---------------------------------------------------------------------------
FILE: email_history
--------------------------------------------------------------------------- Copyright 1996 Ian R. Hardy
(c) 1996 Ian R. Hardy Last update: 13 May 1996 This document is a history thesis paper discussing the evolution of
electronic mail on the ARPANET, a packet-switched research network 
commonly recognized as the forebearer of today's Internet. This paper 
attempts to provide a social history of computer mediated communication 
and Internet culture. This paper is currently in draft form. Any corrections or comments
would be much appreciated. The author may be reached at the following email addresses: Ian R. Hardy
hardy@uclink2.berkeley.edu
hardy@mobileweb.com This document is available via http at the following URL:
http://server.berkeley.edu/virtual-berkeley/email_history DISCLAIMER - PLEASE READ. This document is Copyright 1996 by Ian R. Hardy.  It may be freely 
redistributed in its entirety provided that this copyright notice is not 
removed.  It may not be sold for profit or incorporated in commercial 
documents without the written permission of the copyright holder.  
Permission is expressly granted for this document to be made available 
for file transfer from installations offering unrestricted anonymous file 
transfer on the Internet.This document is provided as is 
without any express or implied warranty.  Nothing in this document 
represents the views of the University of California at Berkeley,
or of David Engerman, the advisor for this thesis. Ian Hardy, the author of this document, is a student at the University of
California at Berkeley studying history and computer science. Acknowledgments
----------------------- Special thanks goes out to David Engerman, Ray Tomlinson, 
Brian Harvey, Judy O'Neill, Alex McKenzie, Robert H'obbes' Zakon, 
Edwin Diamond, Ronda Hauben, and George Fiske for their assistance 
and kindness.         

               The Evolution of ARPANET email   
                             by
                              
                              
                              
                        Ian R. Hardy
                              
                              
                              
                              
                              
                              
                              
                              
                              
                              
                              
                              
                              
                              
                              
                    History Thesis Paper
                  Instructor David Engerman
            University of California at Berkeley
                        Spring, 1996         

     This paper investigates one chapter in the history of computer mediated communication. Specifically, the evolution of human communication on the ARPANET, the technological forebearer of today's Internet, will be examined. Central to this historical inquiry will be the idea that technology, like art or literature, is a significant artifact of human society. The technologies people developed to talk to one another on the ARPANET of the early 1970's, as well as the things they talked about using these technological media, richly reflect the historical requirements and underlying social needs of the day.

     Communication on the Internet - and the alternative social space constructed by the act of communicating electronically - has emerged as a hot topic in several popular and intellectual circles. The growing popularity of computerized communication prompts discussion regarding many fundamental aspects of social interaction, including questions of privacy, community, cultural exchange, ownership of knowledge,  and governmental control of information. This paper attempts to probe computer mediated communication at its roots. It will examine the social history of email on the ARPANET, the first large-scale computer network ever built. After giving a brief overview of the ARPANET and its organizational history, I will begin by tracing the early concepts of communication which informed the creation of the ARPANET, including an analysis of speculative texts and project proposals predating the construction of the network. After this I will endeavor to examine the early methods of text communication which arose following the ARPANET's implementation in 1969, focusing specifically on the evolution of electronic mail. In analyzing these developments, I do not wish to play the role of sentimental engineer, dwelling pointlessly on rusting structures and the past intricacies of enabling technologies. I am interested, rather, in what the development of email reflected in a social sense. What needs prompted this development? Who composed the first groups to communicate using network mail? How can we classify the popularity, problems, and cultural issues which arose from this unique and still very new way of communicating?

     This paper will argue that the ARPANET was originally formulated as a medium for technical resource sharing. Network email, a utility for human communication, never factored into early ARPANET plans. Its creation in 1971 proved completely unanticipated. After email's debut, however, the perceived function of the ARPANET shifted to encompass human communication. ARPANET email garnered stunning popularity and in the process changed the definition of what computer networks were good for. I hope to show that a unique space of social interaction, mirroring larger cultural changes in American society of the 1960's, sprang up around email and was widely embraced by its users. Electronic mail caught on due to its profound relevance to the social needs of those who first took hold of it and used it to communicate. I will discuss this first email generation, a close-knit community of gifted technical researchers. The character of this group differed significantly from the global user population of the current Internet. In its small size and collegial atmosphere, the ARPANET user population more closely resembled a small town community than the big city metropolis of the Internet today. I will outline how this small town society constructed and utilized email, focusing on the social needs fulfilled by computer mediated communication, as well as the unique challenges it brought forth.

     The ARPANET is significant as the first successful distributed, wide-area, packet-switched network of computers ever built.[1] It is the first interactive computer network used to deliver email. The network protocols used to connect the constituent networks of today's Internet derive from communication methods originally developed on the ARPANET, and the current high-speed "backbone" data lines of the Internet follow the same general layout of the original ARPANET topology. The ARPANET of the 1970's and 1980's became the established standard onto which all other wide area networks were linked. The basic network services developed on the ARPANET - file transfer protocol, electronic mail, and remote login - still factor among the most widely used functions on the Internet today. In terms of this investigation, the most crucial aspect of the ARPANET's heritage is electronic mail, a technology which would permanently change the way people viewed computers and human communication.

     Network email developed as an application on the ARPANET. Historically, the ARPANET arose from a complex mixture of Cold War military prerogatives and the academic needs of a computer science research community just coming of age in the United States. The technological planning, design, and construction of the ARPANET was directed and funded by the United States Advanced Research Projects Agency, commonly known as ARPA. The U.S. Department of Defense founded ARPA in 1958 in direct response to widespread fears of U.S. technological weakness after the Soviet Union's 1957 launch of the Sputnik I artificial earth satellite.[2] ARPA's self-stated mission, "to help maintain U.S. technological superiority and guard against unforeseen technological advances by potential adversaries," plainly reflects the Cold War origins of its institutional development.[3] ARPA operated as a governmental research organization with no laboratories of its own: it funded research at public and private institutions in fields considered promising for future defense applications.[4] Computer science, an area of study just beginning to reach maturity at this time, was deemed to hold particular potential and received significant ARPA patronage in the decades following the organization's charter.

     In 1962, ARPA formed a new department, the Information Processing Techniques Office, or IPTO. The IPTO was assigned with the task of researching command and control technologies.[5] This office directed computer science research under the guidance of its first director, a behavioral psychologist from MIT named J.C.R. Licklider. Under Licklider, the IPTO would fund research at top institutions in several emergent fields of computer science, including computer networking.  This research led to the final implementation of the ARPANET in 1969 under IPTO director Robert Taylor.

     ARPANET planners never considered email a viable network application. This is not to say that the sharing of information was overlooked; information and resource sharing were, in fact, among the primary stated goals of the ARPANET's originators. The original goal of resource sharing, however, focused specifically on the research needs of ARPA's computer science subcontractors, who formed the privileged group of the network's first users. Resource sharing in this context meant the electronic distribution of geographically dispersed computing utilities and technical data to the local computer of any ARPA-funded researcher who might require them. This formulation did not originally encompass possibilities of email communication between researchers. The concept that electronic text communication between specialists forms a crucial aspect of scholarly research had not yet surfaced. The idea of using the ARPANET as a medium for human communication was not widely recognized until the creation of network email in 1971. Before this point, ARPANET planners focused on building a network for sharing the kinds of technical resources they believed computer researchers on interactive systems would find most useful for their work: programming libraries, research data, remote procedure calls, and unique software packages available only on specific systems.

     The ARPANET originally developed as a medium for sharing computer programs. Software portability was impossible between the multitude of  experimental computing platforms and operating systems which blossomed at cutting- edge ARPA research institutions.[6] As Lawrence Roberts, head of the ARPANET project in 1967, would later remark, "We had all of these people doing different things everywhere, and they were not sharing their research very well. So you could not use anything anybody else did. Everything I did was useless to the rest of the world, because it was on the TX-2 and it was unique machine."[7] ARPA subcontractors using incompatible systems found themselves unable to share software resources without investing vast amounts of energy porting them from one system to another. The ARPANET, as a proposed network of unified standards, was seen by its planners as a way to bridge this resource gap by allowing researchers to access and utilize resources on computing platforms very different from those at their own institutions. Rather than modifying programs to run on local machines, researchers would be able to simply log in to a remote machine on which the program was already available and use its resources to the greater benefit of ARPA research. ARPANET planners hoped that resource sharing would avoid the reinvention of work already done and allow innovations to come together, interact, and grow in a groundbreaking networked environment of scientific collaboration. Significantly, this idea of collaboration focused on the distribution of technical resources, rather than on resources for interpersonal communication such as email.

     The men who built the ARPANET fell into two groups. The first consisted of researchers at universities and private companies responsible for developing the network technologies necessary for constructing the ARPANET. The second group consisted of researchers working for ARPA's IPTO office as administrative directors. Individuals in this latter group generally hailed as well from research institutions. They worked with the IPTO in short stints to provide direction and allocate funds towards the most viable research programs. Of these administrators, J.C.R. Licklider factors prominently in the early planning of the ARPANET. He was the first director of the IPTO after its inception in 1962. Most importantly, Licklider first evangelized the cause of building a network for ARPA researchers, a cause which would later culminate in the construction of the ARPANET.

     Licklider's papers characterize the paucity of attention given to human communication in the planning stages of the ARPANET. In 1960 Licklider published a work entitled "Man-Computer Symbiosis."[8] In this work can be traced the early roots of his interest in augmenting human knowledge by means of computing machines. This interest would culminate in plans for creating an ARPANET to facilitate technical resource sharing. Licklider's paper presented hopes that, "in not too many years, human brains and computing machines will be coupled together very tightly, and that the resulting partnership will think as no human brain has ever thought and process data in a way not approached by the information-handling machines we know today."[9] To achieve this goal, Licklider suggested several fascinating new technologies which he believed would contribute to a new kind of "symbiotic" relationship between intuitive human cognition and automated computer precision. Many of the technologies proposed in Licklider's paper are today considered essential aspects of computing. He discussed such fields as timesharing, abstract information retrieval systems, graphing and presentation software, statistical packages, library catalog systems, natural language programming, modular programming, as well as intuitive display and user input devices. Licklider would continue, as the first director of the IPTO division of ARPA in 1963, to set the stage for serious research funding in all these areas. Most significant in terms of this analysis, however, is a paragraph in Licklider's report describing the importance of computer networking as a means of augmenting human thought capabilities through resource sharing: 

     It seems reasonable to envision, for a time 10 or
     15 years hence, a "thinking center" that will
     incorporate the functions of present-day libraries
     together with anticipated advances in information
     storage and retrieval and the symbiotic functions
     mentioned earlier in this paper. The picture
     readily enlarges itself into a network of such
     centers, connected to one another by wide-band
     communication lines and to individual users by
     leased-wire services. In such a system, the speed
     of the computers would be balanced, and the cost
     of the gigantic memories and the sophisticated
     programs would be divided by the number of users.[10] 
     In this paper, Licklider never mentions using computers as media for direct human communication. Potential network applications such as email do not surface. The network model laid out by Licklider in this early sketch focuses instead on the sharing of hardware and software resources - "gigantic memories" and "sophisticated programs" - as a means of augmenting human intellect and stimulating the development of ideas. The direct transmission of ideas from one person or group to another is not a part of Licklider's proposed framework. Long-distance communication, already well addressed by the existing postal system, the telephone, radio, and automated telex, may well have seemed too trivial an application for highly specialized computing machines which in those days still filled entire rooms. Licklider's aimed to make computers active participants in the formation of ideas rather than mere calculators of formulaic algorithms or digital transmission belts for ideas already formed. Electronic mail would simply have been too trivial an application to take its place among Licklider's ambitious plans for achieving "man-computer symbiosis."

