also be a maximum number of records (lines) that a message can contain. Some computer systems have formats for sending files which protect specially formatted data from corruption as they are passed on. This form of file transfer is used primarily for the sending of files from a user to one or more known users. The files may be of a technical nature, such as computer programs or data. As a special purpose form of communication, asynchronous file transfer does not occur in the social context of email. The transfer of files may be seen as analogous to sending a package by regular mail -- the package has some utility to the receiver, and is usually expected. To continue the analogy, a transferred file may be accompanied by an email note announcing the file and describing its use. Asynchronous file transfer as human communication: As networks such as BITNET move to adopt the more flexible and efficient means of file transfer associated with Internet, the use of asynchronous file transfer will probably diminish (see the section on synchronous file transfer below). Current limitations of file transfer often prevent specially formatted files (such as executable images) from being transferred across networks uncorrupted. Solutions include standards for encrypting binary files into text-only files, so that they may be transferred as email (uuencode is a popular encryption routine). 2.2.1.3 File and mail servers Defined: File and mail servers are hosts which allow access to their resources to people who are not registered users on that host. On BITNET, Listserv is the program most often associated with file and mail service. Listserv programs run on IBM mainframes, and there are a number of them available over BITNET. Each listserv may maintain and distribute several different mailing lists, keep archives for the lists, and also support the distribution of any number of computer files. Listserv is used by sending synchronous or email messages to the listserv address (see Appendix A). Files and responses from Listserv may be sent via interactive messages, email, or as files. The mail service side of file servers is discussed more fully in the following section on computer conferencing. Qualities and characteristics: The nearest analogies to network file service are the public library, or perhaps the information clearinghouse. Whether the file retrieval is achieved asynchronously or synchronously (see the section on synchronous file transfer below), the file was placed on the server at some time in the past. Whoever produced the file did not know the audience personally. File servers may keep interested users informed of new items on their directories, and some hosts have reputations for particularly rich collections. On BITNET, a particularly important collection of files is stored on the host BITNIC. BITNIC is named after the Bitnet Information Center. The userids LISTSERV, DOCUMENT, and INFO are each associated with a large number of files and archives pertaining to computer networking. COMSERVE @ RPIECS is one of the best known and organized file servers on BITNET and Internet which is not Listserv-based. Although it has a focus on topics having to do with the study of human communication, the database and mailing lists range as far afield as discussions of gender roles and philosophy. File and sail servers as human communication: File servers do not have a strong set of rules associated with their use as a form of human communication. Similar to the public library, file servers generally keep wide ranging databases, without much concern for censorship -- so long as the files submitted for inclusion fit the general purpose of the databases. This can have the same negative effect that occurs in a library, when the producers of files are not well informed, or when the items in the database are outdated but not removed. 2.2.1.4 Computer conferencing Defined: Computer conferencing is a special form of human communication where computers enable the distribution of messages to a wide audience. The most common forms of computer conferencing are mailing lists, and bulletin board systems (BBS). Mailing lists consist of a list of participants, and a central distribution system. On BITNET, these lists are maintained by Listserv (see the previous section). At press time, there are almost 2000 public mailing lists available from BITNET Listserv programs. Topics range from the administrative to the academic to the recreational. Some lists generate only a message or two per month, others generate dozens each day. Internet mailing lists are similarly managed, but use management and distribution systems other than Listserv. Qualities and characteristics: Mailing lists and BBS have substantially different distribution systems. Unlike mailing lists, where each user gets a copy of each message to the list, the general form of a BBS is that messages are not distributed to each subscriber individually. Instead, they are kept in a central location for perusal by interested parties. BBS are not well suited for BITNET, because of the lack of an effective synchronous communication system for browsing a remote BBS. However, BBS are often maintained on individual hosts. The host may be a mainframe connected to BITNET or other networks, or it may be a microcomputer. Advances in microcomputer technology have allowed the development of software which turns a single user microcomputer into a BBS manager, which interested users may access with a modem and a phone line. A special form of computer conferencing is Usenet. Usenet is to the UUCP network what the Listserv mailing lists are to BITNET. UUCP is a network for Unix machines, but it has lost much of its identity to Internet. Now, hosts may access Usenet through Internet, without being part of UUCP. On Usenet, each host keeps its own central database of newsgroups (newsgroups are analogous to BITNET mailing lists). Usenet is hierarchically organized, so that individual hosts may subscribe to a selection of hierarchies, and a choice of groups within each hierarchy. Some newsgroups may be only for local distribution, and senders of messages (sending a message on Usenet is called "posting") may specify anywhere from local to world-wide distribution of their messages. There are hundreds of world-wide newsgroups. An important advantage of Usenet over BITNET mailing lists is that the software used to access the Usenet database on a particular host is considerably more sophisticated than are most email utilities. Rather than having all the subscribed-to mailing list content put in each subscriber's electronic mail box, the program is used to browse the central database. Thus, topics of no interest may be skipped, newsgroups may be subscribed and unsubscribed to at will, and the size of the database remains the same, no matter how many users access however many newsgroups. Computer networking as human communication: All forms of computer conferencing have opened the way for the emergence of new forms of human communication. It is in these forums that the rules for email have been refined, tested, and enforced. Most of the rules associated with email are also associated with computer conferencing (see the section on email, above). Conferences in their various forms allow users with common interests to meet and interact. Users may gain a wide audience for their opinions. Unfortunately, the public nature of most computer conferences leave the networks open for abuse: users who send messages inappropriate to the topic, users who are abusive towards other users, users who decide that what they have to say is important enough to send to every mailing list they can find [11]. The near anarchy of BITNET, and the virtual anarchy of Internet, makes for difficulties in enforcing the rules which most users accept. Disciplinary action is left to the host where the offending user committed his acts, and is often considered ineffective or insufficient. Like human communication, the rules of computer networking are not all explicitly stated -- they are learned through experience. Relatively few institutions have taken the initiative to insure that their users are educated in the ways of computer networking before they are allowed network access. 2.2.2 Synchronous Communication 2.2.2.1 Interactive messaging Defined: Interactive messaging occurs when users send message to other users, and the transmission delay is short enough that electronic conversations can be held. BITNET can be used to send single-line messages to particular users much in the same way as email or files are transferred. Since single-line messages are much shorter than email or files, they are sent with a high priority. Provided that all the links between the conversing users' hosts are up, the delay in sending a line message and receiving a response is only a few seconds. Qualities and characteristics: A program called Relay facilitates the sending of line messages over BITNET, creating an environment modeled after citizen's band radio. Relay allows for users to select private or public channels, and to join in conversation with up to ten or so users scattered across the network [12]. Individual hosts may also have utilities for interactive messaging between users on the host machine. DEC's Phone utility allows for multiple users to participate in a conversation with almost no delay in the sending of messages. The Unix Talk program provides a similar screen-oriented environment. Interactive messaging as human communication: Interactive messaging is a strange form of communication. It is not as effective as email or a telephone call. It is often confusing with more than two parties. However, this form of communication is a favorite of undergraduates, and of new users of computer networking. It is a way of introducing oneself to strangers, and of discussing trivial things [13]. Since interactive messaging is an expensive use of computer resources, many institutions have chosen to limit or eliminate users' access to local or network interactive messaging systems. 2.2.2.2 Synchronous file transfer Defined: Synchronous file transfer may occur when the user has an authorized account on both hosts involved in the transfer, or it may involve anonymous access to one host (where the user does not have a registered account on the second host). In both cases, synchronous file transfer over Internet is facilitated by FTP (file transfer protocol). Other networks may have their own standards for synchronous file transfer. Synchronous file transfer is not available over BITNET. Qualities and characteristics: On Internet, public file servers take the form of hosts which allow anonymous FTP access. FTP allows users to retrieve files synchronously from other hosts. FTP does not have the limitations of email or asynchronous file transfer, because the transfer occurs directly between the two hosts involved. This means that files in any format may be transferred. Anonymous FTP has become a way for computer programs and utilities for a wide variety of different computers to be accessed from a single site. Some hosts have emerged as central information distribution sites [14]. Synchronous file transfer as human communication: Administrators of machines that allow anonymous FTP access are generally more cautious about what is available on their host than those which provide asynchronous file transfer service. This is because the files at anonymous FTP sites are not only text, but are also programs. Computer programs distributed by anonymous FTP are part of the public domain. Programs in the public domain may not be sold for profit [15]. Although the administrators of anonymous FTP sites state that they are not responsible for any problem that users have with software retrieved from their site, the administrators will typically make some effort to insure that the programs they distribute are functional, and free of worms and viruses [16]. Most other human factors are similar to those of asynchronous file transfer. 