Competition keeps Linux lean
(IDG) -- If you hang around Linux people for very long, you're going to hear some bickering about whose widget has the best kung fu: "My distribution will always be superior to yours!"; "Your windows manager is lame!"; "You use what for an editor?"
Name a topic and it's up for debate. That's just the way it is. Linux people have opinions of their own and they aren't known for being shy about expressing themselves. Which can get a little disheartening if you worry, as many do, about not presenting a united front against the forces of the Windows juggernaut.
But it's only disheartening until you understand that the noise you're hearing in these debates is the victory bell, the secret sauce, the very reason Linux is gaining on (and in some cases, overtaking) that once invincible, still heavily entrenched, but now no longer cool OS from Redmond. It's called competition.
The only competition Microsoft has seen in recent years has been the trivial feuding between the Windows 9x and Windows NT development groups. If you look closely at the shipping versions, that frail, half-hearted competition hasn't helped either platform very much. Windows 9x is still a hybrid GUI riding a horse called the DMM (dangerous memory model). Windows NT's stability peaked at release 3.51. Each subsequent release has become more of a kludge and less robust. Windows 2000 promises more of the same.
But look what's happening with the Linux operating system. Look at the steady increase in sophistication, power, and marketshare, not to mention platforms, applications, and distributions. Why? You got it, bucko: competition.
Let's start at the top. Or at the bottom, as it were: the kernel is the basis of the Linux operating system. If you want to do some kernel hacking to add a fix or improvement or polish to the Linux kernel, fine. But your code won't be automatically accepted and made a part of the next kernel release. That's an honor. And that's where the element of competition comes in.
Linux kernel code is administered on something like a feudal basis. Good King Linus sits at the throne and "owns" the bulk of the code. Various other parts of the kernel are "owned" by liege lords. Contributed hacks are funneled to the appropriate noble and either axed or blessed for inclusion. The ruthless efficiency of this competition is reflected not only in the speed and power of Linux itself, but in the improvements seen in each new release.
Now consider the Windows operating system. Competition is part of the Windows development process as well, but it's of a different sort. For one thing, it's all of one house -- a single house modeled after a cathedral. Here, the word cathedral is used not in a spiritual sense, mind you, but in the sense described by Eric S. Raymond in his watershed essay "The Cathedral and the Bazaar."
Moreover, competition in the house of Windows is a competition of evil ideas, malignant marketing, and monopoly management. Quality of the code isn't important. Instead, the holy grail -- marketshare and money -- is reached by embracing competitive innovations, thereby extending the Microsoft monopoly and eventually annihilating the competition.
In the Linux world, competition doesn't begin and end at the kernel. It extends to every aspect of distribution, packaging, and support. It extends to Linux applications, too, and (as we're seeing more and more these days) on down to the tools used to develop those applications.
This spirited competition also extends to those of us writing about Linux. Not just in e-zines and magazines, but in newspapers and books as well. There is, for example, tremendous competition for all the eyeballs currently reading this column.
As a reader posted in response to a recent column, Red Hat isn't Linux. Amen and amen, brother! Neither is Debian. Or SuSE. Or TurboLinux. Or Slackware. Or Caldera. Or Stampede. Or any other distribution. And all of us in the Linux community reap the benefits of that situation.
You don't care for the commercial distributions? Fine. Thanks to the open market that surrounds open software, you can choose not to support any of them.
You prefer a commercial distribution? Great. You have your choice of a number of fine ones. Or you can simply download the Linux source code and create your own. It's all good. Whatever floats your boat.
Personally, I want to see Linux continue to gain marketshare, both in the server world and on the desktop. Unless people and firms are able to achieve commercial success with open source software in general (and Linux in particular), the ceiling will remain far too low for my tastes.
Don't get me wrong. I like the idea of free and/or open software. I love what it has given us in terms of operating systems, applications, and tools. The energy and devotion that produced those gifts simply do not exist in the commercial world. But it's the synergy between the bazaar and the cathedral that will carry us all to that mythical marketing plateau known as the next level. Competition is an important element of that synergy.
As Linux has begun its migration from server to desktop over the past year or two, it has gotten a tremendous boost from the well-publicized, oft-debated competition between KDE and GNOME, each of which is vying for the title of ultimate Linux desktop environment.
Yes, there have been some spirited debates, some untoward comments, some amount of wasted energy by advocates of both camps. But the algebraic result, the net of the competition, has been a huge positive. After all is said and done, competition has been a boon to the morph of Linux from a CLI geektop to a modern user's desktop.
I think most users, like me, would buy Linux applications (like Quake II or WordPerfect or CodeWarrior) even if they weren't available under the terms of a free or open source license.