     After becoming director of the IPTO, Licklider drafted a memorandum discussing plans for building an ARPA network. This document, addressed to "Members and Affiliates of the Intergalactic Computer Network," stressed the need to plan a common set of "network-wide conventions" among research sites to be implemented in a future "integrated network" of timeshared computer systems.[11]  Licklider's memorandum, written in the first year of his brief tenure at IPTO, is among the earliest ARPA documents discussing the possibility of networking research sites together. It contains a lengthy series of examples demonstrating the usefulness of resource sharing. Like "Man-Computer Symbiosis," however, it fails to mention the possibility of any kind of direct communication, such as electronic mail, between researchers over the network. Licklider's memorandum conceptualized the ARPANET as a means of providing greater access to computing resources in order to expand innovation in the ARPA research community. If there existed any interest in direct communication or social interaction over the network, early documentation produced by IPTO does not reflect it.

     In that same year, two ARPA program plans signed by Licklider commissioned UCLA and U.C. Berkeley to conduct a preliminary investigation of the issues involved in implementing computer networks.[12] In July of 1964, Licklider stepped down as director of the IPTO. His successors, first Ivan Sutherland, then Robert Taylor, would continue supporting the innovative areas of computer research originally allocated funding by Licklider.

     Taylor recruited Lawrence Roberts to direct the ARPANET project in 1967. Roberts, a network specialist from MIT's Lincoln Laboratory, authored ARPA program plan no. 723, entitled "Resource Sharing Computer Networks" a year later in June of 1968.[13] This plan called for the implementation of a network tying together ARPA research centers in order "to improve and increase computer resource productivity through resource sharing."[14] In his report, Roberts specified a time frame for the implementation, and enumerated various aspects of the scientific and military value of networking. After much planning and hard work, the ARPANET was constructed and went officially online in 1969 with four nodes located at the Stanford Research Institute, the University of California at Santa Barbara, the University of California at Los Angeles, and the University of Utah.[15] The network would later expand in size to 15 nodes by 1971 and provide a foundation for the development of email.[16]

     Ray Tomlinson sent the first email message across the ARPANET in 1971.[17] This initial network mail experiment, with its significant implications for the future of human communications and social interaction, occurred as a relatively informal event. "The first message," Tomlinson recounts, "was sent from myself on one computer to myself on another computer and its content was completely forgettable; probably 'qwertyuiop' or 'Testing 1-2-3.'"[18] The second message, sent out to other users of the network, was somewhat more interesting. It announced the availability of network email and gave instructions on how to address mail to users on other machines by suffixing "@<hostname>" to the user's login name. Tomlinson's email application was the first attempt to use the ARPANET as a medium for human communication. The stage had been set for a revolutionary new means of electronic interaction between ARPA researchers which would quickly become, as one ARPANET user later commented, "by far the biggest use of the Net."[19]

     Tomlinson named his network mail program SNDMSG. It was designed to run on the TENEX operating system, a popular timesharing package developed by Bolt, Beranek, and Newman for Digital Equipment Corporation's PDP-10 computer. Tomlinson derived the network version of SNDMSG from two preexisting software utilities. The first was an intra- machine email utility, a pre-network version of SNDMSG, consisting of a simple text editor which processed the names of the recipients and the body of a message, and then appended this information to a special file designated as the local recipient's mailbox file. The second utility was an early ARPANET file transfer program called CPYNET. Merging the two utilities together yielded an application that would process outgoing messages in the same manner as if they had been intended for a recipient on the local machine, but could then copy the message information across the network to be appended to a remote user's mailbox file. In this way, early email can be viewed as an extension to the pre-existing technology of local computer mail by means of the brand new technology of network file transfer. It is interesting to note that the original mail delivery process developed by Tomlinson differs little in essence from that employed over the Internet today. The novel use of network file transfer to carry electronic mail never factored in original plans. It was, according to one ARPANET user studying at MIT at the time, "totally an afterthought."[20]

     SNDMSG emerges as the first program to transfer email over the ARPANET. ARPANET email was not, however, the first use of computers for human communication. Electronic mail on individual computers had been in existence for nearly a decade before 1971. This form of communication, referred to as "intra-computer email" to distinguish it from the "network email" arising on the ARPANET, allowed user communities on solitary, timeshared computers to trade messages with one another. Intra-computer email did not proliferate over any kind of network; it was limited to serving the user population of a single machine. Tomlinson notes that intra-computer email was "popular among those who were regular time-shared computer users."[21] Seen in this light, sending email over the ARPA network in 1971 was, as Tomlinson puts it, "a natural extension" to the existing functionality of intra-computer email systems of the 1960's. Most intra-computer email systems offered users only very basic one-to-one messaging utilities. Some messaging systems, however, began in the 1970's to exhibit highly sophisticated features for facilitating electronic group discussions. The EMISARI conferencing system, developed by Murray Turoff in 1970 for the U.S. Office or Emergency Preparedness, was the first such system to arise.[22] The architects of the EMISARI system concerned themselves with developing a computerized medium for enhancing group decision making, information exchange, and opinion surveys between the geographically-dispersed offices of the Office of Emergency Preparedness. EMISARI was first used as an internal system to coordinate official policy regarding Nixon's wage-price freeze program of 1971.[23] Government conferees would access the system directly by means of long- distance telephone calls made from local teletypewriter terminals.     
      
            Although it offered a nondistributed, group- oriented communication medium fundamentally different than that of ARPANET email, Murray Turoff's EMISARI system signified a major contribution to the evolution of computer mediated communication. EMISARI provided a carefully planned medium for enhancing group interchange by means of computer technology. Turoff's work would inform the development of an entire lineage of research systems for electronic group communication. These groundbreaking projects included such systems  as EIES, FORUM, and PLANET. These packages allowed users on the same computer to trade messages and engage in real time textual conversation about a variety of topics. Unlike ARPANET email, research conferencing systems were generally of the intra-computer variety. Geographically- dispersed conferees lacked the benefit of a computer network to connect them together.[24]

     PLANET stood as an exception to this rule. The PLANET computer conferencing package, conceived and designed by Jacques Vallee of the Institute for the Future, serviced the ARPANET for a series of studies starting in 1973.[25] Conference participants could use PLANET's software from anywhere on the network by means of remote login. Use of this system, however, was restricted to designated participants. Although ARPANET email would expand to support group interaction early on by means of mailing lists, nothing approaching the sophistication of early computer conferencing systems like PLANET was ever made widely available on the network.

     There were several business-oriented electronic communication services also available at the time of ARPANET's email debut. TWX, Telex, and private telegraph switching systems such as Western Union all predated the construction of the ARPANET.[26] Such services, however, required the intervention of third-party technicians in the communication process, precluding any possibility of interactive use. The interactive nature of ARPANET mail, which allowed it to be used directly by the sender and recipient, effectively bypassed what one researcher termed "the teletypewriter priesthood of traditional electronic mail systems, such as Telex, TWX, and most facsimile services."[27] In addition, such electronic communication services all involved costly transmission fees which were, to the average ARPANET user, prohibitively expensive when compared to "free" email transported over the government- funded research network.

     "Talking" stood as an additional form of communication popular on the early ARPANET. This means of interaction offered users the experience of real-time text conversations over the network. "Talking" was conducted between people logged into the same ARPANET host by means of remote login. An absence of standardized distributed chat protocols in the early days of the ARPANET required all parties in a real- time discussion to be logged into the same machine. Remote users would take advantage of guest accounts commonly offered on ARPANET systems to log in from afar, check who was actively "talking" on the system, and join the conversation. Some users likened this experience of real- time interaction over the ARPANET, which utilized only very primitive software support, to the experiences of amateur radio operators.[28] "Mostly we were just talking to be talking," Brian Harvey recounts. "It was like HAM radio. If you're a HAM and you meet someone from Brazil or some place, it doesn't matter what you're talking about - the neat thing is that you're talking to somebody who's from Brazil."[29] Due partially to this sense of fascination with a new form of communication, real-time chat on the ARPANET remained popular for some time.

     Tomlinson's SNDMSG program continued its heritage after 1971 by yielding a bumper crop of software offspring. Progressive improvements ensured that electronic mail would be a truly useful and usable application at the user interface level. An example of this can be seen in the evolution of READMAIL, SNDMSG's partner utility for reading messages. READMAIL carried the drawback of displaying every received message in its entirety without providing any kind of subject indexing.[30] This program was later modified at the behest of two men who received far too much email to read it all at once: Lawrence Roberts, the director of IPTO from 1968-1973, and Steve Crocker, the IPTO program manager credited with writing the first ARPANET request for comment.[31] READMAIL therefore gave way to a new mail reading program called RD. Among several innovative features, RD offered users a comprehensive list of available messages, indexed by subject and date. Barry Wessler, also of ARPA, expanded upon RD to provide the ability to selectively delete messages. He christened his utility NRD.[32]

     Some time later, a programmer named Marty Yonke obtained a version of NRD and integrated it with SNDMSG, providing in the process a uniform interface to the functionality offered by both programs. This network mail utility, first named WRD and later redubbed BANANARD, was the first program of its kind to combine sending and reading functions. It also provided friendly on-line help features.[33] John Vittal of BBN followed Yonke's lead. He developed a mailing program based on BANANARD which allowed the additional functionality of message forwarding, automatic handling of address fields in message replies, and a user- configurable operating environment. This utility, bearing the command name MSG, offered features very similar to those of message handling programs used today.[34] An even more sophisticated mailing program, HERMES, was in development by Bolt, Beranek, and Newman at this time.[35] Email programs such as MSG, as well as many of its previously cited ancestors, were in wide use throughout the ARPANET in the years before 1977.

     The emergence of multiple mail handling programs prompted attempts to implement network-wide standards for the delivery of email.[36] These standards proposed a common protocol for delivering email which assured that two different mail programs would be able to effectively trade messages over the ARPANET. Email standards also specified the format of message "header" fields, ensuring an agreed- upon pattern for such crucial information as message sender, addressee, and subject. Before 1977, several standards were generally accepted by participating host computer systems on an informal basis.  RFC's 561 and 680, which were never officially adopted, are examples of such.[37] The official ARPANET standard for electronic mail, RFC 733, was finally adopted in 1977.[38]

     Email caught on fast after its 1971 invention. It became by far and away one of the most popular user applications on the ARPANET. In the words of one ARPANET researcher, the email service sparked by Tomlinson's first SNDMSG utility was "an instant success."[39] Given the original emphasis of ARPANET planners such as J.C.R. Licklider and Lawrence Roberts on purely technical resource sharing, the huge popularity of ARPANET email as a communication medium came to many as a rather unexpected shock. This "great surprise" is evident in a passage written in 1976 by an early email researcher, who likened the surge of network mail usage to a phenomenon of nature: 

     A surprising aspect of the message service is the
     unplanned, unanticipated, and unsupported nature
     of its birth and early growth. It just happened,
     and its early history has seemed more like the
     discovery of a natural phenomenon than the
     deliberate development of a new technology.[40] 
 The idea that network mail "just happened" was widespread among early email users. ARPANET mail, without any kind of preplanning or developmental discussion, surged in popularity within the network community. As Frank Heart, the director of the team which built the physical ARPANET infrastructure, would later comment, 

