History of OSS
The Open Source Software movement both created and was born from the origins of the
Internet. The first section of this writeup will detail the history of OSS, looking at deep roots in the origin of the Internet. The second section deals with characteristics of open source, and the third with connections between open source and the digital divide.
Early days of the Internet
In many ways, the birth of the OSS movement is tied inextricably to the birth of the computer network, starting with
ARPAnet. It was this type of communication tool that allowed mathematicians (the original programmers) and
computer scientists to share
information in a way beyond “distinct from” the
postal service.
In the time period in which the foundations for what would become the Internet were laid, software was shared freely. There were a few reasons why this was possible. One was that software existed solely in an academic realm, with support from defense budgets. This was not inevitable “a random happening”, but a design decision of J.C.R Licklider at the Advance Research Projects Administration (ARPA) Information Processing Techniques Office (IPTO). He wanted the computer network developed in the culture of academia, rather than in defense, to avoid some of the bureaucracies of the military-industrial complex. Through the auspices of Licklider, the IPTO funded such endeavors as Project MAC, which provided research funding to MIT which laid ground for the time-sharing computing explosion in that area. Another fateful project funded by the IPTO was the Augmentation Research Center (ARC) at the Stanford Research Institute (SRI). ARC, head by Douglas Engelbart, created several revolutionary computing ideas such as electronic mail, word processing, the mouse as an input device, windowing environments and hypertext.
An entire history of this area is beyond the scope of this writeup, and is covered in other texts . The importance of this history to the topic at hand is the birth of computer mediated communication software out of the university setting, where a long history of academic sharing meant that software was traded between computing centers freely. Software was seen as the result of research, as well as tools for research, and both were traditionally shared openly in academia.
It was academic research, funded with public money, that created the free, open tools that became the Internet. Vint Cerf and Bob Kahn were hired to help create the first Internet Protocols. Bob Metcalfe was funded by ARPA when he came up with Ethernet. All protocols and software was carefully stored and made available. The code was included in RFC’s (Request For Comments) which were documents available on the Net to collect bug findings, patches and so forth.
This openness was crucial for early adoption of ARPAnet and the extension of early computer networks. By openly sharing protocols, software and standards, universities could be saved from the cost of developing these layers each time they wanted to be part of the network. That cost savings lowered the barrier to entry for universities to participate in the growing computer network.
UNIX and OSS
UNIX also has a role in this early history of open source software. UNIX was developed in the early 1970’s using the
C language, which was being developed in parallel to the
operating system. It was the first time a computer OS was not tied specifically to the
hardware it would run on. UNIX, like all software at the time, was freely available.
However, the original UNIX, which was developed at Berkeley, soon splintered as hardware manufacturers developed proprietary versions of UNIX to increase customer lock-in. It was this closing that would eventually lead to the need of OSS alternatives like GNU, BSD and Linux.
Rainy Days: The Closing of the Internet
In the late 1980’s, the
Department of Defense took responsibility of the ARPAnet when the
National Science Foundation no longer wanted it. Defense workers moved to using the
tools developed on the ARPAnet, like email and
ftp, but soon found that the “open” attitude of the academic researchers did not rest well with the
secrecy needs of a military organization. Academic researchers, for their part, did not appreciate that their free territory suddenly had fences installed by relative newcomers.
To address these issues, ARPAnet was split off and reserved for defense use only, and the academic portions were renamed “the Internet”. However, this is also the time that the government decide to extend membership of the Internet to the private sector. Private companies had always had a role in the Internet, but now the scale was vastly increased. The private sector, like the military sector, soon found that absolute openness did not suit their needs like it did for academics. Information soon became closed, and for-profit software shops claimed copyright and patent protection for their materials. Actually, this was not a new concept in general. Similar protection of information had been done by Microsoft and AT&T in the 1970’s, but the Net had remained largely free of such controls. This was changing.
In the early 1980’s, in the
cauldron of splintered UNIX licenses, NSF moving out of the ARPAnet, and rising software companies, Richard Stallman appeared. A programmer who had worked on many of the pieces of software on the early Net instantiations while at MIT’s
AI lab, Stallman recollects being shocked the first time he asked for source code on the Net and was refused. This was a significant shift in the 15 year history of
network development, and was a significant moment for the programmer.
In reaction to this new attitude towards software, Stallman established the Free Software Foundation (FSF). The term “free” here is further defined by the axiomatic “Free as in speech, not as in beer.” The FSF was the first cohesive voice for those programmers who had made a habit of creating software that they made publicly available, and was seen as a necessary response to the voices of private companies claiming copyright protections.
The FSF created several tools important to the open source movement, including Emacs, a text editing program, but of particular import here is the GNU Public License (GPL), also known as copyleft. Open source licenses are covered in more detail below, but the GPL had several significant pieces that were contrary to common copyright laws. For example, in the GPL:
- Software under the GPL cannot be “closed” i.e. you cannot take the code and claim copyright on it.
- Any software using a portion of GPL software is also under the requirements of the GPL.
- The author of the software must be attributed at all times.