2.2.2.3 Remote sign-on Defined: Remote sign-on is an important form of synchronous communication which is not presently provided by BITNET. It was developed as one of the original purposes of DARPANet and NSFNet (both of which are included in what is now referred to as Internet). Since important computing resources need to be accessed by users with a wide variety of geographic locations, remote sign-on allows for authorized users to access these resources, usually with higher communication speeds and less expense than with a modem and telephone line. Although further distance (in network hops) usually makes for some delays in transmission of data, remote sign-on really is the next best thing to being there. Qualities and characteristics: Unlike anonymous FTP, which uses very similar mechanisms, remote sign-on users must be authorized to use both hosts involved. Otherwise, the environment for remote sign-on is the same as when using a computer directly. The human communication factors are those which apply for using a computer directly: they depend on what the user does on the machine. 3.0. Focus on BITNET BITNET is the most familiar computer network to academics. It links thousands of academic and research institutions around the world. BITNET (from "Because it's time network" [17]) is the official name of the network which exists only in the United States. However, BITNET- like networks also exist worldwide. In everyday usage, BITNET refers to this worldwide network. The governing body of BITNET is BITNIC, run by EDUCOM out of Washington, DC. (Table 2 here: BITNET membership) BITNET allows academicians, students, researchers, professionals, and information professionals to communicate with one another. As these groups integrate computer networking into their daily routines, they are changing the ways in which they work and interact with each other. They are helping to shape an emerging form of human communication. 3.1. Some particulars of BITNET BITNET is a "store and forward" network. This means that as items are transferred from host to host, they must wait for the messages before them. If a link or host is unavailable, the message must wait for the channel to be restored. Shorter messages are put through first, allowing for interactive messaging (above). BITNET supports email, file transfer, and interactive messaging. BITNET does not support synchronous file transfer or remote signon. Internet's boundary with BITNET is most clearly defined by the difference in the network's capabilities: Internet is used for email, synchronous file transfer, and remote signon. Internet routes messages dynamically, allowing for multiple paths between hosts. There are many machines which are hosts on both networks. To become a host on BITNET, an institution must meet the requirement of being not-for- profit. Then, the institution must agree to pay for a link between it and another BITNET host. It must further agree to allow another new host to connect to it. Users can then get access to BITNET with no charge, and without producing a great amount of system overhead. Finally, the institution must agree to uphold the rules of BITNET, and be responsible for proper network usage by its members. This last point is a sticky one, because proper conduct is neither well defined nor strictly enforced at this time. The membership requirements for BITNET have produced a decentralized network. There is usually only one path between hosts on the network, which makes for trouble when hosts become unreliable, or links become unavailable. Some central hosts help alleviate the problem, and also act as gateways between BITNET and other networks. 4.0. Some networking tips 4.1 Social Rules for Computer Networking A useful approach to computer networking is to understand the similarities between forms of communication that are more familiar, but also to understand the unique qualities that computer networking possesses. Computer networks allow for "invisible societies," where groups of computer users have conversations and trade information. To join in a conversation, all one really needs is the electronic address of the mailing list, or access to a machine running Usenet, or a modem and the number for a local or national bulletin board. Although most people do not remember the first time they used the telephone, they probably remember the first time they used email. This does not mean that email has any more difficult rules of conduct than the telephone does. It simply means that it is necessary to invest some time in learning the ropes, just as is necessary for mastering any new social situation. (Table 3 about here: Points for effective computer mediated communication) 4.2 Technical Tips for Computer Networkers Aside from the difficulties in mastering the social rules for computer networking, it is necessary to learn, to some extent, the more technical aspects of networking. This is akin to moving to a new city, and having to learn to successfully navigate the streets. As in moving to a new city, where it is not necessary to know each back road and establishment in order to just get to work and the supermarket, the amount of familiarity with the technical side of computer networking needed depends on the sorts of networking activities in which users will be engaged. Because of the vast number of hosts accessible from any host on any network (not to mention the number of users on each host), there is no central list of all possible hosts and valid routes to each host from each other host. Instead, each host will have a limited picture of the worldwide computer network -- enough to route a message to another host on a particular network, or to identify which gateway a message must be sent to for forwarding to another network. Although conventions for embedding addresses within addresses are almost universal, allowing for each host along the line to correctly process and forward a message, it is not always easy to figure out the correct sequence of gateways to use. Since computers take things quite literally, a message may become lost when a single character is out of place in the address. An emerging convention is for institutions to have a userid CONSULT available on each host. By sending email to this userid, it is possible to get assistance with a variety of topics, including the correct format for an address. Many hosts have a userid POSTMASTER or POSTMAST which is managed by the institutional representative who is in charge of network activities for that host. Other conventions for host representatives vary from operating system to operating system. These representatives may be consulted to confirm whether a message is being received correctly, or to report questionable behavior by a user at that host. The list of information sources on the worldwide computer network is virtually endless. Each host, each user, represents a store of information that may be tapped with a few keystrokes on a computer. Fileservers and anonymous FTP sites provide a tremendous resource to be explored at leisure. As the reference section of a scholarly paper leads to further relevant works, a search on a fileserver or FTP site almost invariably yields more sites to search. It is perhaps unfortunate that no comprehensive index of network information sources exist, but such an index would be outdated before it was complete. It is not necessary to have a map of the worldwide network on hand in order to use it. Indeed, it is questionable whether such a map could ever be constructed [18]. It is also not necessary to be nervous about joining electronic discussions for the first time. As in any other human endeavor, there are more and less desirable people who you will meet; more and less useful information resources. Computer networking may further social contacts, professional affiliations, and information sources. The only way to start your own personal network with yourself at the center is to go ahead and start networking! 5.0 Current Research on Computer Mediated Communication 5.1 Introduction This section examines the literature concerned with research into the phenomenon of CMC. For the purposes of this paper, CMC will refer to any exchange of messages through electronic media which involves the use of computers. Information networking technology is a medium through which CMC takes place. The relationship between technology and social change is complex and difficult to unravel, although many agree that there is a "cultural lag" between the two, with technological advances taking the lead and a myriad of social changes following behind them. Telecommunications and information networking technologies have engendered many changes in the worlds of work, the university and the home [19], not the least of which has been the emergence of CMC, which has been described as a "new form of enhanced human communication" [20]. 5.2 The Problem Much research has been done on face-to-face communication and its related forms, but there has been comparatively little research on CMC, particularly from the point of view of CMC as a set of social interactions which create and maintain social relations among people and groups. There is a literature which focuses on CMC, but it has been preoccupied with the efficiency of this form of communication, its technical capabilities, costs, and the processes involved in its design and implementation in various organizational settings. It has not tended to focus on the ways in which CMC is changing the nature of communication and society [21, 22, 23]. This paper assumes that research from a user-based social science perspective is important and necessary if the impacts of CMC on people, groups, institutions, and societies are to be understood. The research on CMC is broad-based and, over the last two decades, has been produced by researchers working in the fields of communication, management science, computer science, sociology, social psychology, psychology, office automation, and information science ([24]. Researchers have investigated computer conferencing, electronic mail, bulletin boards, and the social psychology of CMC. Most of the research has focused on the impacts of CMC on individuals and small groups; relatively few studies have taken the organization as their unit of analysis [25]. With a few exceptions, these studies have been conducted in either academic settings, typically focusing on college students, faculty or employees, or in industrial settings, typically focusing on managers and occasionally on employees. Only a few studies have examined the effects of CMC on the scientific and academic communities, and even fewer have focused specifically on how scientists and information professionals make use of networks in their research. The field research has tended to examine CMC systems in two different environments: universities and academic science research communities [26, 27, 28, 29, 30, 31, 32, 33, 34, 35], and industry, including research and development divisions [36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46]. Other