InformationWeek

February 8, 1999

Linux: The Ultimate Enterprise OS?

By Sean Gallagher

eff Papows knows when he's made a mistake. And the Lotus CEO made one last year when he said his company wouldn't build a version of Domino for the Linux operating system. Last month, Papows told attendees at the Lotusphere99 conference how he has been accosted by Linux supporters everywhere--in line for "A Bug's Life," and even in his own driveway.

Now Lotus will produce a Linux version of Domino 5.0--and every other Lotus server product. The turnaround, only a few months after Papows said no to the operating system at Lotusphere Berlin, isn't all that surprising. Corel, Informix, Netscape, Oracle, and Sybase, have already made the move, and IBM has a beta version of DB2 for Linux in testing.

But was Papows really wrong, or was he just coerced by the Open Source illuminati? Can Linux, a bunch of code written by hundreds of programmers around the world out of the goodness of their hearts, ever really be an operating system that businesses bet their lives on?

You bet. Linux is rapidly catching up with, and in some cases surpassing, the features of Windows NT. It's got many Unix implementations running scared, too: Linux on Sparc has outperformed Solaris on Sparc in some tests, and it's an order of magnitude easier to install and configure despite Sun's best efforts. All of this is precisely because Linux is a product of the Internet; unlike other operating systems, its development cycle is based on Internet time.

A few years ago, we were asking whether Windows NT could ever make it in the enterprise. Guess what? We're still asking it. Windows NT has developed into a market force, but it still has to address some major scalability and stability issues.

Windows 2000 may solve some of them, whenever it ships, but it probably won't answer all of them. There's a simple reason: Microsoft is using traditional commercial means to produce it. Beta testers of Windows 2000, who are due to see beta three sometime soon, will get to kick tires on test code, but they won't ever be able to make fixes themselves; instead, they'll file bug reports and wait for Microsoft's programmers to get around to their requests.

On the other hand, Linux is always being patched and upgraded, both by its developers and its users. If software developers can't get applications to work with part of the operating system, they can look in the source code to find out why. If a bug in the operating system is found, as it was when the "ping of death" denial of service attack was discovered, a fix in the source code is usually available-- not in months or weeks, but within hours. Another attractive element of Linux is that it can be as thin or as fat as you want it to be. Because of its ad hoc origins, it's highly modular and it can be configured to run in just about any environment. As you read this, someone somewhere is running Linux on a 3Com PalmPilot. Despite the best efforts of Microsoft to put Windows everywhere, Windows NT and Windows CE share little besides look and feel. But there's something even more compelling about the Linux model than express bug fixes, access to source code, and rapid overall development. Linux offers something that no operating system vendor could provide during the past 20 years because of simple business barriers, a potential standard multiplatform, vendor-independent operating system. The open source model of Linux is what makes that possible.

Linux is the stone soup operating system. It started out as Linus Torvald's pet project, the proverbial stone in a pot of water, and its simplicity (combined with some necessity) drew the gifts of developers around the world. Imagine if, rather than spending money on commercial operating system licenses, companies and institutions instead spent a fraction of the money on donating developer time to the Linux effort, porting it to new systems, and increasing its functionality.

Sure, that's a utopian, communal, hippie kind of idea. But it's one that makes business sense. How much of a competitive advantage has your operating system given you lately? How much more of a competitive advantage would you have had if you had spent money on hardware, applications, or more developers, instead of on operating system licenses? It's thoughts like those that make the most hardened capitalist into a hippie.

Microsoft's antitrust lawyers may have a point, one that may not exactly make Bill Gates happy. At some point in the not-too-distant future, Microsoft's monopoly of the operating system market may become moot.

Sean Gallagher is managing editor of InformationWeek Labs.

AuthorITies Archive

Send Us Your Feedback

Top of the Page


Karyl Scott: Enterprise View Karyl will explore the business and technology issues surrounding enterprise systems.	Stuart J. Johnston: Redmond Watch As our eyes and ears in Redmond, Stuart gives his perspective on the latest events at Microsoft.	Charles Pelton: Eye On IT Charles explores IT management issues and strategies that business and technology managers face.	Jason Levitt: Internet Zone Jason focuses on the strange, egregious, and the standard technologies of the intranet/Internet.	Lou Bertin: The Observer Lou offers a view of the good, the bad, and the bizarre developments in the technology business.