Stallman and the FSF remain active and important in the open source world, though many main stream open source developers consider them too anti-commercial and zealous about their mission. Programmers revolting against the strict openness of the GPL have created alternative licenses that are not so demanding about future development.
In 1991, a new model of
shared software development, different than the model pushed by the FSF, arose from the work of a grad student in
Finland. A message went out on the
comp.os.minix newsgroup on October 5, 1991:
Do you pine for the nice days of Minix-1.1, when men were men and wrote their own device drivers? Are you without a nice project and are just dying to cut your teeth on an OS you can try to modify for your needs? Are you finding it frustrating when everything works on Minix? No more all-nighters to get a nifty program working? Then this post might be just for you.
As I mentioned a month ago, I’m working on a free version of a Minix-look-alike for AT-386 computers. It has finally reached a stage where it’s even usable (though may not be, depending on what you want), and I am willing to put out the sources for wider distribution … This is a program for hackers by a hacker. I’ve enjoyed doing it, and somebody might enjoy looking at it and even modifying it for their own needs. It is still small enough to understand, use and modify, and I am looking forward to any comments you might have. I’m also interested in hearing from anybody who has written any of the utilities/library functions for minix. If your efforts are freely distributable (under copyright or even public domain), I’d like to hear from you, so I can add them to the system.
This innocuous message was to receive quite the response. Linus Torvalds initially called his UNIX derivative “Freax”, as in Free+Unix, but was persuaded to change it to Linux. Of the first ten people who downloaded Linux, five sent back code to improve the software .
Soon, a new method of developing software projects developed, mostly encapsulated by this original message from Torvalds. People would send him, and soon a few trusted colleagues, code to improve Linux, and Torvalds would decide whether or not to add that code to the Linux kernel. Release 1.0 was made available in 1994. Currently, close to ten years after this original message, there are more than three million lines of code in the software, donated by hackers, and between 4 and 7 million users .
Even for open source projects, Linux was something new. This kind of distributed, open development process stood in stark contrast to the closed, secretive development processes of commercial software shops, and even university labs. It was one thing to build something and allow the code to be tinkered with, but quite another to build something major, like an operating system, using the efforts of hundreds of developers all over the world.
This new paradigm for software production attracted the attention of Eric S. Raymond. Raymond was the developer of an open source program called Sendmail, and had long been interested in the culture of the hackers. He wrote several influential pieces drawing together the ethos of this community, most notably “The Hacker’s Dictionary” . Raymond observed the success of the Linux developers in creating a cheap, portable, stable version of UNIX, and codified what we saw as the differences in development between this and traditional, closed shop types of development. These observations were codified in “The Cathedral and the Bazaar” , which remains an important introduction to how open source projects develop and operate.
Almost paradoxically,
Netscape has helped and hindered the OSS movement. The first Web browser,
Mosaic, was developed at
University of Illinois Champaign-Urbana at the
National Center for Supercomputing Applications (
NCSA). It was government endorsed, and open. Netscape, founded in part by people who had worked on developing the Mosaic browser, set out to deliberately create an alternative to Mosaic, which they named
Mozilla (
Mosaic-Killer).
Not only did they manage to create this closed version of an open browser, but Netscape then ignored the University of Illinois Champaign-Urbana’s demands for royalties and flooded the Internet with free versions of their software. Effectively, they undercut university control of an open browser standard and replaced it with something proprietary and closed. The U.S. government could have pursued legal action but did not. Champaign-Urbana, lacking support from the U.S. Prosecutor’s office dropped their suit as well, and Netscape succeeded in commercializing the Web.
In January of 1998, Netscape announced they were going to open the source code for the new Netscape browser software, and invited the open source community to participate in its development. Pressure from Microsoft’s Internet Explorer had the Netscape management looking for a way to compete against the software giant, and quoted “The Cathedral and the Bazaar” as inspiration for the decision.
Open Source In 1998
Netscape’s decision to open its source code caused a significant chain of events. The decision was announced in January 1998. In February 1998, Eric Raymond consulted with Netscape on how best to organize their efforts. It was there that he decided that the free software movement was in danger of obsolescence, or at least serious delay, if it continued to distance itself from the commercial sector. The term “
free” with its mixed connotations was hobbling the movement.
A meeting of leaders of the Free Software movement on March 7, 1998 convened to discuss the future of the movement. They determined, with the exception of Stallman, that free software as a term had to be abandoned, and the term “open source” was recommended instead.
At the same meeting, a program of memetic engineering was also established. There were already some infrastructure pieces in place, like Slashdot, Freshmeat and Sourceforge, but the newly crafted Open Source Initiative wanted to extend the conversation into the public sphere. By the end of April 1998, articles on “Open Source” appeared in the Economist, the New York Times and Salon, as well as many of the technical journals.
A watershed moment for the new Open Source movement happened in October of 1998 when Raymond released “The Halloween Documents” which were internal Microsoft memoranda outlining the potential threat of open source to the software giant. Open source had arrived.
Ethics of OSS
Open Source vs. Free Software
The
Open Source Initiative and the
Free Software Foundation have an
ambivalent relationship as they both act as proponents of similar ideas. Richard Stallman of the FSF feels that the term “Open Source” betrays the nature of the free software movement.