     When the mail was being developed, nobody thought
     at the beginning it was going to be the smash hit
     that it was. People liked it, they thought it was
     nice, but nobody imagined it was going to be the
     explosion of excitement and interest that it
     became. So it was a surprise to everybody, that it
     was a big hit.[41] 
     The unexpected nature of email's success is confirmed in the paucity of official literature pertaining to email in the first five years following 1971. It would take network administrators some time to catch up with the reality of the email craze. Initially, the use of network email seems to have been viewed as a somewhat superfluous, and therefore minimally publicized, application of the ARPANET. Email had not been discussed by ARPANET planners, and its use consequently did not factor into the officially recognized purposes of the multi-million dollar government research network. For the first five years or so after Tomlinson sent his first message, discussion of ARPANET email avoided entry into most areas of official discourse despite its great popularity as a means of private communication between ARPANET directors and research contractors. ARPA reports, as well as journal articles and scholarly presentations, all skirted the issue of ARPANET email until approximately 1976. The widespread view was that network mail was superfluous to the established "real" or "scientific" uses of the ARPANET; consequently, even though people used email quite regularly, they chose not to mention it in the domain of official or academic presentations. Other reasons for this official silence owe to the fact that human communication over the network was not considered a major research issue in the early years of the ARPANET. As one author has suggested, email may have seemed such a trivial operation from a technical perspective that even mentioning it in proposals for research investment would have been counterproductive.[42] Or, as Ray Tomlinson, looking back on the dearth of early commentary, noted: 

     Perhaps the lack of comment is because e-mail is
     such a natural use of computer networks that no
     one considered that it required any research. The
     researchers were mostly concerned with how to get
     the bits from the sender to the recipients. E-mail
     as a social phenomenon was scarcely considered.[43] 
     The gist of these observations is that the implementation of electronic mail was not viewed as a significant technical challenge in the early days of the ARPANET, when difficult lower-level issues such as network routing, the creation of host-to-host protocols, and the maximization of network throughput efficiency wholly absorbed the ARPA research community.

     Perhaps another reason for the lack of recorded discussion concerning early email lies in the simple fact that much discourse about email took place using email. In this event, the historian is faced with a rather unique challenge: as users delete messages to free up disk space, any discernible historical record begins to evaporate.  In 1975, for example, an "ARPANET Message Service Group," consisting of 70 researchers concerned with electronic mail, sprung up on the ARPANET and engaged in various discussions using the very same medium they were studying.[44] Because its primary method of information exchange was conducted electronically, much of the written record generated by this group most likely never met any kind of lasting, corporeal form. As email messages are deleted from disk, so also is their recorded content deleted from history. Due to its transient nature, electronic discussion can hope to find lasting recollection only in the collective memories of its human correspondents. This group memory unavoidably gives way to a folkloric tradition, blurring any traces of precise written content. Email as a historical source straddles the divide between written and oral histories. As such it poses a unique historiographical problem. This issue can only be expected to increase in relevance as more and more knowledge becomes constrained within the mutable limits of electronic media.

     By about 1976, email became a topic in a scholarly discourse in which it had not existed even a few years before. ARPA researchers took up the cause of email, once considered best kept quiet, and transformed electronic communication into a "scientific" subject fit for serious research. Several conference discussion groups taking place at the time attest to this development.[45] It is highly significant that the definition of resource sharing, that overarching official objective which informed the creation of the ARPANET, began to adjust itself at this time to include human communication as a vital aspect. As J.C.R. Licklider, along with co-author Albert Vezza, would write in 1978, "The sharing of information is the most important type of resource sharing."[46] This is an interesting change of focus from Licklider's original emphasis on resource sharing as a means of distributing hardware ("gigantic memories") and software ("sophisticated programs") resources among a base of computer researchers. Licklider's redefinition of resource to include communication seems to be a general reflection of a larger transformation in scholarly discourse which slowly took place after the debut of electronic mail. Once email came to be viewed as a genuinely useful and acceptable application of the network, scholarly conceptions of the ARPANET shifted to include human communication as an application in keeping with the network's main operational purposes. Official views moved grudgingly from silence to acceptance.

     Despite this movement in official perceptions towards acceptance of email, technical resource sharing did remain a vital application of the network. Remote usage of supercomputing facilities such as the ILLIAC IV at the University of Illinois, as well as specialized software like the Macsyma symbolic mathematics package at MIT, continued to be popular network applications.[47] Still, it is clear that the creation of email challenged traditional knowledge and engendered a whole new perception regarding the usefulness of computer networks for human communication. In the years following network mail, the ARPANET became a recognized medium for the general transfer of knowledge as well as a way to run sophisticated remote programs stored on other machines. As Licklider and Vezza would write in reference to this new perception of the ARPANET, "Truly effective transfer is a transfer not of data or of information but of knowledge and it flows through human interaction."[48] Licklider had moved full circle from his earlier insistence on technical resource sharing voiced in such papers as "Man- Computer Symbiosis."

     Despite a lack of official statistics for the early years of email, it is clear that the popularity of network mail spread with startling speed. "Network service was an immediate success," a 1976 study notes. "Message flow grew in volume to become the most visible (if not the heaviest) traffic component on the network."[49] Part of this initial popularity in the years directly following 1971 can be attributed to the fact that Tomlinson's software was written for the PDP-10, a computing platform of choice at many ARPA research sites.  In a listing of ARPANET nodes planned for March, 1972, nine of the twenty-four total node locations are shown to be using the PDP-10 for host computing, indicating a clear plurality.[50] Network users offered several reasons of their own for the popularity of email as a tool for human communication. As Licklider and Vezza wrote in "Applications of Information Networks:" 

     It soon became obvious that the ARPANET was
     becoming a human-communication medium with very
     important advantages over normal U.S. mail and
     over telephone calls. One of the advantages of the
     message systems over letter mail was that, in an
     ARPANET message, one could write tersely and type
     imperfectly, even to an older person in a superior
     position and even to a person one did not know
     very well, and the recipient took no offense. The
     formality and perfection that most people expect
     in a typed letter did not become associated with
     network messages, probably because the network was
     so much faster, so much more like the telephone. 
The authors go on further to note: 

     Among the advantages of the network message
     services over the telephone were the fact that one
     could proceed immediately to the point without
     having to engage in small talk first, that the
     message services produced a preservable record,
     and that the sender and receiver did not have to
     be available at the same time.[51] 
     It is clear from these statements that network email occasioned a significant social change in the way ARPANET users communicated with one another. The popularity of email was not merely a result of its great speed and efficiency: the ARPANET community embraced email because it created a new culture of interaction which satisfied several key social requirements. Typed communication, which had before email almost always required the intervention of a professional secretary, became a more personal mode of interchange. Sloppy writing and imperfect grammar between educated correspondents suddenly became tolerated. Traditional barriers of age, rank, and unfamiliarity began to dissolve in the informal social space created by computer mediated communication. Small talk and artificial niceties, the chaff of a small-town America of days past, were blown into the wind in favor of a more direct parlance which maximized the time of the sender and receiver, but also suggested a desire for greater immediacy and directness in conversation. Viewed from this perspective, email offered more to its users than an increased sense of convenience. The early communication technologies of the ARPANET created a unique social space which in many ways reflected larger cultural shifts taking place in American society throughout the turbulent years of the 1960's. While it may be a stretch to call the Department of Defense-sponsored ARPANET of the early 1970's, with its computer-loving constituency of Ph.D.'s and graduate student researchers, a vehicle for American counter-culture on a par with campus "radicals" of 1968, there is no doubt that the cultural space of the ARPANET was far from immune to external developments of social change. Email reflected this. In its disdain for false formality, its distrust of traditional hierarchy, its time-selfishness, speed, and certainly in its ironic juxtaposition of impersonality and emotional directness, electronic mail surged out from the ARPANET as the definitive communication tool of an America wracked with change.

     ARPA researchers recognized the impact of network mail as a new and significant social medium. As one document put it: 

     We in the ARPA community (and no doubt many others
     outside it) have come to realize that we have in
     our hands something very big, and possibly very
     important. It is now plain to all of us that
     message service over computer networks has
     enormous potential for changing the way
     communication is done in all sectors of our
     society: military, civilian government, and
     private.[52]
     
     Many in the ARPANET community felt the social advantages of email significant enough to justify drastic changes in their normal communication patterns. People shunned the phones and post office in favor of network mail service. "Use of the service," the same researcher noted, "has had a substantial impact on the organizations involved, stimulating dramatic shifts of dependence away from the traditional media (postal service, telephone)."[53] Network mail offered its users many conveniences which made it far more efficient than other forms of communication. Messages were transmitted instantly to their destinations. Users could compose messages personally, whenever and wherever they felt prepared to do so. The ability to send and receive messages from any geographic point of the network made email an excellent medium for traveling research professionals, and the decoupling between sender and receiver occasioned by email made it possible for two parties to communicate without being present at the same time. Most significant, however, was the emergence of a distinct social space around the use of electronic mail. This new electronic culture uniquely fulfilled the requirements of its user population, spurning tiresome formality, loosening cultural conceptions of seniority and rank, and encouraging succinct directness in the exchange of knowledge.

     ARPANET users viewed themselves as a unique community. As J.C.R. Licklider would later comment, looking back on the kinds of people who used the ARPANET, 

     I think that the main thing ARPA has had is a
     series of good people running an office, and a
     fantastic community. I guess that's the word. It
     was more than just a collection of bright people
     working in a field. It was a thing that organized
     itself  a little bit into a community, so that
     there was some competition and some cooperation,
     and it resulted in the emergence of a field.[54] 
     The ARPA community Licklider spoke of, while composed of many different people, reflected several common attributes. In accordance with ARPA's charter to "develop imaginative, innovative, and often high risk research ideas offering a significant technological impact that will go well beyond the normal evolutionary developmental approaches,"[55] the research community directed and funded by the IPTO office of ARPA were the best in their fields, and generally knew it. As Brian Harvey notes, "You couldn't get on the Net unless you were very accomplished."[56] ARPA researchers developed much of what still passes for the state of the art in networks, computer graphics, interactive computing, expert systems, and natural language communication systems. Consequently, the research community which embraced the ARPANET was bound by a set of common interests and similar backgrounds in computer-related fields. The resulting spirit of collaboration which arose as a result of interactions within this group often resulted in widespread feelings of animation and excitement. As Frank Heart, looking back at his experiences in the ARPA community, would later comment, 

     You know the people who were involved in this were
     all having a very good time. The ARPANET was a big
     thing in most of their lives. While they might
     have some differences, the overriding single bit
     was that they were all having a great time. And
     they all thought it was very exciting.[57]
     
Researchers on the ARPANET worked at top institutions. In 1971, the year Tomlinson sent his first email message, fifteen research sites communicated on the network. These nodes included institutions such as UCLA, Stanford University, Harvard, UCSB, and MIT.[58] Private think tanks and research groups, such as the Santa Monica-based RAND corporation, or Bolt, Beranek, and Newman, Inc., of Massachusetts, also conducted computer research and development commissioned by ARPA. Bolt, Beranek, and Newman (BBN), the organization associated with Tomlinson's first email experiment, was a private research firm consisting of a number of former MIT scholars in the fields of behavioral psychology, acoustics, and computer science. J.C.R. Licklider, himself a behavioral psychologist who taught at MIT, joined BBN as a vice president in 1957 before organizing ARPA's IPTO office five years later. In addition to being connected with the first email experiment on the ARPANET, BBN was responsible for designing and implementing the ARPANET's physical infrastructure.[59] BBN also held the ARPA contract to supervise the network's overall operation and perform routine upgrades.