Likewise, we'll buy commercial offerings when it makes sense to do so -- that is, when a commercial version offers enough additional value over a competing free or open source offering. By the same token, vendors attempting to peddle Linux wares that are of doubtful quality, or that are only as good as open source equivalents, are going to find it very hard to make a dollar. We, the Linux community, win in either case.
Competition between the free and/or open software world and the commercial software world is just as real, just as vital, and just as beneficial to the Linux community as is competition in any marketplace.
Microsoft has its second monopoly in office suites. Microsoft owns this market just as it owns the market for desktop operating systems. In the domain of Microsoft Office, competition is no longer a dirty word in Redmond, because it just doesn't exist. As a result, prices go up. Quality goes down. And 'nobody's getting fat 'cept Mama Cass.' Or in this case, Bill Gates's bank account.
In the Linux free market (free, in the sense that there are many stalls in the bazaar and you're free to pick the one you want to shop at), we've already seen the benefits of competition for office suites. And we will continue to see such benefits as StarOffice and Applixware continue to mature.
The good news is, there's more competition in the works. KOffice is coming, as is Corel's WordPerfect Office 2000. There will be others as well. Just as Linux has already surpassed Windows in the goodness of its kernel, competition makes it inevitable that the same thing will happen among office suites. The dynamics of competition produce superior products -- and Linux users will reap the benefits.
Look at what's happened in the database systems arena over the past year. Why have Oracle, Sybase, IBM, and others jumped on the Linux bandwagon? I think we have to give the commercial success of Red Hat some credit for this sudden allegiance. The scent of money attracts more money.
And this shift isn't limited to the server-applications arena, either. Firms from the Windows and Mac markets are beginning to migrate as well. Firms like Photodex, for example, which is bringing CompuPic to Linux this year.
If you build it, they will come
One vital area must be addressed before Linux can make its move to the next level: development tools. Developers will soon be coming to Linux from Microsoft Windows, from the Mac, and from elsewhere in Unix. As this happens, demand for the kinds of tools provided by, oh, let's say Microsoft's Visual C++ and its cousins, will grow.
IDEs are the eye of the storm for the near future. And those who provide the shortest ramp-up time for Visual C++ programmers to start producing Linux applications will win the lion's share of the market.
The good news is, the process is happening as we speak. We've already seen Metrowerks announce and launch CodeWarrior, its first development platform for Linux. Others, from both the free/open source world and the commercial world, are following suit. And that's a good thing, because application development tools are critical to Linux's future -- not in terms of who wins or who loses, but in terms of the benefit of the process.
In fact, looking at market competition in terms of "they must die so we can thrive" is a hangover of the Microsoft era. It is the antithesis to the meme that competition is good.
Redneck Software has just announced an open source project for application development called Moonshine. Redneck Software's stated purpose is "to bring software application development tools to Linux."
Metrowerks already had open source competition for its IDE; now it has more.
On the same day that Redneck announced Moonshine, KDE announced a new release of KDevelop: "a C++ development environment which makes the creation and development of GNU Standard Applications an easy task even for beginners."
Yesterday, June 22, Cygnus officially announced Code Fusion, its entry in the parade of IDEs. Cygnus hopes to translate its guardianship of open source compilers into commercial leadership in the field.
The buzz on the street is that there will be several more commercial IDE offerings appearing over the next few months. Each additional tool will open the door for more developers. More developers will mean more applications. More apps will mean more competition. All to the benefit of consumers everywhere.
Some people scoff when I say that Bill Gates is afraid of competition. They believe he's simply ultracompetitive. That's not the way I see it. If you were competing against Bill Gates in a track meet, he would sabotage your starting blocks, bribe the officials, loosen the cleats in your shoes, use a trick count for the start, and do anything and everything else he could think of to avoid having the outcome determined by virtue of performance on the field. I don't call that competitive. I call it cowardly.
Microsoft has managed to create, maintain, and extend a competition-free zone for the Microsoft Windows operating system and for its key applications. In the short term, it's hard to argue with the results reflected in the bottom line.
But just as lack of competition has been the reason for the perennial success of Windows, it is also the cause of its downfall. Without the benefits of real competition, Windows has grown fat, short of breath, and weak on staying power. Linux is only now beginning to flex its muscle. I know which horse I want to carry me into the new millennium. What about you?
Joe Barr is a software professional, writer, and self-proclaimed dweeb. He has been working in the industry since 1974 as a programmer, analyst, consultant, and manager. In 1994 he began writing a monthly column called Papa Joe's Dweebspeak Primer in Austin, TX's Tech Connected magazine. The column exists today as an e-zine and newsletter at www.pjprimer.com, which has run on Linux since its inception.
Torvalds talks: Where Linux is headed
RELATED IDG.net STORIES:
The memes, they are a-changin'
|Back to the top||
© 2001 Cable News Network. All Rights Reserved.|
Terms under which this service is provided to you.
Read our privacy guidelines.