Back to article
Book Review: "The Business and Economics of Linux and Open Source"
Reaching for the Great Linux Business Book
October 24, 2002
For all the talk in the Linux community of converting the "suits," and for all the books that have been written about the open source phenomenon, the great Linux business book has yet to be written. Most open source books are either celebrity profiles, Like Sam Williams' profile of Richard Stallman, "Free as in Freedom," or else they're theoretical treatises for nerds, like O'Reilly's "Open Sources."
And the really strange thing about this is the fact that, as anyone who attended LinuxWorld last August knows, the suits are genuinely interested in this stuff: they want to know what this open source phenomenon is about, what everyone else is doing with it, and--most importantly--how it can save them money.
Enter HP Linux Systems Division General Manager Martin Fink, who takes his stab at the great Linux business book with, "The Business and Economics of Linux and Open Source," (Prentice Hall pp. 242).
Fink, who cut his teeth managing HP's relationship with the Puffin Group--the Ottawa hackers who did the Linux port to PA-RISC--has been involved in the Linux community since 1998, and as the GM behind HP's Linux effort, he clearly understands free software from both the big business and the hacker's perspective. Fink was, after all, the guy who hired Bruce Perens.
And he uses these multiple vantage points in an attempt to create that Holy Grail of Linux business books: a document that any self-respecting geek can plop on his boss's desk, saying, "Read this. It'll save you money," all the while thinking, "If he gets it, it's going to make my life way easier.
Other books have gone for the Grail. Donald Rosenberg's "Open Source: The Unauthorized White Papers," and Russell Pavlicek's "Embracing Insanity" are two of them. And Fink covers a lot of the same ground as his predecessors. There are the obligatory chapters laundry listing the different open source projects, Linux distributions, and software licenses--all very well written.
Fink also provides an excellent explanation of the kernel development process. In fact, if any book does a better job of explaining the dynamics of Linux kernel development to a generalist audience, I have yet to read it. For example, one of the most common questions for companies looking to get involved in Linux is, "How can I get involved and have an influence on the Linux feature set without being shunned as a troublesome outsider?" Fink has the answer: "The best way to influence the Linux feature set, accelerate the release of a new Linux kernel, or see a "roadmap" is very simple," he writes, "get involved. The Linux development community operates under a code of respect and relationships." To people involved in open source, this may seem obvious, but to those who operate under a code of paychecks and progress reports it's easy to miss.
The most exciting part of the book is Chapter Nine: "the Corporate Bazaar," where Fink bravely attempts to explain how the open source development methodology as described in Eric Raymond's "The Cathedral & the Bazaar" might be applied to corporate development projects. I had to re-read the sentence where Fink promises to explain how software project managers could "deliver more product with more functionality, and a higher degree of quality" by using open source methods for internal projects, just to make sure I had read such an audacious claim correctly. I had.
Amazingly, Fink delivers on his promise and goes on to lay out the org chart and discuss the job responsibilities and product roll-out of an open source-style software development project. Interesting stuff.
Fink told me that people who've read the book have given him two "polarized responses" to this chapter, which is based on HP's experiences working with open source projects like Samba and Apache. "One side of it was 'man, what kind of drugs did you wake up on this morning?'," he said, "and the other response was, 'Hey this is really cool; how do we go do it?'."
But while Chapter Nine and a later chapter on how to adjust your business processes to accommodate open source are a must read for those in the tech industry, it may seem too radical for those in corporate IT. This leads to my major complaint with the book. It tries to address both corporate IT managers looking to use Linux and open source software and product managers looking to improve their offerings by making them more open source friendly.
In the end, it better serves the second audience, and I think corporate IT managers looking to learn what open source can offer their businesses may be disappointed. For example, "The Business and Economics of Linux and Open Source" doesn't even talk about the security of open source software. And while there is an excellent chapter on evaluating the real costs of open source software Fink doesn't give any real world examples of what people are actually doing with Linux.
"The Business and Economics of Linux and Open Source" is published as one of Prentice Hall's "HP Strategic Books series" and though Fink does a noble job of protecting the book from the tarnish of the HP marketing machine, the book is clearly at least influenced by the HP way.
For example, it neglects to mention the Ultrasparc in its list of Linux processor architectures, incorrectly lists HP employee Jeremy Allison as the leader of the Samba project (Allison is co-leader, along with Samba founder Andrew Tridgell, who--ever the easy-going Aussie--told LinuxPlanet he "didn't mind" the omission), and claims that HP employee David Mosberger is "one of the few individuals on the short list to succeed Linus," which came as news to both me and David Mosberger.
And I can't help but think that the chapter on Linux standards might have at least mentioned the role Red Hat is playing in the US as a de facto Linux standard had Fink been without industry affiliations.
But on the positive side, Fink says that HP has purchased a number of copies of his book and is giving them away to its customers. They can be sure that they'll be getting a far superior introduction to Linux and open source than any marketing department could dream up.