settings that have been the foci of field research include governmental agencies ([47, 48], public networks [49, 50, 51], libraries [52], and French households [53]. Many of these studies are descriptive in nature, focusing on the hardware and software of existing CMC systems and, to some degree, on the usage patterns. The quantitative data that is presented is also descriptive and typically provides such information as the number of messages sent and received during a given time period, where the senders and recipients are located in the network, the amount of experience the user has with the system, the number of words used in messages, the percentage of time during the work day devoted to CMC and, in some cases, the demographic characteristics of the users of the system. There have also been experimental studies of CMC which have examined interaction patterns in computer conferencing [54] and in synchronous communication [55]. Experimental studies have also focused on the ways in which CMC affects interpersonal communication; these have been conducted in laboratories [56, 57, 58] and through field experiments [59]. The main goals of these studies seem to be to: ¥ Describe existing systems ¥ Assess system performance in order improve the efficiency of existing systems ¥ Determine the factors that enhance or hinder the implementation and use of CMC systems ¥ Observe the ways in which the use of CMC in laboratory settings changes the ways in which people behave and interact. Despite this body of research, the question of the impacts of CMC on the individual and society is far from settled. According to Steinfeld, little research has been done on "...the social behaviors linked to the use" of CMC systems [60]. In the research that has been done there are contradictory or inconclusive findings. In part, this is a problem of the existing research, both in terms of the ways in which the research is conducted and the resulting findings. 5.3 Towards a Solution We have seen that the body of research to date on CMC has not satisfied the need for an understanding of its influence on social life. This research does provide considerable documentation which argues for the positive impacts of computer networking, and similar telecommunications and information technologies, on scholarly, scientific, and organizational activities. For example, increases in the use of computer networking may: ¥ Improve users' access to network services, databases, and other users ¥ Increase the frequency with which users interact with each other in informal ways, such as through electronic mail, bulletin boards, and computer conferences ¥ Increase the frequency with which some users interact with each other in formal ways, such as through teleconferencing and electronic publishing ¥ Change the ways in which many forms of work are done, making them more efficient and productive ¥ Change users' patterns of communication, extending their formal and informal communication networks across disciplinary and organizational boundaries. Many optimistic accounts of the impacts of CMC seem to have reached a consensus that scholarly, scientific, and organizational environments will be positively altered as computer networking becomes a routine element of many users' daily work routines. According to Koch, however, there is a serious problem with many of these accounts, because "there is not an established literature devoted to the impacts of electronic networks " on organizational activities [61]. The literature which discusses the positive impacts that CMC will have on work, research, and communication is incomplete because it offers little empirical evidence that describes specifically how these activities have been improved as a result of such technologies. It also does not adequately explain how institutional and individual characteristics would influence the successful use of electronic networks. A significant problem that has hampered researchers, according to Wulf, is that they "..cannot quantitatively measure...[the]...effects..." of networking [62]. Kiesler, Siegel, and McGuire commented that "the functions and impact of computer- mediated communication are still poorly understood" [63]. The Panel on Information Technology and the Conduct of Research found that there is almost no systematic information on the users and uses of information technology. For example, the Panel cannot estimate how many or what proportion of scientists use computers in different fields, how access to networks and computer facilities is distributed across disciplines, or to what extent useful applications are disseminated throughout the research community [64]. The solution to these problems may be to undertake a program of qualitative research, supplemented by quantitative analysis where appropriate, which seeks to uncover the ways in which computer networks are used in research, work activities, and communication. The research will assess the impacts that CMC has had and might have on individuals, their social interactions, and on organizational structures and processes. This research program would focus on the users of computer networking and would attempt to understand their perspective on the role of networking in their daily activities. Although this research direction seems intuitively appropriate, there have been very few studies that have examined the impacts of networking from a user perspective. All too often, research has been conducted on the assumption that users come to networks with a sophisticated knowledge base and set of skills that will allow them to effectively manipulate networks to accomplish their goals. As a result, the experiences of the "typical user" are left out or discounted. Commenting on the spate of claims that computer networks have improved efficiency and productivity in scientific research, Gould states that there is little empirical evidence on the nature and extent of a network's ability to enhance the research process and increase research productivity. Many claims about existing and proposed research networks are characterized by enthusiasm and idealism either not documented by empirical studies, or documented with narrow anecdotes about very successful, high profile users who are in unusually privileged positions within the research community [65]. Koch states that there is very little evidence on how scientists are actually using electronic networks, yet such information is crucial if operational and policy decisions about scientific networks are to be made [66]. As a corrective, a new direction in research is needed which can gather, analyze and present information about the ways in which computer networking is becoming integrated into academic, corporate, library and governmental settings. 5.4 A New Direction for Research There does not appear to be much research which uses an empirical approach, informed by a social science perspective, to examine the use of CMC as a form of social interaction. When people become engaged in almost any form of social interaction, it becomes, over time, routinized and patterned and "can best be analyzed as...[a]...recurrent social practice" [67]. There develops a set of rules or guidelines which creates boundaries within which the social interaction unfolds. More accurately, the set of rules which both enables and constrains a given form of social interaction emerges from the ongoing interaction itself as a product of people's communicative behaviors. These guidelines influence but do not determine social interaction. Individual success or failure in any given social interaction will then depend, to an extent, on the degree to which there is an understanding of these rules and guidelines. The use of CMC systems will continue to spread throughout academic, business, and governmental environments [68], and the networks over which messages can be transmitted will become increasingly global in coverage. As this happens, the rules associated with CMC will become more firmly established as social practices. CMC will also begin to affect the interpersonal, organizational, and cultural settings within which it is used. The initial research problem, then, is to study the process by which these rules and guidelines are emerging. There is a small body of research which has begun to give this process serious consideration. Within this literature there is an evolving research perspective which focuses on the relationships between the users of CMC and the networking technologies that allow them to communicate. It makes use of a variety of methods and analytic approaches to examine the ways in which social interactions in different settings are affected by the use of CMC. Rather than concentrate on questions of system design and requirements, it seeks to investigate people's uses of systems, taking into account their perspectives on the systems they use. This perspective is important because it provides a methodological context for the study of the impact of CMC on work, research, communicative behaviors, and social interaction. Williams, Rice, and Rogers authored a useful volume outlining some aspects of research into what they called the "new media", of which computer networking is an example. They claimed that scholars...need to consider the use of multiple methods, including more qualitative and triangulation methods of data- gathering and analysis, and the interpretive approaches to research. To date, however, most research on the new media has only used quantitative research methods and has been cast in a positivistic approach [69]. They also argued that there is a need for new theories to explain the "new media" which, they state, are different from traditional media. Computer networking has several characteristics which set it apart from traditional forms of communication, such as the telephone, and people are using networks in novel and unique ways. Traditional theories will prove to be inadequate. McClure, Bishop, and Doty [70], in a recent study of the impact computer networks on scientific research and communication, took the approach suggested by Williams, Rice, and Rogers and successfully put it into practice to assess the ways in which scientists actually use networking. They used triangulation and qualitative methods, but also pushed the perspective further. Their research explicitly set out to capture the perspectives of a wide variety of computer network users by using a wide variety of methods to collect and analyze their data. Among their findings was the insight that many policymakers and system designers were operating with a set of assumptions about network users that were not supported by their evidence. Their work demonstrates the usefulness of a user perspective and that there is much to be learned about the impacts of computer networking on users and the settings in which networks exist. 5.4.1 Examples of Research on CMC [71] There is literature which focuses on the users of CMC that uses a variety of methods to uncover the ways in which social interactions are affected by information technology. This literature, by and large, has not focused on scientists or information professionals as representative user populations. It has, however, arrived at some provocative findings which indicate that similar research questions, methods, and research strategies could be usefully applied to the investigation of computer networking activities. The main media of CMC that most of this research examines are electronic bulletin boards, electronic messaging, and computer conferencing. 