Before the 1980s software was not seen as a commodity but rather the product of collaborative work by academics and corporate researchers. During this time source-code was freely distributed to its users. The source-code of software refers to the text that a computer uses to assemble the program that needs to be run. Having access to the source-code enables the user to modify it and make it compatible with different hardwares and operating systems, as well as fix bugs and add new features. The idea of freely distributing source-code is inline with the core decentralized structure of the Internet. In the 60s and 70s when software development was just getting started it was normal and even encouraged for source-code to be published. User groups for IBM’s first commercial scientific computer, SHARE, and the Digital Equipment Corporation, DECUS, existed to facilitate the exchange of software. Some university computer labs even made it a policy for the programs installed on the computers to come with source-code files 1.
In 1983 IBM made the decision to stop publishing their source-code. 2 At this time IBM was one of the most dominant corporate players in terms of distributing network technologies to users, ie. computer hardware and software and the decision to stop publishing their software’s source-code was met with controversy. In this Computerworld article two IBM employees talk about whether IBM is “playing fair” in making this choice. Martin Gotez argues against this decision in the interest of collaborative software development saying that this decision “...reduces opportunities for companies developing complementary and competitive products”2. Peter Schneider on the other hand says that this decision is beneficial because it protects the company from unauthorized copying. Gotez seems to be viewing the distribution of source-code as conductive of the collaborative values of the academic community that were prominent in computing and networking technology up until this point. Schneider seemed to be focused on how Open Source Software would impact the commercial viability of IBM. The positions illustrated in this article are similar to the discourse around Open Access.
1983 was also the same year that the ARPANET started using TCP/IP protocols under a government mandate.3 Adopting the TCP/IP protocol meant that the network would be able to handle larger amounts of traffic which would allow its user base to expand. Considering this development along with IBM’s decision could potentially mark 1983 as a key year in terms of the commercial potential of networking beginning to be realized.