     Because ARPA researchers were bound by a common professional pursuit, they often developed personal relationships. Friends and colleagues would run into one another at the numerous conventions, colloquia, and workshops conducted by the computer science research community. Before the advent of the ARPANET and email, the ARPA community maintained personal communication with one another primarily through organized professional meetings, as well as through the media of phone calls, memoranda, letters, and research journal articles. Consequently, the ARPANET grew up around a community of people who generally knew one another, or at least knew of one another. This fact illustrates much about early interactions over email. As Brian Harvey put it, "Almost all the time the person you were talking to was someone you knew."[60] This small-town ambience clearly marks a difference from the megalopolis anonymity of today's Internet.

     As more and more people came to depend on the ARPANET for social exchange, efforts materialized to provide directory services listing the people and organizations located along the stretch of the network The Network Information Center, or NIC, offered such services to the ARPANET community.  The NIC was maintained by the Stanford Research Center. Original ARPANET plans mandated that the NIC act as a depository for information relevant to ARPANET users, including documents describing network protocols, and other information pertinent to ARPA resource sharing.[61] In 1970 the NIC published the first ARPANET Directory, an inch- thick book containing descriptions of every host computer on the network. The NIC also made an online database containing Directory information available to network users. The NIC Directory highlighted the unique resources available on each ARPANET machine, inviting network users to "visit" by means of remote login and try them out. Information about any host computer, most of which offered guest accounts to those who cared to visit, could be located at the flip of a page or a query of the NIC database. In this way, the Directory sought to encourage resource sharing and open exchange over the ARPANET. After the advent of electronic mail, the NIC began adding network mailbox addresses to its online identification data base and to the hardcopy ARPANET Directory.[62] This addition of email information was significant in enhancing a sense of community between users of ARPANET mail.  The Directory served as an "email phone book" which made it easier for network users to keep in touch and access professional information about email correspondents.

     The format of electronic mail derives from the scholarly-professional community which first designed it. Beginning with Tomlinson's first message, the format of email evolved according to the model of the institutional memorandum, with its header structure containing "To:," "From:," "Subject:," and "cc:" fields.[63] This traditional form of institutional correspondence, well familiar to the scientists and governmental directors who made up the early ARPANET community, has persisted in email to this day. Some elements gleaned from the memorandum format, such as the "carbon copy" or "cc" header, lost their original denotation in the shift to the noncorporeal medium of electronic mail.

     The institutional format of electronic mail was a fitting one, given its early uses. The ARPA community appears to have first utilized network mail as a computerized substitute for internal correspondence such as memoranda. As one early study noted, 60 to 80 percent of all ARPANET messages were addressed to people in the same office area.[64] Electronic mail over the ARPANET quickly became an integral component in institutional communications patterns. Primarily for this reason the ARPA community began to view the ARPANET as an essential utility on a par with the telephone system, electricity, or jet transportation. "People began to depend upon it," Frank Heart recounts. "Especially as electronic mail became an important component of the use of the system, people began to assume it was going to exist."[65] This general assumption demonstrates a profound reliance on email within the ARPA community. The ARPANET came to be viewed as an indispensable resource precisely because it carried network mail.

     A look at favorite ARPANET mailing lists, which allowed email messages to be broadcast among a large group of participants, indicates a wide variety of discussion topics flowing across the network the early 1970's. Subject matter tended to be for the most part eclectic; however, discourse regarding computers did certainly dominate many mailing lists. As Ray Tomlinson observed, looking back to the subject matter of early electronic mail, 

     Things have not changed much. Most [ARPANET] e-
     mail concerned the interests of its users (as it
     does today). The difference was that for most of
     the people with access to e-mail, the word
     'computer' figured prominently in their resume.
     Topics ranged from bug reports and meeting
     announcements to recipes for pretzels.[66] 
Tomlinson's SNDMSG program supported mailing lists soon after its 1971 implementation. Among the most popular ARPANET mailing lists were NETWORK-HACKERS, dealing with programming and host protocol issues over the ARPA network; SF-LOVERS, for discussion of science fiction literature; WINE-TASTERS, for discourse between glass-tipping connoisseurs of the communication age; and HUMAN-NETS, a forum for examination of human factors in networks and computer science.[67] In the early years before electronic communication became an acceptable component of ARPANET usage, conflicts would sometimes arise between network administrators and users of high-distribution mailing lists such as SF-LOVERS. At times the flow of messages distributed from such lists would be so great as to threaten to jam the network. Generally network administrators would permit the lists to continue only after ARPANET users convinced them that the vast flow of messages posed an important research problem for command and control networks.[68]

     An investigation of 1981 message archives from HUMAN- NETS illuminates the general tone of group discussion over email. In HUMAN-NETS, ARPANET users confronted issues concerning the growing prevalence of computers and computer mediated communication in everyday life. Topics discussed on HUMAN-NETS ranged from issues regarding gays and computing, environmental discussions, debates on computers and adolescent development, to developing etymologies of hacker terms, or "computerese." This last item of discussion prompted several emotional responses, including the following: 

     Date: 24 May 1981 0259-CDT
     From: Clyde Hoover <CC.CLYDE at UTEXAS-20>
     Subject: Inflicting of hacker language upon others
     
     [...]
     
     My feelings about hacker jargon (and all other
     vocational vocabularies), is that they should be left at
     work, and one should talk normal English to the people 
     around you (especially spouses and
     companions).  In short, STOP BEING A HACKER when
     you aren't working and be a regular human being
     relating with other people on their own terms.[69] 
An examination of electronic mail archives from early ARPANET discussion groups provides excellent insight into the concerns of the community which first used network email. For a small selection of early messages from HUMAN- NETS, please refer to the appendix.

     In addition to group communication over mailing lists, the ARPA community also found electronic mail to be a useful medium for the submission and exchange of scholarly papers. In this regard, email provided a form of "virtual journal." This application satisfied the unique requirements of ARPANET's scholarly research community. As the director of a 1976 conference, The Berkeley Workshop on Distributed Data Management and Computer Networks, remarked in the conference digest's introduction page: 

     It is interesting to note that many of the
     abstracts and several complete papers were
     delivered via electronic mail. The last paper
     submitted for these proceedings was copied on a
     high-quality terminal at LBL [Lawrence Berkeley
     Labs] the day after the "deadline"; I was even
     able to correct a few typographical errors before
     copying the paper. This practice is becoming quite
     common for some journals and newsletters, and is
     proving to be a viable medium for such tasks.[70] 
     Scientists, unhappy with the slow spread of knowledge occasioned by traditional forms of scholarly media such as printed journals, found in electronic mail new possibilities for publishing their findings. Several efforts following the advent of email, including one which resulted in the development of the Usenet conferencing network, focused precisely on developing a viable means of computerizing the journal submission and delivery process.[71]

     In addition to advantages, network users observed several problems relating to the new social dynamics occasioned by electronic mail. Put simply, people did not act the same over electronic media. As Hiltz and Turoff noted in a 1978 article on computerized conferencing, users noticed many deficiencies inherent in electronic, textual communication. "The most obvious," they wrote, "are the need for typewriting and the lack of eye contact, facial expressions, gestures, and verbal intonations that lend so much richness to face-to-face and even telephone conversations."[72] The observations of this group apply well to the medium of ARPANET electronic mail, which was surging across the continent in ever thickening streams as sociological studies of computer conferencing were just beginning to move forward. The lack of visual cues and verbal richness characterizing electronic mail, coupled with a less formal writing style and instantaneous delivery, could cause problems when parties in the heat of emotion exchanged messages. As one early ARPANET user has noted, "It was easier to get angry over the Net."[73] The decoupled, delayed nature of email also added to difficulties felt by users wishing to resolve differences or correct misunderstandings over the expansive territory of the ARPANET.

     Researchers on the ARPANET noticed additional problems arising from the new network mail. Concerns for privacy in electronic communication prompted attempts to adapt data encryption technology to the transmission of electronic mail. Message authentication and issues of "digital signatures" designed to provide failsafe identification of email senders also emerged as issues of research at this time.

     One major perceived deficiency of ARPANET mail rested in the design of its addressing structure.[74] The simplicity of the standard email address, commonly adopted to be the "user@hostname" convention worked out by Ray Tomlinson in his original experiments, did not befit a fast growing ARPANET hosting burgeoning user populations. The address scheme was simply too ambiguous. If an email sender knew the name of his or her intended recipient, but not the name of the recipient's host computer, difficulties in determining the correct email address would ensue. People with similar names at different hosts ran the risk of receiving one another's misaddressed messages. Alternative methods of recipient identification were discussed, but realistic solutions proved difficult to come by. Unique numeric identifiers for email users were discussed but quickly deemed to be out of the question: "Persons who do not object at all to having a telephone number go into a three-way tizzy fit when they are assigned a computer number," wrote one frustrated researcher.[75] Proposed uniqueness constraints on user names stood to be just as unpopular. "It can be stated categorically that humans adher [sic] strictly to the Rule of Numero Uno, and do not take lightly to being told that they are JJONES-3 instead of the first and only JONES on the network."[76] The most elegant solution, to establish a comprehensive user data bank on the ARPANET which message software could consult to resolve address ambiguities, never materialized despite numerous research efforts underway by 1977.[77] Lack of sufficient locator or directory services still proves a nagging problem on the Internet of today, where address problems continue to plague user populations of now world-sized proportions.

     The small town, collegial nature of the ARPANET community minimized social problems relating to email. Because the ARPANET community was limited in its early years to people who shared common backgrounds and generally knew one another from a variety of professional interactions, people tended to be aware of accepted modes of behavior. The social space of ARPANET mail was in this way regulated by the collegial nature of its user community. One researcher, discussing network email, took note of this phenomenon: "The closed community of the ARPANET and its research orientation yield a situation different from what could be expected outside."[78] Therefore, issues of email privacy and encryption remained primarily academic. In daily network usage, there existed little fear of eavesdropping or data tampering by friends and colleagues in the ARPA community. Similarly, the prevalence of address ambiguity and user location problems was tightly constrained by a small network population the availability of printed aids such as the NIC email "phone book." Most significantly, the collegial nature of the ARPANET community served to tame most verbal altercations and disagreements arising over email, referred to in computer parlance as "flame wars." In this way the social space surrounding early ARPANET email diverges notably from that encompassing electronic discussions on the Internet today, where "flaming" between correspondents who are strangers to one another can conflagurate to reach brutal peaks of vulgarity.

     After the ARPANET's first international connections linked nodes at the University college of London and the Royal Radar Establishment of Norway in 1973,[79] international issues surrounding email and information sharing slowly surfaced. Around 1976, the year the Queen of England is reported to have sent her first email message over the Atlantic,[80] researchers facing an emergent system of global interchange began to discuss how network information resources might benefit developing nations. Licklider and Vezza, for example, speculated on the possibilities of using computer mediated communication for international benefit. These researchers focused on the knowledge transmission potential of wide area networks: 

     Networks may link the geographically separated
     subcritical foci of cognition in the developing
     world with the concentrated supercritical centers
     of the developed world, bringing the former deeply
     into the interaction patterns of the latter and
     making it much easier for the former to grow and
     advance.[81] 
This quote demonstrates the view that electronic communication provided an important system of knowledge exchange. Licklider and Vezza hoped that this system of exchange might be applied to the benefit of developing nations, or "subcritical foci of cognition." "Indeed," the authors continue, "it seems likely that technology transfer to developing countries could become real and effective through no more than informal extension of patterns of interaction that have become well established in the ARPANET community."[82] These "patterns of interaction," discussed by Licklider and Vezza, refer directly to communication by means of electronic mail. Seen from this perspective, it becomes apparent that the authors considered email to constitute a hopeful torch of knowledge and technology which might be used to illuminate the presumed ignorance of nations shrouded in prescientific darkness. Licklider and Vezza, influenced by the scholarly ethos pervading the early ARPANET community, felt naturally that this ethos might be passed along to users of any network allowing computerized communication, even networks potentially located in the undeveloped world.