5.4.1.1 Field Research on the Effects of CMC on User Behavior Rafaeli used a survey of users and an analysis of transaction logs to investigate the use of a computer bulletin board. He concluded that the use of the bulletin board, although extensive, did not always validate the exaggerated claims of network advocates [72]. Love and Rice [73] and Rice and Love [74] used a multiple methods approach, including network analysis, the content analysis of message transcripts, and statistical inference to study social- psychological elements in electronic messaging and bulletin board use. They found that approximately 30% of the content that was sent over a computer network was "socio-emotional" [75]. This finding challenged the widely held assumption that the narrower channels which are available through CMC effectively restrict messages to "..less friendly...[and]... emotional, and more task-oriented" content [76]. Lynch [77]examined the factors which were critical to the acceptance and use of an electronic mail system at the University of California's Division of Library Automation. He then presented a "taxonomy of...implementations" which was intended to provide an analytic scheme which could be used to indicate the ways in which organizations establish and integrate CMC systems [78] Some of the key factors in the successful adoption of the system were ease of use, accessibility of the terminals, reliability and a "good human interface" [79]. The system worked, in part, because the users trained themselves; no training or support was provided by the organization. He found that as users became more proficient in operating the system, their patterns of use changed, as did their perceptions of the system. On the other hand, he also found that the "direct benefits...[were]... difficult to quantify" [80]. A similar type of study by Grieve and McCabe [81] looked at the implementation of an electronic mail system in the Department of Communications at Ohio State University. Using an analysis of quantitative frequency patterns of message exchanges, the data for which were provided by a system transaction file, and user interviews, they sought to study the characteristics that had accompanied the successful implementation and use of the system. They concluded that the success of the system depended on the ability of the users to quickly observe advantages in their daily routines that could be clearly be linked to the use of the system, and the relative ease with which the system could be used. Sproull and Kiesler [82] used questionnaire data, self reports, interviews, and message transcripts to study electronic mail use in an organization. They developed operational measurements including message length in number of lines, the number of words used in the opening and closing of messages, the number of words indicating positive and negative affect and others to conclude that electronic mail speeds up the process of information exchange and leads to the exchange of new information which would not have been sent over other, more traditional media [83]. They also found that CMC decreased the number and types of social cues that are available to participants as they communicate, producing a deregulating effect on communicative behavior. Safyeni and MacGregor [84] used an interview method to compare manager's and employees' perceptions of two organizational electronic mail systems, one used by executives and upper management and one used by the employees. They concluded that the upper managers' system was perceived by both groups to be better and more efficient, although both systems were perceived to be contributing to organizational efficiency and productivity. In their study of an organization, Sproull and Kiesler [85] found that there was little difference in the messages sent within the organization from superior to subordinate and from subordinate to superior, a phenomenon Rice called "access equalization" [86]. In a review of research on CMC conducted during the 1970's, Rice [87] found support for the findings that the introduction and use of CMC in organizations does change the structure of existing communications networks, which became less rigid and had more lateral channels. He also argued that a consequence of the successful implementation of CMC systems was the decentralization of organizational structure. Research studies by Rice [88, 89] have also focused on the effects of computer networking on organizational structures and activities. This research used multiple methods including message transcripts, network analyses, analyses of messaging tasks, questionnaires, interviews and statistics which describe the amount of time people spent sending messages and participating in conferencing. Rice [90] examined the use of electronic mail in a small governmental agency and found that it complemented rather than replaced print media, the use of which increases with the use of CMC. He concluded that the use of electronic mail "may generate more exchanges of facts, more short reports, more notes to process" and may lead users to spend more time searching for information [91]. In a study of CMC and organizational innovation, Rice argued that the introduction of CMC systems changed an organization's structures and processes leading to "persistent, although malleable and non- revolutionary, differences" in the organization [92] CMC led to boundary spanning communication patterns, new forms of information acquisition, remote agenda setting, and desktop publishing. He argued that CMC supported such activities as information and opinion exchange, querying and responding, generating ideas, solving problems, resolving disagreements, staying in touch, and getting to know others in the organization. CMC did not facilitate persuasion, group decisionmaking, or the exchange of confidential information. A study conducted by Rice and Case in an academic setting argued that people who used electronic message systems did not necessarily attribute greater benefits to using CMC as they gained more experience with their systems [93]. They found that, in this setting, CMC reduced users' reliance on telephone and paper based messaging, although almost half of the users reported that they noticed no change in the ways in which they used traditional communication channels. They conclude that electronic messaging helps develop and sustain invisible colleges and can allow users to surmount resource allocation problems which are caused by status differences within and among organizations [94]. 5.4.1.2 Laboratory Research on the Effects of CMC on User Behavior Social psychologists have conducted laboratory studies of the effects of CMC on individual behavior. Matheson and Zanna [95] compared face to face interactions with CMC to investigate the ways in which the latter changed the nature of interpersonal communication. They found that subjects in the CMC trials experienced different levels of "private" and "public self awareness" than did those in the face to face trials [96]. Subjects exhibited a reduced awareness of the social context of the communicative interaction, social and cultural standards and personal norms, and more uninhibited behavior. Smilowitz, Compton, and Flint [97] replicated the famous Asch conformity studies to see if the presence of others would affect individual judgments when the individual communicated through a simulated computer conference. This research was designed to test the effects of the "egalitarian nature" of computer mediated communication [98]. They found that people were less likely to conform to the judgment of a group when they were making their decisions from a remote location through a terminal than they were when they were in the physical presence of a group whose unanimous judgment was in direct opposition to theirs. Subjects in the CMC trials were more willing to critically evaluate information presented to them than to simply accept it [99]. Kiesler, Seigel and McGuire [100] used a social psychological approach to compare face- to-face interaction, simultaneous electronic communication from remote locations and electronic mail in a laboratory setting. They found that there were distinct differences in styles of participation, decisionmaking, and group interaction depending upon the experimental condition. The conditions which used electronic channels to mediate communication were characterized as lacking in the verbal and non-verbal cues which accompany face-to-face and voice interactions. One interesting implication that they drew from their findings is the possibility that the relative cuelessness of CMC may create the conditions for "deindividuation", which may, in turn, lead to increased uninhibited behavior among users [101]. They describe CMC as "undeveloped culturally" with no strong norms to regulate its use. Studies have also been conducted which seek to uncover the impacts of computer mediated communication on arousal, feelings, and expressive behavior [102]. 5.4.1.3 Research on CMC and Scientific Communication Several studies have examined the scientific research community's use of computer mediated communication. Hiltz and Turoff conducted an 18 month study of a sample drawn from five different research communities and gathered detailed empirical evidence about changes in user behavior and preferences related to the features and capabilities of computer based communication systems as a function of experience [103]. They found that these research communities were heavy users of networks for messaging and conferencing. Interviews with users revealed that they wanted CMC systems to provide group conferences, notebooks for text composition, self defined commands and simple messaging formats. Their analysis led them to observe that system designers should always base their planning on "..feedback from the experiences of users in current systems" [104]. Shapiro and Anderson wrote on the need for social norms which would regulate the "quality and appropriateness" of electronic communication [105]. Although their report is largely based on anecdotal evidence gathered from their experiences with Usenet in the Rand Corporation, they did not present it as empirical research. Rather, their goal is to "accelerate the process by which social customs and behavior appropriate to electronic mail" and computer conferencing so that these media can be used effectively by the scientific community [106]. The report has value as a document representing two users' attempt to understand the evolution of patterns of communicative behavior over a computer network. Brotz [107], wrote about the "Laurel" system, a messaging network used by Xerox. He claimed that he had "discovered many patterns of message system user behavior that appear to apply to message systems in general" by observing people as they used the "Laurel" system over a period of several months [108]. After discussing five major problems that had occurred during the time that he observed system use, Brotz offers some "biased suggestions about standards" that he believed would improve system effectiveness and reduce incidents of abuse [109]. McClure, Bishop, and Doty [110] conducted an extensive study of the impacts of computer networks on the research process and scientific communication. They also identified and analyzed trends and policy issues in networking and offered recommendations for policymakers concerned with the development of the National Research