The Cathedral and the Bazaar
socio-political manifesto of the free software movement
Forget the enigmatic title of The Cathedral and the Bazaar for a moment. This is essentially four long, polemical essays on the open source movement, written by one of its prime movers in the period between 1992 and summer 1999. ‘Open Source’ is a term used to describe the idealistic notion of freely sharing technological development – particularly the software code written by computer programmers. The first and earliest essay sets out the principles of the open source movement. The second inspects the attitudes and moral codes of its members (the hackers) who submit their work to peer review and what Eric Raymond claims is a ‘gift culture’. The third looks at the economic conundrum of how the open source movement sustains itself without a regular income. The last essay is an account of activism relating to the Microsoft anti-trust case.
Basically, it’s an impassioned argument in favour of a new strategy in software development which has arisen from the decision by Linus Torvalds to release the source code of his operating system Linux. He released it not only for free use, but also invited volunteers to help him develop it further. Raymond argues that this represents – dare one say it? – a paradigm shift – a democratic sharing of ideas and repeated testing rather than the development of a product in commercial secrecy.
This is where the title comes in. The ‘cathedral’ is a metaphor for work ‘carefully crafted by individual wizards or small bands of mages working in splendid isolation, with no beta to be released before its time’. The bazaar represents an open free-for-all approach ‘differing agendas and approaches…out of which a coherent and stable system [can] seemingly emerge only by a succession of miracles’.
He inspects the arguments which have been made in criticism of the open source movement, and whilst I wouldn’t say that he demolishes them exactly, he does come up with some interesting points about a system which he is presenting as a revolutionary alternative to the common commercial model. ‘It is often cheaper and more efficient to recruit self-selected volunteers from the Internet than it is to manage buildings full of people who would rather be doing something else’. If the principles of the open source movement really do work in the long term, this will stand a lot of MBA wisdom on its head.
However, his arguments for the advantages of releasing open source on Netscape (in autumn 1998) seem to evade the issue that NS was under intense pressure from Microsoft. He’s making an argument from technological altruism, when deep down the motive might have been economic. But he does explain how a company such as Red Hat can sell open source code (Linux) for a profit, when it’s free for anyone who wants it. They sell – ‘a brand/service/support relationship with people who are freely willing to pay for that’ – and other companies are free to do the same thing if they wish.
As the book reaches its breathy conclusion, the fourth essay becomes a rather personal and excited account of how the open source movement was established in 1998/9 – largely to support Netscape in its fight against Microsoft. No doubt there will be updates to this statement issued at the appropriate web site [www.opensource.org] following each stage of the fight in court.
Some of the anthropological parallels and excursions into political economy seem slightly fanciful, and at times his polemic becomes a sociological study of hackers’ motives – a trap which in literary studies is known as the ‘intentional fallacy’. That is, we shouldn’t judge outcomes on the strength of what we perceive to be the author’s intent. It’s also very idealistic – though the latest edition of WIRED carries an article about open source warriors selling their services on the open market, and Raymond argues that there is no necessary contradiction in this.
It’s the first book on high-tech developments I’ve come across which provided the slightly bizarre experience of a text printed with double line spacing and one-sentence paragraphs. This I imagine reflects the influence of the email originals written for reading on screen. Another interesting feature is that the majority of the bibliographical references are to articles on the Net, not to printed books – though I still think he should have tried to produce an index and bibliography.
He claims that even this book is in a state of evolution via updates following peer review – and that’s exactly as it should be for such a subject. It’s written in a concise, deeply compacted style, with few concessions to an average reader’s technical knowledge, and he’s occasionally cryptic to the point of obscurity: ‘Before taxonomising open-source business models, we should deal with exclusion payoffs in general’.
This is a crusading text, and anyone concerned with the sharp end of software development and the battles of operating systems will be fascinated by his arguments. This revised and expanded paperback edition includes new material on recent technological developments which has made it one of the essential texts on Open Sources
© Roy Johnson 2000
Eric S. Raymond, The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary, Sebastopol (CA): O’Reilly, 1999, pp.268, ISBN: 0596001088
If you found this article interesting you might want to Tell-A-Friend.