     The authors, however, did not consider email to be completely benevolent in this regard. The prospect of "electronic imperialism,"[83] in which developing nations might be dominated by reliance on network communications maintained by technological superpowers, was considered by Licklider and Vezza to be a potential negative consequence of a global ARPANET. "From the point of view of the imperialist," they suggest, "it may well be that packet networks will be to the not-so-distant-future what clipper ships (or were they packet ships?) and clipper aircraft were to the not-so-distant past."[84] In their report, Licklider and Vezza also consider questions of transborder data flow, including restrictions on the movement of sensitive technological information and totalitarian information control by national governments. It was clear to these researchers that network descendants of the ARPANET would grow to constitute "the nervous system of the world," In such case, they write, "arrangements will have to be dictator-proof."[85]

     Today's Internet follows the ARPANET as exactly the global "nervous system" Licklider and Vezza originally envisioned. While it is unclear whether the Internet constitutes a torch of knowledge to illuminate the developing world, it is clear that email has overtaken the Internet to become a profound component in the way we interact today. In the formative years of the ARPANET, countless crucial lessons were learned. People realized that computer networks offered more than a sophisticated way to share programs. With the advent of Ray Tomlinson's first email utility in 1971, network users suddenly discovered the attraction of communicating with one another in an entirely new social space. A unique culture of electronic interaction sprang up around ARPANET email. This email culture mirrored larger shifts in the landscape of American society, shifts which satisfied the needs of its practitioners by spurning outmoded formality, loosening cultural conceptions of seniority and rank, toppling hierarchical restrictions, and encouraging succinct directness in the exchange of knowledge. A whole new etiquette infused ARPANET email which continues to thrive within Internet email to this day. The Internet megalopolis has inherited the medium of electronic mail, its culture, advantages, and persisting problems, from small town ARPANET roots. Many problems which did not concern the tight-knit community of the ARPANET emerge as larger issues in the contemporary, expanded environment of the Internet. It is the sincere hope of this author that I have been able to provide some background regarding the growth of computer mediated communication. With luck, it will allow users of the Internet today to confront the unique challenges facing them with a firm understanding of the cultural tradition they inherit.  
---------------------------------------------------------------
                           NOTES
--------------------------------------------------------------- 
[1] Distributed: As opposed to a centralized, or star, network,
in which data communication between every network site is
controlled from a central location, a distributed network is
characterized by spreading data routing and control
decisions  to every node.  Wide-area: Wide area networks
cover geographic distances greater than the one mile limit
of most local area networks. The original ARPANET linked
sites from coast to coast on the North American continent.
Packet-switched: As opposed to circuit-switched networks,
such as the telephone system, which maintain a direct
connection between two communicating sites for the duration
of a data transaction, packet-switched networks break
continuous data into discrete "packets," which are
transmitted one at a time and individually routed to their
destination, where they are later reassembled. These three
properties produced a network of computers which facilitated
email by allowing the transport of data between divergent,
geographically dispersed computing platforms with speed and
efficiency. [2] Edwin Diamond, Stephen Bates, Mark Horowitz, "The Ancient
History of the Internet," American Heritage, 46(6) (Oct.,
1995), 34. [3] "Advanced Research Projects Agency - Mission," ARPA
Homepage, http://www.arpa.mil/. Accessed April, 1996. [4] Judy O'Neill, "The Role of ARPA in the Development of the
ARPANET, 1961-1972," IEEE Annals in the History of
Computing, 17(4), 1995, p. 76. [5] This office was originally designated the "Command and
Control Research Office." Upon joining it as director in
1963, J.C.R. Licklider changed the name to "Information
Processing Techniques Office" to reflect a sense of wider
mission - a mission which included the support of a broad
spectrum of cutting-edge U.S. computer science research.
O'Neill, "Role of ARPA," p. 77. [6] Software Portability refers to the ability to compile or
run the same program on multiple machine architectures. [7] Lawrence Roberts, interview by Arthur L. Norberg, tape
recording, San Mateo, California, 4 April 1989, OH 159,
Charles Babbage Institute, University of Minnesota,
Minneapolis, Minnesota, from transcript, p. 4. [8] J.C.R. Licklider, "Man-Computer Symbiosis," IRE
Transactions on Human Factors in Electronics, 1(1), (March,
1960), p. 4. Licklider was working for the Massachusetts
research firm, Bolt, Beranek, and Newman, at the time he
wrote this article. [9] Licklider, "Man-Computer Symbiosis," p. 4. [10] Licklider, "Man-Computer Symbiosis," p. 7. It is
interesting to note that Licklider's date projection is a
near-perfect forecast of the actual implementation of the
ARPANET, which took place beginning in 1969. [11] J.C.R. Licklider, "Memorandum to Members and Affiliates of
the Intergalactic Computer Network," 23 April 1963, RG 330-
69-A-4998, Box 3, Folder: 350-1, National Archives and
Records Administration. The "members and affiliates" named
in the memo's distribution list included: D.C. Engelbart,
John H. Wensley, and Roy Amara of the Stanford Research
Institute; John McCarthy of Stanford University; Harry D.
Huskey, Edward Feigenbaum, and David C. Evans of U.C.
Berkeley; George W. Brown of UCLA; Alan J. Perlis and Allen
Newell of the Carnegie Institute of Technology; Robert M.
Fano, Fernando J. Corbato, and Marvin Minsky of MIT; and
Keith W. Uncapher of the RAND Corporation. Many of these men
factor as prominent contributors to the history of ARPA
research and, indeed, computer science research in general. [12] ARPA, Program Plan No. 93, "Computer Network and Time-
Sharing Research," 5 April 1963, RG 330-78-0013, Box 1,
Folder: "Program Plans," National Archives and Records
Administration; and
ARPA, Program Plan No. 95, "Remote Stations and Programs for
Computer Network," 5 April 1963, RG 330-78-9913, Box 1,
Folder: "Program Plans," National Archives and Records
Administration. [13] ARPA, Program Plan No. 723, "Resource Sharing Computer
Networks," 3 June 1968, RG 330-78-0013, Box 1, Folder:
"Program Plans," National Archives and Records
Administration. [14] ARPA Plan, "Resource Sharing Computer Networks," p. 1. [15] Robert H'obbes' Zakon, "Hobbes' Internet Timeline v2.3,"
http://info.isoc.org/guest/zakon/Internet/History/HIT.html,
or email timeline@hobbes.mitre.org. [16] University of California at Los Angeles (UCLA); Stanford
Research Institute (SRI); University of California at Santa
Barbara (UCSB); University of Utah (UTAH); Bolt, Beranek,
and Newman, Inc. (BBN); Massachusetts Institute of
Technology (MIT); Rand Corporation (RAND); Systems
Development Corporation (SDC); Harvard University (HARVARD);
Lincoln Laboratory, MIT (LINCOLN); Stanford University
(STANFORD); University of Illinois (ILLINOIS); Case Western
Reserve (CASE); Carnegie-Mellon University (CMU); and Ames
Research Center, NASA (AMES). See Zakon, "Hobbes' Internet
Timeline v2.3." [17] Ray Tomlinson, personal communication with author, email, 9
April 1996. [18] Ray Tomlinson, personal communication with author, email,
10 April 1996. [19] Brian Harvey, interview by author, tape recording,
Berkeley, California, 21 March 1996. [20] Harvey, Interview. [21] Tomlinson, 10 April 1996. [22] Murray Turoff and Star Roxanne Hiltz, "Meeting through your
Computer," IEEE Spectrum (May 1977), p. 58. Murray Turoff
and Star Roxanne Hiltz, The Network Nation: Human
Communication via Computer (Massachusetts: Addison-Wesley,
1978), p. 46. EMISARI is an acronym for Emegency Management
Information System and Reference Index. [23] Turoff and Hiltz, "Meeting through your Computer," p. 59. [24] This situation would change with the advent of public
packet-switched network services, such as Telenet, first
introduced in 1975. Such services provided an alternative to
ARPANET, the use of which was limited to the ARPA research
community. Many conferencing systems used public network
services such as Telenet to allow remote login by far-flung
conferees. The basic messaging system supported by
conferencing systems, however, in which messages were
located on a single computer and not transmitted over the
network, would remain intra-computer in nature. [25] Jacques Vallee, "The Forum Project," Computer Networks 1
(1976), p. 39. [26] The principle of packet-switching, which characterized the
system of data transmission over the ARPANET, was based on
store-and-forward message switching techniques derived in
part from observing the operation of telegraph systems.
Packet switching was first elaborated in 1964 by Paul Baran
of the RAND corporation, and later, independently by D.W.
Davies of the National Physical Laboratory, U.K., in 1965.
Please refer to: Paul Baran, "On Distributed Communications
Networks," IEEE Transactions on Communications Systems COM-
12 (1964), p. 1, and Martin Campbell-Kelly, "Data
Communications at the National Physical Laboratory (1965-
1975)," Annals of the History of Computing, 9(3/4) (1988),
p. 266. [27] Raymond Panko, "An Introduction to Computers for Human
Communication," NTC '77 Conference Record, (New York: IEEE,
1977), p. 21:1/1-6. [28] Significant software support for real-time, distributed
chatting on the network would not arise until Jarkko
Oikarinen's creation of Internet Relay Chat (IRC) in 1988.
Please see Zakon, "Hobbes' Internet Timeline v2.3." [29] Harvey, Interview. [30] T.H. Myer and John Vittal, "Message Technology in the
ARPANET," NTC '77 Conference Record, (New York, NY, USA:
IEEE, 1977), p. 21:2-1. [31] Requests for Comment, or RFC's, were documents which grew
from being informal proposals between ARPANET programmers
developing new network applications to be the standard form
of protocol specification on the Internet today. Steve
Crocker's central position in the early proposal process
ensured that he received a daily mountain of email from
across the expanses of the ARPANET. [32] Myer and Vittal, "Message Technology," p. 21:2-5. [33] Myer and Vittal, "Message Technology," p. 21:2-5. [34] Myer and Vittal, "Message Technology," p. 21:2-6. [35] T.H. Myer and D.W. Dodds, "Notes on the Development of
Message Technology," Berkeley Workshop on Distributed Data
Management and Computer Networks, (1976), LBL-5315, Lawrence
Berkeley Laboratories, p. 146. [36] All network applications require published standards making
clear the methods of data exchange, or protocols, which are
to be used by computers running that application over the
network. This process of standardization ensures that
different kinds of computers, each speaking a different
"language" of binary signals, can interact in such a way as
to provide a standard service to the user running the
application. Standards on the ARPANET were known as RFC's,
or Requests for Comment, and were generally written by
informal committees. Internet standards today continue to be
called RFC's. [37] Myer and Vittal, "Message Technology," p. 21:2-2. [38] David Crocker, John Vittal, Kenneth Pogran, and D. Austin