and Education Network (NREN). This study used multiple methods to collect and analyze data and concentrated on the perspectives and experiences of network users and managers in academia, the private sector and Federal laboratories. They found that the basic research process was not appreciably changed by the use of networks, although findings suggest that network use may provide researchers with a competitive advantage over researchers that are unable or unwilling to use high-speed research networks...The findings also indicate that it is likely that the modi operandi of a large number of scientists have not been affected by networks. Nor is it likely that in the immediate future electronic networks will affect their research process. [111] Networks, however. have improved the efficiency of the scientific communication process by extending scientists' personal communication networks, expanding the pool of human and bibliographic resources, and enabling remote collaboration. They also found that there was a pressing need for education and training programs for network users, many of whom did not have the skills they needed to fully exploit networking in their work. (Table 4 here: Research on CMC: Findings) 5.4.2 CMC As Social Interaction How, then, does the literature treat computer networking as a form of social interaction? In comparison to face to face communication, CMC takes place over a much narrower channel. It removes or suppresses the verbal and visual cues, "especially facial expression and eye contact" that allow participants in face to face communication to create meanings and monitor the the interaction as it unfolds [112]. With the removal of obvious status cues, computer networking has the potential for creating greater equality among users, unless there are ways of including status cues in the text of the message. In CMC, the immediate feedback of a physical encounter is blocked and can only occur as textual feedback during the exchange of messages, thereby reducing its effect only to the creation of subsequent messages. As a consequence of the quality of narrowing, users' attention becomes focused on the message or text as the source of the cues and other information which will help the recipient develop some conception of the sender and her/his intentions, motivations, status, and interests. The ways in which the presence and absence of various types of cues affect communication is a topic which has received research attention, although not with computer networks as the focus of attention. Studies of the effects of "cuelessness" have traditionally focused on the differences between face-to-face communication, telephones, com- munication over microphones and speakers from remote locations, and even communication among blind people [113, 114]. This work has produced several interesting findings that may have relevance for the study of computer networking as social interaction. One is that the presence or absence of visual contact affects "three principal aspects of social interaction: content, style and outcome", but in paradoxical ways [115]. Content and style are dependent on the social cues that are available in the particular form of communication, and the fewer cues that are present, the more task-oriented and depersonalized the interaction will become. This occurs until a point at which the lack of cues may have an opposite effect, as on a telephone hotline, where the anonymity may produce intimacy instead of psychological distance. Another finding concerns the concept of "social presence" [116], which is introduced as a "subjective" quality of a form of communication which affects user interaction; "media which allow psychologically close encounters are...high in social presence" [117]. Users understand the relative social presence of different forms of communication and will choose appropriate forms for their various interactions. CMC also removes constraints of time and space [118]. Rules that govern face-to-face interaction, such as turn taking, the obligation to immediately answer a question, the avoidance of interruptions and the display of deference to others of higher status do not apply to asynchronous communication that originates in remote locations. Senders and receivers of electronic mail conduct their interactions in contexts of relative anonymity and autonomy. They can choose where and when to send their messages, whether or not to respond, and, because they are not bound by "norms about sticking to the subject", they can decide what the topics and formats of their messages will be [119] . There is also a quality of impersonality to e- mail communication, due in part to the lack of immediacy which accompanies face-to-face and other synchronous forms of communication. Its effect are not clear and researchers seem to be struggling with the significance of impersonality for CMC interaction, although there is agreement that it is present [120, 121, 122, 123]. One interesting implication of impersonality in computer networking is that it may increase the chances that interactions will be more uninhibited and impulsive and less socially desirable [124, 125, 126]. Users will be less aware of the social context into which their messages will be transmitted, and less committed to existing guidelines of propriety [127] They will be more likely to form and hold to individual judgments in the face of mounting social pressures to conform to a majority [128]. As a consequence, the messages that are sent can be expected to contain a higher than expected degree of socioemotional content, without regard for the type of computer network under consideration [129]. This can range as high as 30% of the message content [130]. Users may engage in the expression of negative affect, or "flaming" [131], or they may take advantage of the constraints of computer networking to establish "some very warm and personal relationships" [132]. This creates a situation where message styles are "unpredictable" and the uninhibited interaction of users may lead to "deindividuation" [133]. There is also the impact of CMC on the larger society in which it is used. The research on e-mail systems has posited an interesting range of relationships between CMC and the social world, three of which will be briefly mentioned. First, in what seems to be the broadest and most abstract of these relationships, there is a link between various forms of CMC and the maintenance of the social order. The use of keyboard based CMC may be reinforcing the rules and practices of bureaucratic rationality [134]. The use of CMC systems may be a form of social interaction around which a new and different set of social norms are emerging [135, 136, 137, 138]. It may even be a critical element in the emergence of a CMC based subculture [139]. Second, the spread of CMC systems through a variety of environments is thought to be changing the fundamental nature of interpersonal communication by adding a new dimension or channel with its own unique characteristics [140, 141]. The use of CMC may be changing the ways in which groups make decisions and reach a consensus [142, 143] The spread of CMC through the scientific research community may produce increases in productivity because of the new ways in which scientists can communicate with each other [144]. It may also be producing a wide range of cognitive, affective, and behavioral impacts on individual users, groups, and institutions [145]. Third, CMC systems are also thought to be significant factors in creating organizational innovation and change [146, 147, 148] The impacts of CMC systems will be felt in the domains of individual work patterns, work groups, and organizational structure [149, 150]. CMC systems may have the potential to alter the ways in which organizational members exchange information [151] and may alter organizational structure, both by decentralizing power and authority and by creating new communication networks with more cross-boundary lateral channels [152]. (Table 5 here: CMC as Social Interaction: Findings) 6.0 Conclusion It is clear that computer networking is an emerging form of communication that is becoming increasingly pervasive in many organizational settings. Although it has received research attention over the last two decades, much of the work has focused on characteristics of networking other than interactive ones. Descriptions of existing systems are useful, but more to the consumer of networking services than to the researcher. Experiments which simulate conditions of networking interactions are illuminating, but they have difficulties overcoming the artificiality of the laboratory or experimental setting. Field studies are closing in on the problem of understanding computer networking as a social interaction, but they typically have not posed the types of research questions that would have allowed them to capture the interactive features of CMC and the emerging sets of rules and guidelines which are shaping the subculture of network users. There is an apparent gap in the literature [153] which could be filled with a series of field studies which, informed by sociological, communication, and social- psychological perspectives, set out to investigate the ways in which users interact over computer networks. This work is important because CMC is a new form of communicative social interaction that is not yet well understood, either by researchers or users, although it is becoming increasingly important in many different communication settings. Furthermore, "people... [generally]...receive no formal instruction in an etiquette of electronic communication, and what informal socialization exists tends to be received haphazardly" [154]. Such a situation calls out for rigorous investigation. Some basic research questions are: ¥ How do people use e-mail in different organizational settings to regularly and routinely interact with each other? ¥ What patterns of interaction are emerging? How have they developed over time? How and why are they changing? ¥ How can these patterns of interaction be analyzed as "social practices"? There are also related questions which must be asked: ¥ How do people learn to use e-mail? Is there a formal training process? Is it a process of trial and error? ¥ What are the rules and guidelines which are emerging as people regularly use e-mail? To what extent are these rules and guidelines common to all e-mail systems and to what extent are they unique to specific systems? ¥ What is the significance of the concepts of "cuelessness" and "social presence" for under- standing e-mail as a social interaction? We know something of the future of CMC: standards will continue to emerge for the technical side of networking; voice mail, graphics, and color will be integrated into email and other networking activities; the worldwide network will branch into all sectors of society. Although we cannot predict all of the advances to come in the future, we predict that CMC will continue to play an increasingly important role in our everyday lives. CMC is different from forms of communication with which we are familiar. It shares characteristics of the telephone, the letter, the inter-office memo, and other phenomena for which rules of conduct are well-known, but it is also unlike any other form of human communication. The social norms and rules associated with CMC are not yet well developed. As more members of society participate in this exciting new way of communicating, the norms and rules will continue to emerge and solidify. As CMC becomes commonplace in society, new applications will emerge which have so far been in the domain of fiction. There are exciting times ahead. NOTES 1. Bitnet is a trademark of CREN (the Corporation for Research and Educational Networking). Unix is a trademark of AT&T. Ethernet is a trademark of Xerox. Vax is a trademark of Digital Equipment Corporation. Other trademarks may apply for items mentioned in this work. 2. 