Linux, Linux: Enough Already

Okay, I’m willing to believe that Linux is breaking out of its stronghold in dorm rooms around the world and actually getting some use in enterprises. I’m even willing to believe that it might get popular in some of those enterprises. But what I can’t believe is that Linux is any kind of serious threat to the growth of Windows NT.

Here’s why. Ultimately, despite the quasi-religious fervor of its supporters, Linux is fundamentally another Unix. Why should it be any more of a threat to NT than is, say, Solaris? One potential answer, obviously, is that Linux boxes are cheaper than Solaris systems. Linux runs on commodity hardware -- hardware that was commoditized by the popularity of Microsoft systems, which is a nice irony -- and the OS is free. But for enterprises, unlike college students, the price of the system itself is a small part of the longer-term cost. Ultimately, Linux systems are only a little bit cheaper, and for an enterprise, who cares? I don’t know any serious organization that, to save a small amount of money, is willing to bet its future on an operating system written and maintained by a bunch of self-selected volunteers.

I don’t doubt the dedication or the talent of those volunteers. What I, on occasion, do doubt is their sanity. They are producing a valuable thing, and they’re just giving it away. This is certainly their right, but what troubles me is that someone, somewhere, is absorbing the value that’s being created. I have something close to an ethical problem here: These enthusiastic Linux developers seem in some ways exploited, in that someone else is deriving the economic value produced by their work.

It’s hard to imagine a happier bunch of exploited workers, which brings us to the real reason for the Linux phenomenon. These developers, like lots and lots of other people, hate Microsoft. There’s no better way to express this today, and even to feel like you’re taking action against Bill Inc., than by joining the Linux movement. I participated in GartnerGroup’s Windows NT conference recently, and the hottest topic at the event was Linux. Red Hat’s CEO attended, too, and people were lining up to have their picture taken with him. The man is a folk hero, as, of course, is Linus Torvalds.

Linux supporters sometimes base their predictions of a glowing future on claims of Linux’s technical superiority over NT in some key areas. Even if these claims are true, which is certainly debatable, they’re irrelevant. As long as two technologies are each good enough to solve the relevant business problems, which one is technically better will have no effect on which gets more widely used. Who wins depends on support, marketing, availability of applications and various other characteristics that are largely independent of the technology itself. MS-DOS didn’t crush OS/2 because it was a better technology, but because it offered a better total solution for the relevant business problems.

My advice is to use Linux where it makes sense, in places where a cheap Unix box is a good fit. For organizations without much Unix installed today, there are few situations where this may apply. If competent Unix administrators are already on hand, Linux boxes can be cheap, reliable platforms for running DNS servers, Apache Web servers and other kinds of single-purpose applications. To even consider Linux, though, an organization must already be Unix-friendly. This means that, in reality, every installed Linux box is more likely to be a lost sale for Sun, HP or IBM than for Microsoft.

In fact, the company that should be most worried about Linux is Sun. Scott McNealy, Sun's chairman and CEO, hasn’t been especially complimentary about Linux in public, which is no surprise. As the Unix community’s guiding light and the owner of today’s best-selling enterprise Unix, the appearance of a serious competitor for both of those titles has got to make Sun nervous. I believe that five years from now Linux will still exist, and I’ll bet it will even be a fairly popular Unix. But its growth will come largely at the expense of Sun, not Microsoft. It’s another example of the Law of Unintended Consequences: Linux fans are driven by Microsoft hatred, but the company they’re really hurting is the one that’s done more than any other to make Unix a successful, mainstream system. --David Chappell is principal of Chappell & Associates (Minneapolis), an education and consulting firm. Contact him at david@chappellassoc.com.