Henderson, Jr., "Standard for the Format of ARPA Network
Text Messages(1)," ARPANET Request for Comment 733, 21
November 1977. [39] Elizabeth Feinler, "The Identification Data Base in a
Networking Environment," NTC '77 Conference Record, (New
York: IEEE, 1977), p. 21:3-1. [40] Myer and Dodds, "Notes," p. 145. [41] Frank Heart, interview by Judy O'Neill, tape recording,
Cambridge, Massachussets, 23 March 1990, OH 186, Charles
Babbage Institute, University of Minnesota, Minneapolis,
Minnesota, from transcript, p. 15. [42] Maurice Wilkes, "Networks, Email, and Fax," Communications
of the ACM, 33(6), (June, 1990), p. 631. [43] Tomlinson, Interview, 10 April 1996. [44] Panko, "Computers for Human Communication," 21:1-2. [45] Among these are the Berkeley Workshop for Distributed Data
Communications (1976), NTC '77 (1977), 1976 International
Conference on Communications (1976), and the Eighth
International Symposium on Human Factors in
Telecommunications (1977). [46] J.C.R. Licklider and Albert Vezza, "Applications of
Information Networks," Proceedings of the IEEE, 66(11),
(November 1978), p. 1330. [47] Harvey, Interview. [48] Licklider and Vezza, "Applications," p. 1342. [49] Myer and Dodds, "Notes," p. 145. [50] Lawrence Roberts and Barry Wessler, "The ARPA Network,"
Chapter 13, Computer-Communication Networks, Norman
Abramson, Franklin Kuo, ed. (New Jersey: Prentice-Hall),
1973. [51] Licklider and Vezza, "Applications," p. 1331. [52] Myer and Dodds, "Notes," p. 145. [53] Myer and Dodds, "Notes," p. 145. [54] J.C.R. Licklider, interview by William Aspray and Arthur
Norberg, tape recording, Cambridge, Massachussets, 28
October 1988, OH 150, Charles Babbage Institute, University
of Minnesota, Minneapolis, Minnesota, from transcript, p.
28. [55] "Advanced Research Projects Agency - Mission," ARPA
Homepage. [56] Harvey, Interview. [57] Heart, Interview, p. 19. [58] Zakon, "Hobbes' Internet Timeline v2.3." [59] BBN's contributions in this area included the development
of the ARPANET's Interface Message Processors (IMP's), which
were Honeywell 516 minicomputers responsible for switching
message packets between each network node. The network
implementation project was led by Frank Heart in 1969. [60] Harvey, Interview. [61] Feinler, "Identification Database," p. 21:3-1. [62] Feinler, "Identification Database," p. 21:3-1. [63] Myer and Vittal, "Message Technology," p. 21:2-2. [64] Panko, "Computers for Human Communication," p. 21:1-6. [65] Heart, Interview, p. 9. [66] Tomlinson, Interview, 10 April 1996. [67] Gleaned from sources quotes in: Hauben, Ronda, "The
Evolution of Usenet News: The Poor Man's ARPANET," speech
presented at the MACUL Conference, Detroit, Michigan, 12
March 1993. [68] Hauben, "Evolution." [69] Usenet History Archives,
ftp:weber.ucsd.edu/pub/~usenet.hist/. [70] Donald Austin, "Introduction," Berkeley Workshop on
Distributed Data Management and Computer Networks, (1976),
LBL-5315, Lawrence Berkeley Laboratories, p. iii. [71] Richard Roistacher, "The Virtual Journal," Computer
Networks 2 (1978), p. 18; and Sandra Emerson, "Usenet: a
Bulletin Board for Unix Users," Byte, 8(10), (October,
1983), p. 239; and Usenet History Archives. [72] Turoff and Hiltz, "Meeting through your Computer," p. 60. [73] Harvey, Interview. [74] Feinler, "Distributed Database," p. 21:3-4; Myer and
Vittal, "Message Technology," p. 21:2-4. [75] Feinler, "Distributed Database," p. 21:3-4. [76] Feinler, "Distributed Database," p. 21:3-4. [77] One paper proposed using the Network Information Center's
ARPANET Directory as a starting point for such a service. [78] Myer and Vittal, "Message Technology," p. 21:2-7. [79] Zakon, "Hobbes' Internet Timeline v2.3." [80] Zakon, "Hobbes' Internet Timeline v2.3." [81] Licklider and Vezza, "Applications," p. 1342. [82] Licklider and Vezza, "Applications," p. 1342. [83] Licklider and Vezza, p. 1342. [84] Licklider and Vezza, p. 1342. [85] Licklider and Vezza, p. 1344.  
---------------------------------------------------------------
                        BIBLIOGRAPHY
--------------------------------------------------------------- 
                       Primary Sources
                               Scholarly Papers and Convention Records
-------------------------------------------------- Donald Austin, "Introduction," Berkeley Workshop on
Distributed Data Management and Computer Networks, (1976),
LBL-5315, Lawrence Berkeley Laboratories, p. iii. Barry Boehm, "A Panel Session - Planning Community
Information Utilities," AFIPS Fall Joint Computer
Conference, (1971), p. 669-673. Elizabeth Feinler, "The Identification Data Base in a
Networking Environment," NTC '77 Conference Record, (New
York, NY, USA: IEEE, 1977), p. 21:3-1. Thomas Marill, "TDA Message System," Berkeley Workshop on
Distributed Data Management and Computer Networks, (1976),
LBL-5315, Lawrence Berkeley Laboratories, p. 143-144. T.H. Myer and D.W. Dodds, "Notes on the Development of
Message Technology," Berkeley Workshop on Distributed Data
Management and Computer Networks, (1976), LBL-5315, Lawrence
Berkeley Laboratories, p. 144-154. T.H. Myer and John Vittal, "Message Technology in the
ARPANET," NTC '77 Conference Record, (New York, NY, USA:
IEEE, 1977), p. 21:2-1. Raymond Panko, "An Introduction to Computers for Human
Communication," NTC '77 Conference Record, (New York, NY,
USA: IEEE, 1977), p. 21:1/1-6. Lawrence Roberts, "Communications and System Architecture,"
Memories, Terminals, and Peripherals, Proceedings of the
1970 IEEE International Computer Group Conference, (New
York, NY, USA: IEEE, 1970), p. 373. Lawrence Roberts and Barry Wessler, "Computer Network
Development to Achieve Resource Sharing," Spring Joint
Computer Conference, (1970), 543-549. Richard Roistacher, "The Virtual Journal," Computer Networks
2 (1978), p. 18-24. Robert Stotz, et al., "SIGMA - An Interactive Message
Service for the Military Message Experiment," National
Computer Conference (1979), p. 839-846. Murray Turoff, "Delphi and its Potential Impact on
Information Systems," Fall Joint Computer Conference (1971),
p. 317-326. Walter Ulrich, "OnTyme: A Computer Message System," NTC '77
Conference Record, (New York, NY, USA: IEEE, 1977), p. 21:5-
1. Jacques Vallee, "The FORUM Project," Computer Networks 1
(1976), p. 39-52. Jacques Vallee and Robert Beebe, "Topics and Notepad: New
Management Tools for the Disseminated Task-Force," NTC '77
Conference Record, (New York, NY, USA: IEEE, 1977), p. 21:4-
1. David Walden, "Experiences in Building, Operating, and Using
the ARPA Network," 2nd USA-Japan Computer Conference
Proceedings, Tokyo, (August, 1975), p. 21-3-1. Steve Walker, "The Future of the ARPA Network," Berkeley
Workshop on Distributed Data Management and Computer
Networks, (1976), LBL-5315, Lawrence Berkeley Laboratories,
p. 16-17. Books
-------- Lawrence Roberts and Barry Wessler, "The ARPA Network,"
Chapter 13, Computer-Communication Networks, Norman
Abramson, Franklin Kuo, ed. (New Jersey: Prentice-Hall,
1973). Murray Turoff and Star Roxanne Hiltz, The Network Nation:
Human Communication via Computer (Massachusetts: Addison-
Wesley, 1978). Journal and Periodical Articles
------------------------------------ Paul Baran, "On Distributed Communications Networks," IEEE
Transactions on Communications Systems COM-12 (1964), p. 1. Sandra Emerson, "Usenet: a Bulletin Board for Unix Users,"
Byte, 8(10), (October, 1983), p. 219-236. Starr Roxanne Hiltz, "Computer Conferencing: Assessing the
Social Impact of a New Communications Medium," Technological
Forecasting and Social Change, 10, (1977), p. 225-238. Valerie Lamont, "New Directions for the Teaching Computer:
Citizen Participation in Community Planning," Technological
Forecasting and Social Change, 5, (1973), p. 145-162 Lawrence Landweber, "CSNET - The Computer Science Research
Network: History, Status, and Future Plans," Computer
Compacts, 1(1), (February, 1983), p. 9-14. J.C.R. Licklider, "Man-Computer Symbiosis," IRE Transactions
on Human Factors in Electronics, 1(1), (March, 1960), p. 4-
11. J.C.R. Licklider and Albert Vezza, "Applications of
Information Networks," Proceedings of the IEEE, 66(11),
(November 1978), p. 1330-1345. Brock N. Meeks, "An Overview of Conferencing Systems," Byte,
(December, 1985), p. 169-184. Lawrence Roberts, "The Evolution of Packet Switching,"
Proceedings of the IEEE, 66(11), (November, 1978), p.1307-
1313. Murray Turoff, "The Design of a Policy Delphi,"
Technological Forecasting and Social Change, 2, (1970), p.
149-154. Murray Turoff, "Delphi Conferencing: Computer-Based
Conferencing with Anonymity," Technological Forecasting and
Social Change, 3, (1972), p. 159-204. Murray Turoff, "An On-Line Intellectual Community or "MEMEX"
Revisited," Technological Forecasting and Social Change, 10,
(1977), p. 401-412. Murray Turoff and Star Roxanne Hiltz, "Meeting through your
Computer," IEEE Spectrum (May 1977), p. 58-64. Murray Turoff and Star Roxanne Hiltz, "The Evolution of User
Behavior in a Computerized Conferencing System,"
Communications of the ACM, 24(11), (November, 1981), p. 739-
751. Jacques Vallee, "Modeling as a Communication Process:
Computer Conferencing Offers New Perspectives,"
Technological Forecasting and Social Change, 10, (1977), p.
391-400. Jacques Vallee, "The User Abusers - Modern Networks in
Perspective," Computer Design, (September, 1982), p. 203-
208. Jacques Vallee, Arthur Hastings, Gerald Askevold, "Remote
Viewing Experiments through Computer Conferencing,"
Proceedings of the IEEE, (October, 1976), p. 1551. Interviews
------------- Brian Harvey, interview by author, tape recording, Berkeley,
California, 21 March 1996. Frank Heart, interview by Judy O'Neill, tape recording,
Cambridge, Massachussets, 23 March 1990, OH 186, Charles
Babbage Institute, University of Minnesota, Minneapolis,
Minnesota, from transcript. J.C.R. Licklider, interview by William Aspray and Arthur
Norberg, tape recording, Cambridge, Massachussets, 28
October 1988, OH 150, Charles Babbage Institute, University
of Minnesota, Minneapolis, Minnesota, from transcript. Lawrence Roberts, interview by Arthur L. Norberg, tape
recording, San Mateo, California, 4 April 1989, OH 159,
Charles Babbage Institute, University of Minnesota,
Minneapolis, Minnesota, from transcript. Ray Tomlinson, personal communication with author, email, 9
April 1996. Ray Tomlinson, personal communication with author, email, 10
April 1996. Internet
---------- "Advanced Research Projects Agency - Mission," ARPA
Homepage, http://www.arpa.mil/. David Crocker, John Vittal, Kenneth Pogran, and D. Austin
Henderson, Jr., "Standard for the Format of ARPA Network
Text Messages(1)," ARPANET Request for Comment 733, 21
November 1977. Robert H'obbes' Zakon, "Hobbes' Internet Timeline v2.3,"
http://info.isoc.org/guest/zakon/Internet/History/HIT.html,
or email timeline@hobbes.mitre.org. Usenet History Archives,
ftp:weber.ucsd.edu/pub/~usenet.hist/. ARPA Memoranda and Project Plans
-------------------------------------------- J.C.R. Licklider, "Memorandum to Members and Affiliates of
the Intergalactic Computer Network," 23 April 1963, RG 330-
69-A-4998, Box 3, Folder: 350-1, National Archives and
Records Administration. J.C.R. Licklider to Dan Bobrow and Bert Sutherland, RG 330-
78-0085, Box 2, Folder: "Networking 1968-1972," National
Archives and Records Administration. ARPA, Program Plan No. 93, "Computer Network and Time-
Sharing Research," 5 April 1963, RG 330-78-0013, Box 1,
Folder: "Program Plans," National Archives and Records
Administration. ARPA, Program Plan No. 95, "Remote Stations and Programs for
Computer Network," 5 April 1963, RG 330-78-9913, Box 1,
Folder: "Program Plans," National Archives and Records
Administration. ARPA, Program Plan No. 723, "Resource Sharing Computer
Networks," 3 June 1968, RG 330-78-0013, Box 1, Folder:
"Program Plans," National Archives and Records
Administration. 
                              