9600 baud is 9600 bits per second. Since characters are usually 8 bits long (one byte), this amounts to 1200 characters per second. 1200 characters fill just less than one full screen of data. 3. LAN's are able to operate at higher speeds because they usually do not use analog signals for transmission. Since computers only understand digital signals, and telephone are used for analog transmissions, a device called a modem (for modulator-demodulator) is employed to translate the signals from digital to analog and back again. 9600 baud is close to the maximum cost efficient modem speed. LAN's use such technologies as Tokenrings and Ethernet to connect the nodes. Digital channels allow signals to be transmitted almost as fast as the host computers can send and receive them. Note that technologies for sending analog transmissions without phone lines are able to operate at significantly higher speeds. Microwaves can be used to transmit signals at rates of one million bits per second or more. 4. For a more complete description of worldwide computer networks, see John Quarterman [155]. BITNET and Internet, the topics of this paper, are largely non- commercial networks. There are also special- purpose, for-profit, and private networks that are parts of the matrix discussed by Quarterman. 5. Microcomputers and other small computers are able to linked to Internet because of the network's hierarchical structure. It is not necessary for each node on the network to have an explicit map to each other node. Instead, information is passed from node 1, the sending node, to node 2, the node lowest in the hierarchy representing the target node's address which the ending host recognizes. Then, node 2 continues to route the message down the hierarchy to the target node. 6. The scholarly journal, Social Systems, is devoted to the study of human communication networks, but often with a focus on computer mediated communication. Williams, Rogers, and Rice [156] have written a volume that discusses methods for analyzing networks. 7. See BIT2PLAN PROPOSAL, an online file available from Listserv@ BITNIC. 8. RFC822 ("Request for Comments # 822) is on online file available from Document @ Bitnic. 9. The notion of rules for human communication is embedded in a definite theoretical framework, although a casual notion of the role of rules in communication is not inappropriate for the current discussion. An introductory discussion of the literature concerning rules in human communication can be found in Littlejohn (1989). 10. Computer networkers understand a number of special key sequences used to indicate affect. The sequence " ;-) " is used to indicate when the sender is not to be taken seriously (if you don't get it, turn this page on its side). 11. On Bitnet, discussion of these topics is on the Policy-L mailing list (policy- l@bitnic). See the archives from December, 1989 for a discussion of one recent incident of a user sending a message to over 100 mailing lists. In the end of the discussion, as in previous incidents, the Bitnet policy-makers and enforcers realized that they had very limited power to take disciplinary action against either the user, or the host. (This was a particularly interesting case, because the offending user was acting in an official capacity as spokesperson for the host institution.) 12. For more information on Relay, send a message with the text "/help" to Relay @ Bitnic (do not type the quotation marks). 13. Bitnet usage guidelines explicitly forbid sending of line messages to users at random. However, this rule is difficult to enforce, unless the recipient of the messages reports the infraction. 14. Many Internet hosts have anonymous Ftp access. Not all have such a wide variety of available information, though. One of the best sources for computer programs for a variety of systems is currently supported by the Federal Government: simtel20-wsmr.army.mil. 15. Public domain is contrasted with shareware and with copyrighted software. Shareware is software which may be distributed at will, but users of the software have a moral obligation (and possibly a legal obligation) to pay for the software if they decide to use it. Copyrighted software is produced for a profit, and may not be legally copied or redistributed. It is the corporations which produce copyrighted software which are most familiar to the public, because they are able to devote their profits to advertising. However, both shareware and public domain software are frequently of comparable quality to copyrighted software. All three may be produced by professional programmers and software development teams, although public domain software and some shareware is produced by independent programmers. Public domain and shareware software are able to cater to the specific needs of smaller audiences, because they do not have the necessity for high sales that profit-oriented corporations usually do. 16. For a discussion of the impacts of the Robert Morris affair on networking, see Cerf ([157], Chapman [158], Eisenberg, et al. [159], Farber [160], Rochlis, et al. [161] and Saltzer [162]. 17. Actually, the original term was "Because it's There," coined by Ira Fuchs, recognized as a founder of Bitnet. As a result of a query by the secondary author to "Policy-L @ Bitnic" mailing list, it was found that the original title was changed to "Because it's Time" because IBM and other sources were to be approached for funding, and Fuchs and others wanted the name to sound more serious. Some BITNIC documents are still signed "Because we're here." 18. Such a map would almost certainly be fractal in nature. 19. Dean Harper, "Computer Technology and Social Relations," Computers and the Social Sciences, 1, (1985) 127. 20. Elaine B. Kerr and Starr Roxanne Hiltz, Computer Mediated Communications Systems (New York: Academic Press, 1982), 3. 21. Charles W. Steinfeld, "Computer Mediated Communication Systems," Annual Review of Information Science and Technology 21, (New Jersey: Knowledge Industry Publications, Inc., 1986). 22. Sara Kiesler, Jane Seigel, and Timothy W. McGuire, "Social Psychological Aspects of Computer Mediated Communications," American Psychologist, 39(10), (October, 1985), 1123. 23. Ronald Rice, "The Impacts of Computer Mediated Organizational and Interpersonal Communication," Annual Review of Information Science and Technology 15, (New Jersey: Knowledge Industry Publications, Inc., 1985). 24. Steinfeld, "Computer Mediated Communication Systems". 25. Susan E. Koch, "How Electronic Networks are Changing Science: Implications for a National Research Network", Final Report submitted to the Office of Technology Assessment, U. S. Congress, (1989), 4. 26. Shelly Grieve and Barbara G. McCabe, "E- mail Use in a University Department of Communication," ASIS '86: Proceedings of the 49th Annual Meeting of the American Society for Information Science, (New Jersey: Learned Information Pub., 1986). 27. R. G. Rittenhouse, "Computer Based Messaging in Research Organizations", Communicating Information: Proceedings of the 49th Annual Meeting of the American Society for Information Science, (New Jersey: Learned Information Pub., 1986). 28. Laurel Nan Hellerstein, "The Social Use of Electronic Communication at a Major University," Computers and the Social Sciences, 1, (1985), 191. 29. H. V. Dreher, "Electronic Mail: An Exemplar of Computer Use in Education," Educational Technology, 24(8), (August 1985), 37. 30. Ronald Rice and D. Case, "Computer-Based Messaging in the University: A Description of Use and Utility," Journal of Communication, 33(1), (Winter, 1983) 131. 31. Starr Roxanne Hiltz, Online Communities: A Case Study of the Office of the Future (Norwood, N.J.: Ablex Pub. Co, 1984). 32. Starr Roxanne Hiltz and Murray Turoff, "The Evolution of User Behavior in a Computerized Conference System," Communications of the ACM, 24(11), (November, 1981), 738. 33. Starr Roxanne Hiltz and Murray Turoff, The Network Nation: Human Connection via Computer (Reading, Massachusetts: Addison Wesley,1978). 34. Clifford A. Lynch, "Practical Electronic Mail Though a Centralized Computing Facility," Communicating Information: Proceedings of the 43rd ASIS Annual Meeting. Vol 17. (New Jersey: Knowledge Industry Pub. Inc., 1980). 35. Robert Johansen, Robert DeGrasse Jr., and Thaddeus Wilson, Group Communication through Computers: Vol 5: Effects on Working Patterns, The Final Report on Research, (supported by National Science Foundation. Report R-1, California: Institute for the Future, February, 1978). 36. Ronald Rice and D. Shook, "Access to, Usage of and Outcomes from Electronic Messaging," ACM Transactions on Office Information Systems, 6(3), (1989) 255. 37. Urs E. Gattiker, Barbara A. Gutek, and Dale E. Berger, "Office Technology and Employee Attitudes," Social Science Computer Review, 6(3) , (Fall, 1988) 327-40. 38. F. Safyeni, E. Lee, and J. MacGregor, "An Empirical Investigation of Two Electronic Mail Systems," Behavior and Information Technology, 7(4), (October, 1988) 361-72. 39. Ronald Rice, "Computer Mediated Communication and Organizational Innovation," Journal of Communication, 37(4), (Autumn, 1987) 65-94. 40. Lee Sproull and Sara Kiesler, "Reducing Social Context Cues: Electronic Mail in Organizational Communication," Management Science, 32(11), (November, 1986) 1492- 1512. 41. Ronald Rice and J. Torobin, "Expectations About the Impacts of Electronic Messaging," ASIS '86: Proceedings of the 49th Annual Meeting of the American Society for Information Science. (New Jersey: Learned Information Pub., 1986). 42. David L. Althiede, "Keyboarding as a Social Form," Computers and the Social Sciences, 1, (1985) 97-106. 43. Norman Z. Shapiro, and Robert H. Anderson, Toward an Ethics for Electronic Mail: Report prepared for the National Science Foundation, (Report # R-3283-NSF/RC, California: Rand Corporation July, 1985). 44. Ronald Rice and James Blair, "New Organizational Media and Productivity," The New Media: Communication, Research and Technology, Ronald P. Rice (ed.), (California: Sage Press 1984). 45. Douglas K. Brotz, "Message System Mores: Etiquette in Laurel," ACM Transactions on Office Systems, 1(2), (April, 1983) 179-92. 46. Margrethe Olson, and Henry C. Lucas Jr., "The Impact of Office Automation on the Organization: Some Implications for Research and Practice," Communications of the ACM, 25(11), (1982) 838-47. 47. Ronald Rice, "Outcomes Associated with New Media Use are Contingent on Task Characteristics," Managing Information Technology: ASIS '89. Proceedings of the 52 Annual Meeting of the American Society for Information Science (New Jersey: Learned Information Pub., 1989). 48. Rice, "Computer Mediated Communication". 49. Gail Love, and Ronald Rice, "Electronic Emotion: A Content Analysis and Role Analysis of a Computer-Mediated Communication Network," ASIS '85: Proceedings of the 48th Annual Meeting of the American Society for Information Science, Vol. 22, (New Jersey: Learned Information Pub., 1985). 50. Sheizaf Rafaeli, "The Electronic Bulletin Board: A Computer Driven Mass Medium," Computers and the Social Sciences, 2, (1986) 123-36. 