By the second year of operation [âŠ] an odd fact became clear. ARPANETâs users had warped the computer-sharing network into a dedicated, high-speed, federally subsidised electronic post-office. [âŠ] Not only were they using ARPANET for person-to-person communication, but they were very enthusiastic about this particular serviceâfar more enthusiastic than they were about long-distance computation.
By 1973, email accounted for three-quarters of all ARPANET traffic (Hafner and Lyon 1996, p. 194). While the ARPANET sharing imaginary at the time of development focused on resource efficiency, the practices of users exerted a profound paradigm shift that was not anticipated. Les Earnest, a computer scientist at Stanford University, notes,
I was surprised at the way the use of email took off, but so were the others who helped initiate that development ⊠We thought of [the ARPANET] as a system for resource sharing and expected that remote login and file transfers would be the primary uses.
(cited in Abbate 1999, p. 232, endnotes)
Instead, sharing via email and mailing list was used to develop a sense of community and relational ties between ARPANET users. Databases contained lists of topics such as science fiction, which allowed individuals to not only communicate with large groups but also to develop a community identity among geographically disparate people. Increasingly, the resources of the ARPANET were seen to be the community of users rather than the use of remote devices or access to specific programs.
This social dimension of email also drew on existing perceptions of social hierarchies of power. Stephen Lukasik, who directed ARPA from 1971 to 1975, was himself an early advocator of email, and this influenced the practices of those working under him. Program managers perceived that they could develop closer ties with Lukasik by using email and could gain certain advantages in the workplace, such as budgetary rewards. Consequently at âARPAâs headquarters, the appeal of the network had nothing to do with computers and everything to do with access to powerâ (Abbate 1999, p. 108).
Even in the early 1970s, illicit file and software sharing as a regular practice in the ARPANET. Earnest recalls the development of file-sharing practices:
[A] thing that happened a lot in the 1970s was benign theft of software. We didnât protect our files and found that both programs and data migrated around the net rather quickly, to the benefit of all. For example, I brought the first spelling checker into existence around 1966 but it wasnât picked up by anyone else, whereas the improved version (around 1971) quickly spread via ARPANET throughout the world.
(pers. comm cited in Abbate 1999, p. 101)
Initial sharing of software took place within the closed communities of those who had access to ARPANET. In the early 1970s, computer researchers did not imagine many other people were using their network. Sharing software was a way of developing relationships within their (limited) community with expectations of mutual participation and reciprocity. The files and software shared were perceived to be part of the commons of the community.
During this pre-internet period, the network was still relatively inaccessible to those outside of the pioneering research communities. Built by technologists and researchers, it was also built for technologists and researchers and relied on comprehension of access commands and data ranges (Baym 2010). Although the network structure reflected intentions to promote resource sharing through âopen architecture networkingâ (Leiner et al. 2009, p. 24), access to such structures were limited.
The first significant areas for popular networked sharing were Bulletin Board Systems (BBSes) and USENET. BBSes provided early public access to networked communities. The first public BBSes were established in 1978 by computer hobbyists Ward Christensen and Randy Suess, who met through the Chicago Area Computer Hobbyistsâ Exchange (CACHE). Users accessed BBSes over a phone line using a modem, meaning they no longer required access to a mainframe computer to participate. BBSes functioned as repositories of files and software and allowed communities of users to read notices and exchange messages similar to contemporary internet forums.
USENETâestablished in 1979 as a âpoor manâs ARPANETâ (Quarter-man 1990, p. 243)âused the protocols of mailing lists to post messages to discussion networks known as newsgroups (Hauben and Hauben 1996). USENET provided a means of linking those shut out of the ARPANET community. Unlike BBSes, USENET was decentralised, meaning it did not have a centralised server and did not require a dedicated administrator. Any user could create a newsgroup and users could select which newsgroups they wished to participate in. The functionality of sharing computation had been further adapted for communication:
Computer users flocked to USENET because it offered new possibilities for social interaction, bringing together âcommunities of interestâ whose members might be geographically dispersed and allowing people to participate anonymously if they chose.
(Abbate 1999, p. 201)
Systems for sharing such as BBSes and USENET were developed by users for users, based on existing networking protocols, and it was expected that they would manifest characteristics in keeping with the attributes of community: shared interests, ongoing participation, common resources, reciprocity and shared language (Whittaker, Isaacs and OâDay 1997, p. 137)...