                     Secondary Materials 
Journal and Periodical Articles
------------------------------------ Martin Campbell-Kelly, "Data Communications at the National
Physical Laboratory (1965-1975)," Annals of the History of
Computing, 9(3/4) (1988), p. 221-247. Edwin Diamond, Stephen Bates, Mark Horowitz, "The Ancient
History of the Internet," American Heritage, 46(6) (Oct.,
1995), 34-42. Judy O'Neill, "The Role of ARPA in the Development of the
ARPANET, 1961-1972," IEEE Annals in the History of
Computing, 17(4), 1995, p. 76-81. Maurice Wilkes, "Networks, Email, and Fax," Communications
of the ACM, 33(6), (June, 1990), p. 631. Books
-------- Forster, E.M., "The Machine Stops," in E.M. Forster: The New
Collected Short Stories, (London: Biddles, 1985), p. 108-
141. Howard Rheingold, The Virtual Community, (Massachussets:
Addison-Wesley, 1993). Other
------- Hauben, Ronda, "The Evolution of Usenet News: The Poor
Man's ARPANET," speech presented at the MACUL Conference,
Detroit, Michigan, 12 March 1993.  
---------------------------------------------------------------
Appendex I: ARPANET Timeline from 1957 to 1990
--------------------------------------------------------------- The information for this timeline was gleaned from "Hobbes'
Internet Timeline v2.3" and other sources listed in the
bibliography.  1957        The USSR successfully launches and orbits the
            Sputnik I artificial earth satellite, prompting
            widespread fears of technological inferiority
            in U.S.
1957        J.C.R. Licklider becomes a vice president of
            Bolt Beranek and Newman (BBN).
1958        Advanced Research Projects Agency (ARPA)
            created by the Department of Defense.
1960        Licklider, J.C.R., "Man-Computer Symbiosis"
            mentions connecting computers together by wide-
            band communication lines. March.
1961        Director of Defense Research and Engineering
            (DDR&E) assigns a Command and Control project
            to ARPA.
1962        J.C.R. Licklider joins ARPA as director of its
            Command and Control research office.
1962        The Information Processing Techniques Office
            (IPTO) of ARPA established to coordinate ARPA's
            Command and Control research. Orginally named
            the Command and Control Research Office.
1963        Licklider, memorandum: "to members and
            affiliates of the Intergalactic Computer
            Network." (cf. O'Neill). April 25.
1964        Licklider leaves IPTO in July. His successors:
            first Ivan Sutherland, then Robert Taylor. (cf.
            O'Neill)
1964        Paul Baran, RAND, "On Distributed Networks."
1967        ACM Operating Systems Symposium in Gatlingberg,
            Tennessee: Initial plan for ARPANET unveiled.
            At same conference, paper by Davies, Bartlett,
            and Scantlebury presents the proposed NPL
            network.
1968        Tet Offensive in Vietnam.
1968        ARPA Program Plan No. 723, 3 June. "RESOURCE
            SHARING COMPUTER NETWORKS" Prepared by Lawrence
            G. Roberts.
1969        BBN wins ARPA bid to implement the physical
            ARPANET network.
1969        ARPANET  online. 4 nodes: UCLA, UCSB, Stanford
            Research Institute (SRI), University of Utah.
            First IMP installed September 1 at UCLA.
1969        First request for comment: "Host Software" by
            Steve Crocker.
1969        Tymshare Corporation begins work on Tymnet,
            using non-distributed routing.
1970        Lawrence G. Roberts and Barry D. Wessler,
            "Computer Network Development to Achieve
            Resource Sharing"
1970        Turoff, Murray, "The Design of a Policy Delphi"
1970        National Physical Laboratory (NPL) in Great
            Britain's packet-switched network becomes fully
            operational.
1970        ARPANET nodes begin using network control
            protocol (NCP).
1971        Turoff, "Delphi and its potential impact on
            information systems"
1971        Tymnet described at the Spring Joint Computer
            Conference.
1971        EMISARI Conferencing developed at the Office of
            Emergency Preparedness.
1971        ARPANET Reaches size of 15 nodes.
1971        Ray Tomlinson of BBN sends first network email
            message across the ARPANET using SNDMSG.
1972        ARPA renamed DARPA.
1972        International Conference on Computer
            Communications with first public demo of
            ARPANET organized by Robert Kahn.
1972        Internetwork Working Group (INWG) created to
            establish agreed-upon protocols. Chaired by
            Vinton Cerf.
1972        Telnet specification (RFC 318)
1972        BBN forms Telenet Communications Corporation to
            develop private packet-switched network
            systems.
1972        Turoff, Murray, "Delphi Conferencing: Computer-
            Based Conferencing with Anonymity"
1973        First international connections to the ARPANET
            between University College of London (United
            Kingdom), and the Royal Radar Establishment
            (Norway).
1973        Robert Kahn and Vinton Cerf begin work on
            gateways, internetting.
1973        File transfer specification.
1973        Research with FORUM conferencing on the ARPANET
            begins.
1974        Scrapbook information processing system
            developed for network at NPL by a team led by
            David Yates. Includes data base, word
            processing, and network email.
1974        Telenet introduces the first public packet
            network service.
1975        EIES conferencing system goes online
1976        Vallee, Jacques, Hastings, Arthur C., and
            Askevold, Gerold, "Remote Viewing Experiments
            through Computer Conferencing"
1976        Vallee, Jacques, "The FORUM Project: Network
            Conferencing and its future applications"
1976        uucp (Unix to Unix Copy Protocol) developed at
            AT&T by Mike Lesk. Released in 1977 with UNIX
            version 7. Would eventually be used for
            implementation of Usenet news.
1976        Elizabeth, queen of England sends out her first
            email message.
1976        Hiltz, Starr Roxanne and Turoff, Murray,
            "Metting through your Computer"
1977        Hiltz, Starr Roxanne, "Computer Conferencing:
            Assessing the Social Impact of a New
            Communications Medium."
1977        Vallee, Jacques, "Modeling As a Communication
            Process: Computer Conferencing Offers New
            Perspectives"
1977        Turoff, Murray, "An On-Line Intellectual
            Community or "MEMEX" Revisited"
1977        THEORYNET developed by Lawrence Landweber,
            Richard DeMillo, and Richard Lipton at Univ.
            Wisconsin to provide email facilities for over
            100 computer science researchers.
1977        Mail specification (RFC 733)
1977        Hiltz and Turoff, The Network Nation, textbook
            dealing with "Human Communication via Computer"
            is published.
1978        Licklider and Vezza, "Applications of
            Information Networks"
1978        Roberts, Lawrence G., "The Evolution of Packet
            Switching"
1978        RCPM dial-in Bulletin Board System (BBS)
            started by Ward Christianson and Randy Suess.
1978        First Usenet connection established between
            Duke and University of North Carolina by Tom
            Truscott, Jim Ellis and Steve Bellovin.
1979        First Multi-User Dungeon (MUD), MUD1 by Richard
            Bartle and Roy Trubshaw at Univ. of Essex.
1979        Jim Ellis hands out 5 page "Invitation to a
            General Access UNIX network" at the January
            1980 Usenix Conference in Boulder, Colorado.
            This letter presents Usenet as a medium for
            spreading the Usenix newsletter and journal
            articles.
1980        Software for Usenet version "A" put on tape for
            general distribution at Delaware summer Usenix
            meeting. The phrase "Poor man's ARPANET" is
            used in a handout to describe Usenet.
1980        Hiltz and Turoff, "The Evolution of User
            Behavior in a Computerized Conferencing System"
1981        Simple Mail Transport Protocol (SMTP), RFC 788.
1981        "B" version of Usenet news written by Matt
            Glickman, a high school student, and Mark
            Horton, a grad student at U.C. Berkeley.
1981        BITNET starts.
1981        Minitel deployed across France.
1981        CSNET project funded by NSF.
1981        Vallee, Jacques, "The User Abusers - Modern
            Networks in Perspective"
1982        Phase I of CSNET implementation (email)
            achieved.
1982        TCP/IP protocol suite established. Replaced
            NCP.
1982        EUNet (European Unix Network) established to
            provide email and Usenet services.
1982        Orginal Usenet backbone created by Gene
            Spafford.
1983        CSNET/ARPANET gateway in place.
1983        ARPANET splits into ARPANET and MILNET.
1983        Berkeley releases 4.2BSD incorporating TCP/IP
1983        Fidonet BBS software developed by Tom Jennings.
            Supports the growth of a hobbyist electronic
            mail and conferencing network.
1984        NSF takes over backbone administration from
            ARPA.
1984        Domain name server developed.
1984        Number of hosts on what is now the Internet
            breaks 1000
1984        JUNET (Japan Unix Network) established using
            uucp.
1984        Neuromancer written by William Gibson. In this
            book, Gibson coins the phrase "cyberspace."
1984        Brock N. Meeks, "An Overview of Conferencing
            Systems"
1985        Original NPL network in England replaced with
            updated system.
1985        Whole Earth 'Lectronic Link (WELL) started.
1985        Fidonet echomail developed by Jeff Rush.
1986        The Usenet "Great Renaming" begins. Ends 1987.
1986        IPTO ceases to exist as a discrete office after
            DARPA reorganization. Technical scope of IPTO
            expands and it becomes the Information Science
            and Technology Office (ISTO).
1986        Cleveland Freenet begins.
1986        NNTP (network news transfer protocol) developed
            to enhance Usenet performance using TCP/IP.
1986        Number of hosts on the Internet breaks 10,000.
1987        Merit Network, Inc., IBM, and MCI sign
            agreement to manage the NSFNET backbone.
1987        Harry Spencer of University of Toronto creates
            the "C" version of Usenet news.
1988        Usenet: "Breaking of the Backbone Cabal"
1988        The Morris Internet worm is unleashed.
1988        NSFNET backbone upgraded to T1 (1.54Mbps).
1988        Internet Relay Chat (IRC) developed by Jarkko
            Oikarinen.
1988        First gateway connecting Fidonet and the
            Internet.
1988        Number of Internet hosts breaks 100,000.
1989        First commercial electronic mail carriers
            appear on the Internet: MCI Mail and
            Compuserve.
1989        Electronic Frontier Foundation founded by Mitch
            Kapor.
1990        ARPANET officially decommisioned.   ---------------------------------------------------------------------------
Appendix II: Selected Messages from HUMAN-NETS, 1981.
--------------------------------------------------------------------------- 
These messages were gathered from the Usenet History
Archives, maintained by Bruce Jones at
ftp:weber.ucsd.edu/pub/~usenet.hist/. They are intended to
give a sampling of some topics discussed at one time on the
HUMAN-NETS mailing list. These messages should by not be
interpreted as providing a complete representation of HUMAN-
NETS discussion. 
------------------------------ Date:  6 May 1981 0359-EDT (Wednesday)
From: Gary Feldman at CMU-10A
Subject:  Sociology of Computer Science With respect to the assertion from Human-Nets V3 #92

  ... this is because there are a fair number of gays in
computer
science ... Is there any reason to believe that the percentage of gays
in computer science is different from the percentage in the 
general population (or more properly, in the work force)?  
I can relate this to two observations:

   1. The commentary to the Hacker Papers (Psychology Today,
Aug 80), Weizenbaum (1976), and others observe that many 
people appear to use computer interactions as a substitute for 
human interactions.

