4.7 The rise of Open Source into the main stream

As we have seen it was normal to share code in the infancy of computing , but very few people used computers. Computers were an arcane subject. In the late seventies computers started to come into more widespread use, and it became possible to earn money on software. This created conflict between sharing code and keeping it secret, so you could sell it. This is what Stallman experienced.

Bill Gate’s open letter to hobbyist, written at the time, said that the hobby marked lacked good software, and went on to state “Who can afford to do professional work nothing?”. As the use of computers grew, so did proprietary software. The “hobby” marked which Microsoft was founded on, would eventually surpass the mainframe and mini computer marked. Proprietary software companies were successful in distributing it’s software to the new masses that entered computing during the eighties and nineties.

Stallman had taken the discussion about software into the realm of ethics. He preached that proprietary software was morally wrong. Free software is a question of freedom, you should have freedom to make changes to a program and give it to your neighbours. FSF’s perceived hostility to intellectual property and slightly communist sounding ideology did not sit well with the business world, especially in the US where communism had been seen as the most dangerous threat by the government until the late eighties. The meaning of the word “free” have two meaning in English; free as in “libre” and free as in “gratis”. How can you earn money on something that is “gratis”?

In the decade since launching the GNU project, Stallman had built a reputation as an excellent programmer. He had also developed a reputation for being non-compromising both in terms of software design and people management. Ian Murdock, founder of Debian, and Eric Raymond had distanced themselves from FSF because of Stallman’s “micro-management” style. Eric Raymond had up until 1992 contributed significantly to GNU Emacs, but distanced himself for the same reason. In 1996 FSF experienced a full-scale staff defection, blamed in large parts on Stallman.

Brian Youmans, a current FSF staffer hired by Salus in the wake of the resignations, recalls the scene: "At one point, Peter [Salus] was the only staff member working in the office."

(Williams 2002)

Hackers like Linus Torvalds and others in the Linux crowd did not share this moral view on software. Torvalds do not find proprietary software morally wrong. He started Linux development for the fun of it, not to fight proprietary software. In an interview Torvalds said:

I’m generally a very pragmatic person: that which works, works. When it comes to software, I _much_ prefer free software, because I have very seldom seen a program that has worked well enough for my needs, and having sources available can be a life-saver.

(Yamagata 1997)

There were growing dissent in the free software community towards FSF’s strongly moral stance. There were a large contingent of developers within the free software community having a more pragmatic view on software, not only Linus Torvalds. The free BSD Unixes were released under a license with minimal restriction, permitting proprietary derivatives from the code base. This crowd believes in the freedom to build great software with minimal restrictions.

The previously mentioned Eric Raymond encountered Linux in late 1993 and what he saw came as a shook to him. He assumed that hacker amateurs could not muster the resources to produce a multitasking operating system. He involved himself in the kernel development, and pondered upon what made Linux development work so well.

Brook’s Law predicts that as the number of programmers N rises, work performed scales as N, but complexity and vulnerability to bugs rises at N2. So if a project is delayed it will only be more delayed if you add programmers. Raymond was determined to discover how the Linux community had avoided the N2 effect.

After three years of participation he developed a theory. He tested the theory on the procmail project. Based on his experience with Linux and procmail he wrote The Cathedral and the Bazaar. In 1997 he presented this essay on a Linux congress in Germany and a Perl conference in USA, with standing ovation from the audience. The news about this essay spread like fire on the net.

Netscape, a pioneering company in web technology, had been targeted for destruction by Microsoft. Microsoft used it’s near monopoly of the desktop by including Internet Explorer in it’s operating system offerings. Netscape feared that if Internet Explorer achieved marked dominance, Microsoft would be able to bend the web protocols away from open standards and into Microsoft’s own proprietary standards, which only Microsoft’s servers could serve.

In January 1998 Netscape announced that it would release the source of the Netscape browser. The content of The Cathedral and the Bazaar was a major influence in this decision. It took several month after the announcement before the code was released. The code for the Netscape browser was a hard to understand patchwork, with inadequate documentation, developed under tight business deadlines. The code badly needed reorganisation, which took several months.

Eric Raymond saw this as an demonstration case for the principles he laid out in The Cathedral and the Bazaar. He hoped that if the source code release was successful the hacker culture, and thereby free software, would raise out of it’s ghetto and into the mainstream, if not he feared that it would confirm business manager’s assumption that free software were not commercially viable.

Raymond offered his help to Netscape in developing the license for the software and working out a strategy. While he was meeting with Netscape he also met a number of key people in the free software community, like Linus Torvalds and Bruce Perens. During this meetings a strategy for getting free software into the mainstream was made. In February 1998 Eric Raymond and Bruce Perens founded the Open Source Initiative (OSI). Deciding on using the term open source to avoid the “libre” vs “gratis” confusion associated with the term “free”, Raymond and his supporters set out on a marketing campaign outlined in his book (Raymond 2001).

I March Tim O’Reilly, founder of a publishing company specialising in software related books and an early supporter of Raymond’s initiative, gathered a number of key developers living at the west-coast in the US. This was set in place to get support for OSI. The invitation list included Torvalds, Larry Wall (creator of Perl), Eric Allman (creator of sendmail) and Paul Vixie (creator of Berkley Interned Naming Daemon (BIND), the most used DNS server.). This was later called The Free Software Summit. In this meeting it was agreed upon to use the term open source with a 9 to 15 vote, according to O’Reilly (Williams 2002). Stallman was not invited to this meeting which would create controversy later.

The OSI marketing campaign can be said to have been a success. It created a lot of attention in the US media. Other companies began to announce that it would involve themselves in open source. Oracle, a big database vendor, said it would port it’s database to Linux. IBM involved themselves in the Apache project. Apache is the web server most frequently used on the net. IBM used Apache in it’s WebSphere product and contributed back to Apache, even if the Apache License did not require it.

The open sourcing of Netscape did not become a success, but it spun of the Mozilla Foundation. The Netscape source was still too hard to understand and had to many interlocking dependencies. It was decided to build a new layout engine for the browser from scratch. The layout engine turns HTML into what is rendered in the browser. This engine was built using a Component Object Model (COM) framework and given the name Gecko. Gecko made the development of the web browser highly modular. The Mozilla Foundation have produced the Firefox web browser and Thunderbird e-mail client, which is increasing in popularity. The open source initiative gained so much momentum that the early failures of Netscape to open source it’s browser did not matter.

Bruce Perens resigned from OSI after one year regretting that OSI had positioned itself in opposition to FSF. In a e-mail from February 1999 he wrote:

Most hackers know that Free Software and Open Source are just two words for the same thing. Unfortunately, though, Open Source has de-emphasized the importance of the freedoms involved in Free Software. It’s time for us to fix that. We must make it clear to the world that those freedoms are still important, and that software such as Linux would not be around without them.

(Perens 1999)

Stallman considered briefly to adopt the term open source, but concluded that: Open source, while helpful in communicating the technical advantages of free software, also encouraged speakers to soft-pedal the issue of software freedom.

The pragmatic view represented by OSI, emphasises the technical and efficiency advantages and represents the Linux community as an example of an efficient method. The moral view represented by FSF, emphasises freedom and the right to change and distribute software. This differences have not been settled, so the term FLOSS have been invented to include both the free software and the open source crowd. This is the term I will be using in this paper.