The Linux Bandwagon

• 10/07/1998

Excellent question, Emily. Is this freeware creation a tempest in a teapot, or is it the Windows killer that Microsoft Corp. detractors have long hoped for?

Of course, the answer to that last question is no, unequivocally no, and anyone who says otherwise is a fool. Just because something is free and perhaps even better than the leading product does not mean it will succeed, certainly not in the computer business. Look how long customers put up with one buggy version of Windows after another until 3.0 rolled out, and they didn’t flock to the superior Macintosh GUI in the interim. It was the commitment to DOS that kept customers in the Microsoft camp, and it is the overwhelming commitment to Windows that will keep the vast majority of them from jumping to alternative operating systems.

But what about Linux? What’s known about it? Why has it attracted tens of thousands, possibly millions, of users, many of them borderline Linux fanatics? Will it become mainstream, at least to the extent that, say, Unix is mainstream? Is the possibility of such bolstered by the fact that mainstream companies such as Oracle Corp. and Informix Software Inc. have jumped aboard the Linux bandwagon?

Linux owes its development and its name to Linus Torvalds, a Finn who did the lion’s share of the coding as a student in the early 1990s. His relative obscurity ended this past summer when he became a cover guy for Forbes magazine.

The operating system itself is available free of any charges via the Internet. Linux has an open source code model, meaning testers can modify the code to meet individual specifications. Try that with NT! In theory, having such access to the code should translate into fewer third-party support calls because the in-house developers will better understand how the applications interact with the operating system. In theory.

Companies such as Caldera Inc. and Red Hat Software Inc. are Linux distributors selling various products that leverage Linux’s many outstanding attributes. The products carry what seem like mythical price tags, such as Red Hat’s $29 price for its Extreme Linux. Of course, there is no support for the product other than that found on Internet chat groups. Oracle will develop and sell a Linux version of its Oracle8 database. Informix has already begun selling a Linux-compatible copy of Informix SE, its small-system database offering.

Meanwhile, International Data Corp. (Framingham, Mass.), which does its best to track Linux’s installed base (a tough task when most users don’t bother to register), says that 20 percent of the server operating system market will have Linux written all over it within 4 years, up from just 6 percent last year. Certainly that would make Linux mainstream, by any definition. If the numbers hold up, that is.

What do the users say about this worst-kept secret? For one thing, they’d like to see some big company really get behind the Linux movement big time. Maybe IBM/Lotus? That has yet to happen, and I have to believe it is because the risks far outweigh the perceived potential for reward.

Also, they’d like to see applications that take maximum advantage of Linux, but they want those applications to read the massive installed base of legacy applications -- namely, all those Windows-based desktop applications out there. Right, and I want to be an astronaut.

Linux is nowhere near as resource-intensive as NT, users report. And it runs on a wide range of hardware platforms, although installation is said to be no easy task. But it does come loaded with compilers, drivers, source code and network software. Users rave about its speed and overall performance.

I’ve always maintained that things in the computer industry take a lot longer to happen than most people think, and there are two underlying, interconnected reasons for this. The first is the existing installed base of hardware and software, and today that installed base on the desktop -- and increasingly at the low end of the server platform -- is all Windows. The second reason is the installed base of skills, the people component. Changing that is far more difficult and slow than changing products.

Thus, I have to conclude that Linux, with all its apparent outstanding attributes, will remain a niche player for the foreseeable future. Maybe you should consider putting Linux up on a server in a test cell with Windows-based workstations attached to it. But in any case, it just doesn’t make sense at this early juncture to make major commitments to Linux, free or not. --Bill Laberis is president of Bill Laberis Associates Inc. (Holliston, Mass.) and former editor-in-chief of Computerworld. Contact him at [email protected].