   2. Many gays experience a sense of isolation before they
manage to meet other gays, thus giving them a need for the sort 
of interactions mentioned above.  In the language of Transactional 
Analysis, it could be claimed that computers can provide "strokes" 
which are as useful as those provided by people. One case which comes to mind is the young man from Michigan
who committed suicide last year and who made the news with his
disappearance the year before.  He was a computer whiz, was
involved with gay groups, and (as an aside) was a Dungeons and
Dragons player. He was also (according to news accounts) a very 
lonely person. The answer to this question may provide insight into the
general question of what sort of people get involved with 
computers, and why are there so many compulsive computer 
users.  It also relates to the discussion about communication of 
emotion via computers (which also started as a plea against 
loneliness). P. S. wrt stereotypes: some of the most macho football
players on frat row ARE gay. ------------------------------ Date: 05/15/81 01:19:00
From: FFM@MIT-MC
Subject: English Murdering & flame about human
telecommunicating I can't seem to understand all the sighing and moaning about
the "death of the English language". It seems alive and well to
me. All languages have always had slang and argot and various other
unapproved-of features. Every so many years someone gets up
and moans about horrid things that are being done to whatever 
language they happen to worry most about. Some countries, most notably France actually have 'bodies'
dedicated to 'keeping the language pure' which in thier case means
trying to fine people who use the word "hot-dog".  We could go this
route and order that MLA stylesheets be attached to all terminals and
followed scrupulously in all communications. I however do think there is a difference in written and
verbal communications. Cue words like like "Hmmm", "I see" and 
"Ya-know" are common in verbal communications, along with 
things of an outright nosensical nature(if taken literally) like "How 
do you do?". Some of us do not take these literally but realize 
they have symbolic meanings like "Was unaware of that..", 
"I follow you.."  and "we are sort of on
the same base...you kind of agree?? ..".  Verbal
communication needs cues to go along smoothly and to work well, 
if it did not have these things it would be a rather disconcerting 
and distressing. I think/feel that computer communications(done between
humans via computers) lie somewhere between written and verbal
communications in style and flavor. There is an ambience of 
informality and stream-of-conciousness style that pervades it but 
coupled with ideas that are well thought out (usually) and deeper in 
insight than average verbal communications.  Does this make any sense 
to anyone 'sides myself? As far as the medium being used because people really don't
want to communicate in a 'really human way'(read snailmail(??),
phone or in person)....I really wonder about the validity of that
statement.  The most important thing about electronic mail is that it is
asynchornous, if I send you a message at 2am because I had a sudden
brainstorm, you won't be rousted out of bed and wonder if someone 
close is in real trouble, which would happen if you were a day person 
and I called you at 2am. Snailmail is very slow and only really winning if
one does artwork in letterwriting, which I sometimes do... It is however a more 'cowardly' medium in that if I send you
something that might provoke ire, there is little you can do to me
immediately in a physical manner, and the most likely thing that will
happen is you might yell at me in a letter or a send which is much
less bothersome than if you were doing it into my face. And even
if you(good old hypothetical you..)got angry there is a chance
,you and I would be cooled down quite a bit before we met face to
face... However if our hypothetical communications produced some
more pleasenter passions there would be only so far we could go
without needing to see each other in person. However the same
problems apply to snailmail writers, there is only so much you can do in
words... However there are advantages in that not initially seeing a
person in the flesh. There are all sorts of things we attach to
people's appearence and the clothes they wear etc. Sometimes nice to
start relationships with- -out those things being initially the
most important, much vaunted-first- -impressions. I feel that electronic mail is definitely a different medium
than 'written' communications or verbal 'communications', after
lots of thinking/feeling it over I can not see it as a medium that
is by nature tremendously inhuman. It can definitely be used in
inhuman ways or in human ways and it has limits as to what it can do but
I honestly can't see it one way or the other. It can't replace holding
someone in your arms but neither can a letter or a phone call or a
pillow for that matter... Enuff flaming.... Have fun
Sends Steve ------------------------------ Date: 17 May 1981 07:17:18-PDT
From: ARPAVAX.olson at Berkeley
Subject: furthering feldman's cry -- Re: HN(v3)#98

        A study of the demography of computer science would
be interesting.  I have no idea whether the gay population in
the field would be disproportionately high, but see no reason to
suspect so. Most of my gay friends are involved with computers 
somehow, but this could well be because I met most of them at work.

        The suggestion that gays are using computers to get
"strokes" is more interesting, but I doubt that the trend is limited
to gays; it seems instead to be the norm for the programmers I know.
The feeling of control I get (well, sometimes) over the huge beastie in
the corner, as well as a sense of accomplishment and usefulness,
make me enjoy my work.  That's why I'm still at it, and not loading
boxes at Sears.  Most of the people I spoke to about this voiced
similar opinions -- one hard-core hacker told me that he'd rather
work with machines than people, because "machines are more reasonable.
People get bitchy, but machines never do."

        That's really a frightening statement.  The whole
idea of computers is to >improve< human communication, not 
provide a substitute for it (item: this digest).  Assuming that my
friend's feeling is, indeed, the norm, hackers must be generally ill-
prepared for "real life" (as "The Hacker Papers" suggests).  We may
have just met the enemy (and yes, he may be us) -- hackers are, after
all, people, too (after a fashion).  If we as a group continue to
be predominantly machine-oriented, it's going to be ugly the
next time the system crashes.

        So what should >our< role be?  Is it important for
the people who write the programs that make the world go 'round to be
more than just information processors?  How do we make them that way?
Awaiting the reply of any sociologists out there... Mike ------------------------------ Date:  20 May 1981 13:53 cdt
From:  VaughanW at HI-Multics (Bill Vaughan)
Subject:  influencing the language It would be nice if some of the MIT community (who seem at
times to dominate these lists) would recall that there are indeed
members of the lists who do not use such terms as foo, hack, frob, moby
and mung in their everyday language; that those people probably have
their own jargon (many folk seem to use "fred" where an MIT'er would
use "foo"); but particularly that some parts of the MIT jargon (I have
in mind "win" and "lose") are also in the standard language with
different semantics and are therefore likely to be misinterpreted by
the rest of us.
   
(A word to the wise: "to hack" and its derivatives have
strongly different semantics in most of the world than at MIT, such
that one would not willingly admit to being a hacker - especially not
on an interview!  In industry, the practice we call "hackery" is
the antithesis of good design.) ------------------------------ Date: 24 May 1981 0259-CDT
From: Clyde Hoover <CC.CLYDE at UTEXAS-20>
Subject: Inflicting of hacker language upon others

        Thinking about the recent discussion of the
linguistic miscongenation particular to the computer field, and how
other non-hacker types react to it, I get the feeling that the non-
hackers are right in a quizzical and/or negative reaction to such
jargon.

        I work programming computers about 40 hrs/week, and
when I finish with my work day, I really don't want to think about
or talk computers until I go to work the next day (whole weekends
have been known to pass sans a cybernetic thought on my part), unless
of course I run into another computer person (which I do
occasionally).

        This feeling is particullary interesting because I
live in an environment that has its own pecular slang, and do find
myself using that home slang at work, at times to the consternation of my
colleages.

        My feelings about hacker jargon (and all other
vocational vocabularies), is that they should be left at work, and one
should talk normal English to the people around you (especially
spouses and companions).  In short, STOP BEING A HACKER 
when you aren't working and be a regular human being relating with 
other people on their own terms.

        It does take a shifting of the mental gears, but
I've found that I can deal with people MUCH better when I don't 
bombard them with terms they don't understand, and in turn they can 
really relate to you as a PERSON, not as some 'computer jock.'

        One of my colleages stated this dichotomy that I
feel one should maintain in respect to ones's vocation vs the rest of
life very aptly as :
        "With human beings I will communicate;
                with computers I will interact."

        Cheers,
                Clyde Hoover
        (writing this at a bizzare hour of Sunday morning because
         it finally struck my fancy to add my 2c worth). Date: 05/20/81 10:47:54
From: TRB@MIT-MC
Subject: Influencing Language At Worcester Tech, I think our favorite word in the hacker
lexicon had to be "CUSP."  In DEC parlance, or at least TOPS-10
parlance, CUSP was the acronym for Commonly Used System Program.  
CUSPs (usually after having been rewritten by local hackers) were excellent
programs, and CUSP took on the meaning of something that was winning.  
The usage was almost always: "She's a CUSP of a <thing>."  She's a CUSP 
of a car, she was a CUSP of a movie.  Anything good was a CUSP, except
your girlfriend, as you didn't want to call your girlfriend commonly used.
As cusp was already in the dictionary, we confused our share of
dentists, astrologers, and mathematicians... Conversely, the word NUSP (Never etc.) caught on to a lesser
extent, to describe DEC standard software and other losing things.
"The drum is fried again? What a NUSP!"  Ah, the good old days. ------------------------------ Date: 27 May 1981 0209-EDT (Wednesday)
From: Gary Feldman at CMU-10A
Subject:  sociology and psychology of computer use Permit me to carry the doom-crying one step further.  I am
curious whether the increasingly easy access to computers by
adolescents will have any effect, however small, on their social 
development. Keep in mind that the social skills necessary for 
interpersonal relationships are not taught; they are learned by 
experience.  Adolescence is probably the most important time period 
for learning these skills. There are two directions for a cause-effect relationship.
Either people lacking social skills (shy people, etc.) turn to
other pasttimes, or people who do not devote enough time to human
interactions have difficulty learning social skills.  I do
not whether either or both of these alternatives actually occur. I believe I am justified in asking whether computers will
compete with human interactions as a way of spending time?  
Will they compete more effectively than other pasttimes?  
If so, and if we permit computers
to become as ubiquitous as televisions, will computers have
some effect (either positive or negative) on personal development
of future generations? I am not trying to be anti-technology.  In fact, my hunch is
that the answer to the above questions is either no or only slightly.
However, as an ethical computer scientist, I believe in asking these
questions in advance. One aid in answering these questions is to get psychological
profiles of people involved with computers (not necessarily
demographic data). A direct psychological survey would be most 
precise. However, getting indirect data such as gender, marital status, 
membership in fraternities/sororities, etc.  would also be useful, if
properly interpreted.  The only reason for picking on sexual
preference (apart from the unfounded claims that have been made in 
these digests) is the slight correlation between sexual preference and 
other psychological factors. Anyone have other ideas for evaluating the psychology of
using computers?  I would certainly like to see some sound
research efforts in this direction, although I don't for a minute believe
that the economics of research would permit such efforts.  
---------------------------------------------------------------------------
END OF FILE: email_history Please refer to the copyright notice 
at the beginning of this document.
The author may be contacted for
questions, corrections, or comments
at: Ian R. Hardy
hardy@uclink2.berkeley.edu
hardy@mobileweb.com
---------------------------------------------------------------------------