51. Ronald Rice and Gail Love, "Electronic Emotion: Socioemotional Content in a Computer Mediated Communication Network," Communication Research, 14(1), (1987) 85- 108. 52. Becki Whitaker, "Electronic Mail in the Library: A Perspective," Library Trends, 37(3), (Winter, 1989) 357-365. 53. R. S. Wassa, "Convivial Messaging Systems: Startling Facts and Figures About Electronic Mail for French Households," Information Society, 5(4), (1987) 265-9. 54. Hiltz and Turoff, "The Evolution of User Behavior" 740. 55. Sara Kiesler, David Zubron, Anne Marie Moses, and Valerie Geller, "Affect in Computer Mediated Communication: An Experiment in Synchronous Terminal-to-Terminal Discussion," Human Computer Interaction, 1(1), (1985) 77-104. 56. Love and Rice, "Electronic Emotion". 57. Kimberly Matheson and Mark P. Zanna, "The Impact of Computer Mediated Communication on Self Awareness," Computers in Human Behavior, 4(3), (1988) 221-33. 58. Michael Smilowitz, D. Chad Compton, and Lyle Flint, "The Effects of CMC on an Individual's Judgment: A Study Based on the Methods of Asch's Social Influence Experiment," Computers in Human Behavior, 4(4), (1988) 311-22. 59. Adrian Furnam, "Assertiveness Through Different Media," Journal of Language and Social Psychology, 3(1), (1986) 1-11. 60. Steinfeld, "Computer Mediated Communications Systems" 197. 61. Koch, "How Electronic Networks are Changing Science", 2. 62. William Wulf, In Computer Networks and High Performance Computing ... Hearing. (Subcommittee on Science, Technology, and Space. Washington, D. C., GPO, August, 11, 1988) 94. 63. Kiesler et al., "Social Psychological Aspects" 1493. 64. Panel on Information Technology and the Conduct of Research, Information Technology and the Conduct of Research: The User's View, (Washington, D. C.: National Academy Press, 1989) 9. 65. Stephen Gould, High Performance Computing: An Overview, (Congressional Research Service, Washington D.C.: Library of Congress, January 11, 1990) 16, 17. 66. Koch, "How Electronic Networks are Changing Science", 2. 67. Anthony Giddens, Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis, (Berkeley: University of California Press, 1979) 66. 68. There is home or consumer use of e-mail services in France and other European countries and, to a much lesser extent, in this country. There is not, however, much literature available on these services which goes beyond the descriptive level, so this topic will not be discussed in this paper. 69. Frederick Williams, Ronald E. Rice, and Everett M. Rogers, Research Methods and the New Media, (New York: The Free Press, 1988) 50. 70. Charles R. McClure, Ann Bishop, and Philip Doty, The Impacts of High Speed Electronic Networks on Scientific Communication and Research: Final Report submitted to the Office of Technology Assessment, (Washington D. C.: Office of Technology Assessment, October, 1989,). 71. Portions of this literature review are adapted from Rosenbaum[163] 72. Rafaeli, "The Electronic Bulletin Board. 73. Love and Rice, "Electronic Emotion". 74. Rice and Love, "Electronic Emotion". 75. Ibid., 108. 76. Love and Rice, "Electronic Emotion" 266. 77. Clifford A. Lynch, "Practical Electronic Mail Though a Centralized Computing Facility," Communicating Information: Proceedings of the 43rd ASIS Annual Meeting. Vol 17, (New Jersey: Knowledge Industry Pub. Inc. 1980). 78. Ibid., 35. 79. Ibid., 35. 80. Ibid., 36. 81. Grieve and McCabe, "E-mail Use in a University Department". 82. Sproull and Kiesler, "Reducing Social Context Cues". 83. Ibid., 1500. 84. Safyeni and MacGregor, "An Empirical Investigation". 85. Sproull and Kiesler, "Reducing Social Context Cues". 86. Ronald Rice, The New Media: Communication, Research, and Technology, (California: Sage 1984) 165. 87. Rice, "The Impacts of Computer Mediated". 88. Rice, "Outcomes Associated with New Media". 89. Rice, "Computer Mediated Communication". 90. Rice, "Outcomes Associated with New Media". 91. Ibid., 182. 92. Rice, "Computer Mediated Communication" 67. 93. Rice and Case, "Computer-Based Messaging" 133. 94. Ibid., 150. 95. Matheson and Zanna, "The Impact of Computer Mediated". 96. Ibid., 229. 97. Smilowitz et al., "The Effects of CMC". 98. Ibid., 313. 99. Ibid., 320. 100. Kiesler et al., "Social Psychological Aspects". 101. Ibid., 1132. 102. Kiesler et al., "Social Psychological Aspects". 103. Hiltz and Turoff, "The Evolution of User Behavior" 740. 104 .Ibid., 739. 105. Shapiro and Anderson, Toward an Ethics iii. 106. Ibid., iii. 107. Brotz, "Message System Mores". 108. Ibid., 179. 109. Ibid., 181. 110. McClure, The Impacts of High Speed Electronic Networks. 111. McClure, The Impacts of High Speed Electronic Networks 43. 112. Hiltz and Turoff, The Network Nation 81. 113. D. R. Rutter, Communicating by Telephone: International Series in Experimental Social Psychology: Vol. 15, (Oxford: Pergamon Press 1987). 114. N. J. Kemp and D. R. Rutter, "Cuelessness and the Content and Style of Interaction," British Journal of Social Psychology, 21(1), (1982) 43-9. 115. Rutter, Communicating by Telephone 69. 116. Rice [164] uses the concept of "social presence" in a discussion of the differences in the types of information transmitted by different media; it is "the degree to which a medium is perceived to convey the physical presence of the participants". He takes this to mean that the fewer cues there are, the less social presence a form of communication has. The concept, as expressed by Rutter [165], does not seem to be limited to the presence or absence of cues, since it depends on users' interpretations of a form of communication as a whole in terms of its ability to enhance or inhibit feelings of psychological intimacy or closeness. The latter explanation would seem to be more applicable to computer networking which, as will be discussed below, does allow intimacy and closeness to be expressed. 117. Rutter, Communicating by Telephone 132. 118. Love and Rice, "Electronic Emotion". 119. Hiltz and Turoff, The Network Nation 29. 120. Smilowitz et al., "The Effects of CMC". 121. Sproull and Kiesler, "Reducing Social Context Cues". 122. Kiesler et al., "Social Psychological Aspects". 123. Kerr and Hiltz, Computer Mediated Communications. 124. Sproull and Kiesler, "Reducing Social Context Cues". 125. Brotz, "Message System Mores". 126. Hiltz and Turoff, The Network Nation. 127. Matheson and Zanna, "The Impact of Computer Mediated". 128. Smilowitz et al., "The Effects of CMC". 129. Rice and Love, "Electronic Emotion". 130. Love and Rice, "Electronic Emotion". 131. Shapiro, and Robert H. Anderson, Toward an Ethics. 132. Hiltz and Turoff, The Network Nation 27. 133. Kiesler et al., "Social Psychological Aspects" 1134. 134. Althiede, "Keyboarding". 135. Harper, "Computer Technology". 136. Alex Czajkowski and Sara Kiesler, "Computer Mediated Communication: Or the New Next Best Thing to Being There," National Forum, 64(3), (Summer , 1984,) 31-4. 137. Brotz, "Message System Mores". 138. Hiltz and Turoff, The Network Nation. 139. Hellerstein, "The Social Use of Electronic Communication". 140. Rice and Love, "Electronic Emotion". 141. Rice, "The Impacts of Computer Mediated". 142. Kiesler et al., "Social Psychological Aspects". 143. Kerr and Hiltz, Computer Mediated Communications. 144. Johansen et al., "Group Communications through Computers". 145. Kerr and Hiltz, Computer Mediated Communi- cations. 146. Gattiker et al., "Office Technology". 147. Grieve and McCabe, "E-mail Use in a University Department". 148 Rice, "Computer Mediated Communication". 149. Olson and Lucas, "The Impact of Office Automation". 150. Bruce Christie, Face to File Communication: A Psychological Approach to Information Systems, (Chichester: John Wiley and Sons, 1981). 151. Sproull and Kiesler, "Reducing Social Context Cues". 152. Rice, "The Impacts of Computer Mediated". 153. A cursory scan of recent issues of Dissertation Abstracts reveals that CMC is becoming a rather popular topic for researchers receiving Phd.'s in communication, management science, and social psychology. For example, there is research on CMC which has been conducted in prison [166], in organizations [167, 168, 169] and on a European e-mail network [170]. Perhaps there is a new generation of researchers who will be working to fill this gap in the literature. 154. Kiesler et al., "Social Psychological Aspects" 81. 155. John Quarterman, The Matrix: Computer Net- works and Conferencing Systems Worldwide (Bos- ton: Digital Press, 1990). 156. Williams et al., Research Methods and the New Media. 157. Stephen Littlejohn, Theories of Communication 3rd Edition, (California: Wadsworth Pub. Co. 1989). 158. Vint Cerf, "Ethics and the Internet," Comm- unications of the ACM, 32(6), (June, 1989) 710. 159. Gary Chapman, "CPSR Statement on the Computer Virus," Communications of the ACM, 32(6), (June, 1989) 699. 160. Ted Eisenberg, David,Gries, Juris, Hartmanis, Don Holcomb, M. Stuart Lynn, and Thomas Santoro, "The Cornell Commission: On Morris and the Worm," Communications of the ACM, 32(6), 706-709. 161. David, J. Farber, "NSF Proposes Code of Networking Ethics," Communications of the ACM, 32(6), (June, 1989) 688. 162. Jon A.Rochlis and Mark W.Eichil, "With Microscope and Tweezers: The Worm from MIT's Perspective," Communications of the ACM, 32(6), (June, 1989) 689-698. 163. Jerome H. Saltzer, "Teaching Students about Responsible Use of Computers," Communications of the ACM, 32(6), (June, 1989) 704. 164. Howard Rosenbaum, "Background and Development of a National Research and Education Network," (In press, 1990). 165. Rice, "Outcomes Associated with New Media" 177. 166. Rutter, Communicating by Telephone. 167. Gina C. Daddario, Bytes and Bars: The Social Impact of Computer Mediated Communication in a Minimum Security Prison, (University of Mass- achusetts 1988). 168. Steven Raymond Phillips, Electronic Per- suasion: The Uses of Electronic Mail for Inter- personal Influence in Organizations, (University of Southern California 1989). 169. Anne Patchell Crum, Factors which Influence Electronic Mail Use, University of Georgia 1988). 170. Linda A. Kurth, Message Responses as a Function of Communication Mode: A Comparison of Electronic Mail and Typed Memoranda, (Arizona State University 1987). 171. Eklunda Kersten Elisabet Severinson, Dialogue Processes in Computer Mediated Com- munication : A Study of Letters in the COM System, (Linkoping University, Sweden 1987). 