| Waiting for Linux, Waiting for Godot |
by Malcolm Dean (Nov. 29, 2001)
Does waiting for Linux to achieve World Domination have anything in common with waiting for a character who never shows up on stage? DesktopLinux.com Contributing Editor Malcolm Dean explains . . .
Waiting for Linux, Waiting for Godot
There's one thing about Linux: it's downright frustrating. It's unfamiliar, it requires a new vocabulary, and it's simply hard to figure out what it really means.
I see that quiet frustration when I meet Microsoft users. Most of them have heard about Linux, but see it nowhere in their current environment, or their future. They know it's supposed to be great, but for what? They've heard it's the next thing, the latest and greatest. But what about these strange new concepts like "distributions," Open Source, licensing, and the meaning of "free." And, bottom line, does it run Microsoft Office?
At the beginning of the millennium, while I was News Editor for a Linux magazine, I wrote that 2000 would be the beginning of the end for Microsoft, and 2001 would be the end of the beginning for Linux. With Microsoft releasing an important new version of Windows in October 2001, and many Linux companies desperately clinging to life, how could this be so?
In a way, it all comes down to cereal. Until not long ago, cereals were always grown and processed locally, so that every bread would have a unique local flavor. Then powerful, giant mills took over, and bread became standardized. Wheat, for example, was shipped all over the world, becoming an essential commodity in countries where it had never grown. Nowadays, you buy your cereal in colorful boxes smothered with carefully researched logos and images. You eat Fruit Loops for breakfast, each bite carefully manufactured to please the customer's eye and palate, and distance the consumer from the original substance. Our culture now demands processed products proferred in sophisticated packaging.
Long, long ago (perhaps before some of you experienced your initial boot cycle), in a very different world called the Mid-Twentieth Century, no one had invented boxed software. There was no such thing as a "computer store" where you could browse aisles devoted to software products for the Macintosh or Windows. There was no such thing as a Personal Computer. Software was something which came with the computer, or which you had to write yourself.
The first Personal Computers emerged in the late 1960s, but they were only for the rich and famous. Honeywell offered its H316 "Kitchen Computer" in the Neiman Marcus catalog for US$10,600 (much more than the average breadwinner, as they were called, earned in a year). A decade later, they were becoming a cottage industry. Enthusiasts contributed software freely and copied code freely, even code which was originally licensed to someone else.
Which bothered William Henry Gates III enormously. Bill Gates complained in 1976 that Micro-Soft's Altair BASIC was widely copied and re-sold illegally, and would everyone now like to pay up? "Nothing would please me more than being able to hire ten programmers and deluge the hobby market with good software," he said. And he did.
When IBM introduced its Personal Computer in 1981, the world was transformed. In those days, IBM behaved like Microsoft does today, and it even had the same nickname, the Evil Empire. IBM's blessing meant that Personal Computers would crop up everywhere, and they would be accompanied by lots of BOXED PRODUCTS, some of which were Operating Systems.
Shocking as it may seem now, PCs were not necessarily delivered with a Microsoft Operating System. For a brief period, CP/M was a viable alternative, and it ran on many types of microcomputers. But IBM's blessing begat Microsoft's success, and DOS (which in those days stood for Disk Operating System, not Denial of Service) became the default.
Around 1985, computer trade shows were still dominated by DOS. In the corner, there might have been a few booths featuring software for Windows, but no one took it seriously. It was buggy, slow, and boring. But with the release of Windows 3.1, there was a reversal of fortune, and the number of DOS booths dwindled steadily.
Years of titanic struggle left us with two predominant architectures -- Wintel and Apple, their default Operating Systems, and their BOXED PRODUCTS. This is a nifty universe if your ambition is to be a consumer of said boxes. It is a universe driven by the manufacturers of these boxes, who promise a future of comfort and productivity, if only consumers will return to the stores regularly to purchase upgrades, much like cattle returning to the trough.
Now, try to imagine if in 1976 software enthusiasts everywhere had rejected William Henry Gates' belief in proprietary commercial software? Suppose they said, en masse, that it would be better to collaborate on all kinds of projects, to freely and openly share development. Welcome to Microsoft's Room 101.
And it gets worse. Because, besides being free, there really is no such "product" as Linux. Linux is a kernel, the essential core of an Operating System. It only becomes useful when it is combined with utilities and programs in a "distribution," which may or may not be a BOXED PRODUCT.
Because Linux is not necessarily a boxed product, newcomers have a hard time understanding exactly what they are faced with. You can sense the impact of this new reality in the language Microsoft now uses to describe its products. Windows XP is called "the latest Operating System," "based on the NT kernel." Even the New York Times, a publication which should know better, parrotted the phrase. Before Linux, new releases were simply "versions," and there was no mention of NT's kernel. It's a clever ploy to condition consumers to see Linux as less than the latest, and the NT kernel as an object of desire.
Although Microsoft appears to be hard at work convincing customers that the future belongs to boxed products (theirs), much of Bill Gates' activities over the past few years shows that he realizes that the future will look very different, and almost certainly does not belong to Operating Systems. G2 Computer Intelligence, Inc. provides an interesting summary of over 170 Microsoft acquisitions and investments approaching $20 billion by 2000, showing that Redmond is reaching in all directions to ensure it doesn't miss the next boat.
Then there's .NET. Remember the auto ad which said something like "lead, follow, or get out of the way?" Seeing a brave new world in which all kinds of devices and operating environments communicate routinely, Microsoft had little choice but to attempt to establish a leading environment which would place it at the center of the next stage in computing. But while Microsoft is providing "boxed" .NET products such as Visual Studio for its development community, it cannot escape the fact that the most important parts of .NET are open standards, not proprietary products. Worse, several projects are underway to implement Open Source versions of .NET, which will make it possible to remove Microsoft from the equation entirely.
Even Linux can be removed in this new world. "Out of the box," if you'll pardon the expression just this once, Linux runs DOS, Windows, and UNIX applications to some degree or other. Most of the OSes competing with Linux, FreeBSD, NetBSD, and the UNIX variants coming from major vendors such as IBM, Caldera, and others, are growing Linux compatibility layers. Even independent OSes such as SkyOS and MenuetOS plan to have a Linux compatibility layer. And applications such as VMWare and Win4Lin are making it easy to host OSes and applications on Linux.
In this organic environment, you'll see rapid development of new types of applications and technologies which have been held back by years of monolithic, and according to the courts, monopolistic thinking. Matt Dillon, a major contributor to FreeBSD, wants to add the ability for computing processes to migrate from processor to processor. IBM's Web-optimized z900 mainframes can have superfast channels between thousands of virtual Linux machines. UNIX vendors are carefully studying the strengths of their products, and paving the way to work with Linux where it has complementary strengths, such as in networking and on the desktop.
Clearly, we are not far away from the day when changing an Operating System is about as difficult as changing a GUI or an office suite -- not something you look forward to, but definitely feasible. You'll choose the OS most appropriate to your circumstances, unencumbered by concerns about which applications are available. They all will be, or they will die. Many of these new OS options will be a form of Linux, and many won't. Some will be BSD variants, and others will come from the "big" UNIXes. And there will be entirely new and exciting Operating Systems, as well.
So waiting for Linux is like waiting for Godot, a character who never arrives on stage. Linux is not Windows, and if you expect it to become the next Windows, you'll be sitting there when the house lights go down.
Copyright © 2001 by Malcolm Dean
About the author: Contributing Editor Malcolm Dean is a writer and IT strategist based in Los Angeles.
Talk back! Do you have comments or questions about this article? talkback here
(Click here for further information)
|
|
|
news feed