Buy This Book Or We'll Kill This Software Paradigm.

Actually, according to Eric Raymond, author of The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary (O'Reilly & Associates, www.oreilly.com), the proprietary, closed-source model of software development is probably doomed anyway in the long run, except in fairly narrow niches. But you should buy the book anyway: it's a brilliant exposition of the hacker culture and how its approach to software is transforming the software industry.

The book is so full of brilliant insights and apothegms that I used up almost an entire sheaf of Post-its while reading it and this brief review can do no more than whet your appetite for perhaps the most interesting book about software you'll ever read. To my mind, one of the central insights is Raymond's description of the hacker culture (hacker, not cracker) as a gift culture operating in a Lockean property space he calls the noosphere and how this cultural structure makes it possible for open-source projects to overcome Brook's Law.

Frederick P. Brooks, Jr. is the author of the seminal work The Mythical Man-Month, originally published in 1975, in which he observed that the performance of programming teams does not scale in a linear fashion. In fact, if you throw additional programmers at a project that is late, you are only likely to make it more late. According to Brooks, the only way to speed up a late project is to remove features. Case in point: Windows 2000. This is because the work done by N programmers is proportional to N, while the bugs (unwanted interactions between programmers' pieces of the project) scale as N-squared.

So what about Linux, the effort of thousands of programmers around the world? If Brooks is right, Linux should never have happened or, if it did, it would have taken so long that the developers would have to seriously think about the Y3K problem. Instead, Linux has developed into an enterprise-strength OS in record time and goes on from strength to strength at a pace that proprietary operating systems cannot match.

According to Raymond, this is the result of the nature of the hacker gift culture. A gift culture is one in which reputation and reward are based on what one gives away in contract to a command culture (rewards granted by autocracy) or an exchange culture (such as capitalism). Perhaps the most familiar example of a gift culture is that of the native Americans of the Pacific Northwest such as the Kwakiutl and their famous potlatch ceremonies. The largest of these featured the total destruction of villages and the dispersion of great amounts of wealth to establish beyond doubt the reputation of the chief as a "great man." (If you've ever experienced a Seattle winter, you'll understand just what a demonstration of insouciance this represented.)

Gift cultures are only possible in situations where the necessities of life are easily obtained; in the Pacific Northwest of the time, it was said that you could walk across the rivers on the backs of the salmon during spawning season. Likewise, Raymond sees the hacker culture as existing on top of the successful free market exchange culture. Hackers gain reputation by the code they develop and contribute to projects; naturally, only open-source code can be used to demonstrate one's hacking ability. The space in which they operate exhibits a Lockean understanding of property rights: they are established either by starting a new software project (i.e., homesteading never-before exploited land); taking over a project when the originator decides to move on or can't support it anymore (transfer of title); or taking over an abandoned project (adverse possession).

Raymond's hypothesis is that the structure of hacker culture has evolved as the optimal social organization for generating quality software. He writes: "The verdict of history seems to be that free-market capitalism is the globally optimal way to cooperate for economic efficiency; perhaps, in a similar way, the reputation-game gift culture is the globally optimal way to cooperate for generating (and checking) high-quality creative work."

That should give you the flavor of this very enjoyable book; but I can't resist adding a couple of the apothegms that stood out to me.

"Given enough eyeballs, all bugs are shallow." This is the raison d'etre of the "release early and often" practice of open source software: more users early in the development cycle means more bugs found and squashed. How is this different from the release 1.0 horrors of commercial software? You'll have to read the book and, finally, in what may serve as the epitaph of proprietary, closed-source software in the business world, Raymond notes that: "The brutal truth is this: when your key business processes are executed by opaque blocks of bits that you can't even see inside (let alone modify) you have lost control of your business [emphasis in original]. You need your supplier more than your supplier needs you--and you will pay, and pay, and pay again for that power imbalance." He adds that: "The logic is compelling: closed source code is an unacceptable strategic business risk" and states that, eventually, he expects the choice of closed-source code in the face of an open-source alternative to be viewed as f iduciary irresponsibility. Ask not for whom the bell tolls, Redmond; it tolls for thee.