172. BITNIC, BITNET OVERVIEW, (LISTSERV @ BITNIC December, 1989). -------- Table I: Network Activities in Brief Activity/source: Email Network(s): All Type of Activity: Asynchronous communication, to individuals or groups who may or may not be known personally known Sample: Email to any userids on any network Human aspects: For a wide variety of communication types, memos, letters, formal, and informal communication Activity/source: Listserv and other file- and mail-servers Network(s): Listserv on BITNET, other programs available on networks Type of Activity: Asynchronous communication, mass distributed Sample: LISTSERV @ BITNIC distributes the POLICY-L mailing list Human aspects: Similar to email, but broadcast to a list of userids. Also makes files available, in the form of electronic "libraries" Activity/source: Interactive messaging Network(s): BITNET Type of activity: Synchronous communication to a single user Sample: Send a line message to any Bitnet users Human aspects: The recipient must be signed on to his/her account. Difficult to have a conversation. Activity/source: Usenet, BBS Network(s): Usenet on Unix systems, over UUCP and Internet. BBS on a variety of mainframes and microcomputers on local to international scales. Type of activity: Asynchronous communication to a wide audience Sample: Post to news.test on Usenet, with national distribution Human aspects: Similar to email. Better defined norms of conduct. Usenet and BBS software allow for a more conversational atmosphere, often with more discussion on a topic than usual for Bitnet mailing lists. Activity/source: FTP Network(s): Internet Type of activity: Synchronous file transfer Sample: Anonymous FTP to host nic.ddn.mil to retrieve the current List of Mailing Lists. Human aspects: Allows access to tremendous information resources, quick transfer between hosts. Danger of retrieving inaccurate information, programs sabotaged by worms or viruses. Activity/source: Remote login Network(s): Internet, others Type of activity: Synchronous, simulates being logged directly onto the remote host Sample: Logging in to one of the NSF supercomputer facilities. A valid account is required. Human aspects: The next best thing to being there -------- Table II: Bitnet Membership Bitnet connectivity is as follows: Members Nodes Location 500 Members 1750 nodes US (BITNET) 650 Sites 800 nodes Europe (EARN) 110 Sites 180 nodes Canada (NetNorth) 20 Sites 30 nodes Latin America (ALNS, RUNCOL...) 65 Sites 100 nodes Asia (GulfNet, Hong Kong, ...) Bitnet Worldwide: In 35 countries Africa: Ivory Coast Asia: Japan, Korea, Singapore, Taiwan, Hong Kong Europe: Austria, Belgium, Denmark, France, Finland, Great Britain, Greece, Iceland, Ireland, Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, Turkey, West, Yugoslavia Latin America: Argentina, Brazil, Chile, Colombia Middle East: Israel, Saudi Arabia North America: Canada, United States, Mexico This information was taken directly from BITNET OVERVIEW (December, 1989), an online file available from LISTSERV @ BITNIC [172]. -------- Table III: Points for Effective Computer Mediated Communication These points apply for personal email, mail to a group of people, mail sent to a distribution list, or items posted to a Usenet or other bulletin board-type systems. o Read carefully what you receive, to make sure that you are not misunderstanding the message. o Read carefully what you send, to make sure that your message will not be misunderstood. o If you are using humor or sarcasm, make sure it is clearly labeled as such. o Remember that the absence of cues associated with face to face communication provide for an environment where it is easy to misunderstand what is being said. o Know your audience. Make sure that the person or list of people you are sending your message to are the appropriate one(s) to com- municate with. o Be tolerant of newcomers. Do not abuse new users of computer networks for their ignorance -- be patient as they first learn to crawl, and then to walk. o Avoid cluttering your messages with excessive emphasis (such as stars, arrows, and the like). It may make the message hard to follow. o If you are responding to a message, either include part of the original message in your message, or make sure you unambiguously refer to the original message's contents. It's very possible that people will read your response to the message before they read the original. A convention is to proceed each line of the original message you include with the > character.) o Always include a descriptive subject line in your message. If you are responding to another message, your subject line should be the same, with "Re: " at the beginning. If you are going off track, change the subject line. For example, if the original subject was: "Email," the followup subject would be "Re: Email." A spin-off subject would be: "My view of the world (was Re: Email)." o Keep messages to only one subject. This allows readers to quickly decide whether they need to read the message in full. Second subjects may be missed. -------- Table IV: Research on CMC: Findings Field Research Some of the findings that have emerged from field research on CMC in different settings are: o Users' actual behaviors when interacting through bulletin boards fall short of the claims of supporters of networking technology. o As much as 30% of the message content in some computer conferences may be described as socio-emotional. o As users' time on electronic networks increases, their patterns of use change as do their perception of the networks. o It is very difficult to quantify the direct benefits that users experience when using electronic networks. o Successful implementation and use of networks may depend on the extent to which users can perceive clear advantages to their use and the relative ease with which the networks can be used in daily work routines. o One significant factor which mitigates against the acceptance and use of networks in organizations is the lack of user training. o CMC suppresses social cues, leading to more uninhibited or deregulated communicative behavior. o Electronic networks may reduce status differences within organizations. o CMC systems affect organizational structures and processes, increasing boundary spanning communication and creating more lateral communication channels. o The use of CMC may lead to organizational decentralization. o In organizations, CMC can support such activities as, information and opinion exchange, querying and responding, generating ideas, solving problems, resolving disagreements, staying in touch, getting to know others in the organization. o In organizations, CMC does facilitate persuasion, group decision- making, or the exchange of confidential information. o In academic settings, electronic networks are used to create and support invisible colleges and to promote resource sharing. Experimental Studies Some findings that have emerged from experimental studies of CMC are: o Subjects exhibit a reduced awareness of the social context of the networking interaction, social and cultural standards and personal norms, and more uninhibited behavior. o CMC use changes users' self-perceptions of public and private awareness. o In group decisionmaking, people are less likely to conform to the judgment of a group when making their decisions from a remote location through a terminal than they are when they are in the physical presence of a group whose unanimous judgment is in direct opposition to theirs. o Subjects in CMC trials are more willing to critically evaluate information presented to them than to simply accept it. o There are distinct differences in styles of participation, decisionmaking, and group interaction depending upon the experimental condition. o The conditions which use electronic channels to mediate com- munication are characterized as lacking in the verbal and non- verbal cues which accompany face-to-face and voice interactions. o The cuelessness of computer mediated communication may create the conditions for "deindividuation", which may, in turn, lead to increased uninhibited behavior among users. o Computer mediated communication is "undeveloped culturally" with no strong norms to regulate its use. o CMC use may lead to increases in physiological and psychological arousal for users. Studies of CMC in Scientific Communication Some findings that have emerged from studies of CMC and scientific communication are: o Some research communities are heavy users of networks for messaging and conferencing. o Users want CMC systems to provide group conferences, notebooks for text composition, self defined commands and simple messaging. o Network designers should always base their planning on feedback from the experiences of users in current systems. o There is a need for social norms which will regulate the quality and appropriateness of electronic communication. o The scientific research community is heterogeneous, in terms of the ways in which users which different populations wish to use net- works and in the networking skills that have. o There is little provision made for the education and training of network users. CMC as Social Interaction Findings: How, then, does the literature treat CMC as a form of social interaction? o CMC occurs in a much narrower channel of communication which removes or suppresses the verbal and visual cues, especially facial expression and eye contacts. o With the removal of obvious status cues, CMC has the potential to allow greater equality among users,unless there are ways of including status cues in the text of the message. o The immediate feedback of a physical encounter is also blocked and can only occur as textual feedback during the exchange of messages, there by reducing its effect only to the creation of subsequent messages. o Users' attention becomes focused on the message or text as the source of the cues and other information which will help the recipient develop some conception of the sender and her/his intentions, motivations, status and interests. o The content and style of communication are dependent on the social cues that are available in the particular form of communication and the fewer cues that are present the more task-oriented and depersonalized the interaction will become. o The concept of social presence describes a subjective quality of communication which affects user interaction and if CMC is found to be high in social presence, it will allow psychologically close encounters to occur among users. o CMC removes constraints of time and space. o Rules that govern face-to-face interaction, such as turn taking, the obligation to immediately answer a question, the avoidance of interruptions and the display of deference to others of higher status, may not apply to asynchronous communication that originates in remote locations. o There is also a quality of impersonality to computer networking, due in part to the lack of immediacy which accompanies face-to-face and other synchronous forms of communication. Its effects are not clear. o CMC may increase the chances that interactions will be more uninhibited and impulsive and less socially desirable. Users may be less committed to existing guidelines of propriety. o Users may engage in the expression of negative affect, or "flaming", or they may take advantage of the constraints of CMC to establish some very warm and personal relationships. o There is a link between various forms of CMC and the maintenance of the social order. The use of keyboard based CMC may be reinforcing the rules and practices of bureaucratic rationality. o The use of CMC systems may involve a form of social interaction around which a new and different set of social norms are emerging. o The use of computer networking systems may be a critical element in the emergence of a CMC based subculture. o The spread of CMc systems through a variety of environments may be changing the fundamental nature of interpersonal communication by adding a new dimension or channel with its own unique characteristics. o The use of CMC may be changing the ways in which groups make decisions and reach a consensus. o The spread of CMC through the scientific research community may produce increases in productivity because of the new ways in which scientists can communicate with each other. o It may also be producing a wide range of cognitive, affective and behavioral impacts on individual users, groups and institutions. o CMC systems may be significant factors in creating organizational innovation and change. o CMC systems may have the potential to alter the ways in which organizational members exchange information.