Microsoft recently posted an exposé of Linux myths, declaring that Windows NT/2000 is in every way superior to Linux. The Microsoft Linux Myths page includes truth, lies, spin, leaps in logic, and irrelevancies. I'm not going to address any of them.
Others have addressed the misstatements on the Microsoft page. They have done so quite well. But that's not why I don't want to address the specifics.
As of this writing, the LinuxWorld poll that asks how people reacted to the Microsoft Linux Myths page has these results:
42 percent "yawned"
39 percent "laughed loud and long"
9 percent "became outraged"
8 percent "agree with Microsoft"
In other words, our readers probably think Microsoft's page is irrelevant. But that's not why I don't want to address the specifics, either.
Here's the reason: I'm the one who created that poll, and in retrospect, I can see that I made a serious error. One of the responses should have been "celebrate." That's the choice I would have made. And the fact that the Linux Myths page is cause for celebration is more important than what the Linux Myths page actually says.
Pressed for time
What I'm celebrating -- and what you should be, too -- is that we didn't read this story about Linux Myths in ComputerWorld, InfoWorld, PC Week, Information Week, or any of the other major trade journals.
As little as three years ago, that's where you would have read this story. And most readers wouldn't have known that Microsoft virtually wrote it from beginning to end.
Here's how it would have happened. Microsoft would do one or more of the following:
Commission a research group to gather statistics that support the conclusion that Windows NT/2000 has the most promising future in the market Commission benchmark tests which, by design, show that Windows NT outperforms Linux Commission, conduct, or manufacture a tightly controlled security test that demonstrates that Windows NT/2000 doesn't really have the security holes it is reputed to have Encourage (or assign) people to frequent public forums, spreading discontent about Linux and promotion of Windows NT/2000, thus spreading a negative perception of Linux Then someone from Microsoft or one of Microsoft's public relations firms would call the trade journals with an idea for a story. This person would present the "independent" benchmark and security test results, "independent" research results, and "independent" testimonies as evidence that Windows NT/2000 has nothing to fear from Linux. Since the evidence would all be in Microsoft's favor, the natural conclusion to draw would be that the hype about Linux must be based on myths.
The reporters and editors would then get a couple of quotes, write the story, and prepare it for publication. Even if they suspected that the data had been manufactured, they couldn't ignore the story. It's probably inevitable that Windows NT/2000 will destroy Linux, they'd think. Can the publication really afford to be the one journal that didn't see it coming? Besides, so what if the data are manufactured? Everyone does it.
The winds of change
So why didn't it happen that way this time? Has the trade press suddenly developed critical thinking skills? Hired a staff of investigative reporters? Developed spotless integrity?
Hardly. The primary difference is that, in the past, nobody in the trade press was motivated to expose the story as a sham. For a long time, Microsoft was a trade publication's primary source of revenue, both directly and indirectly.
Microsoft's advertising was (and often remains) the direct source. But one thing has changed. As long as vendors could make gobs of money developing hardware and software products for DOS and then Windows, even competing vendors were Microsoft's best friends.
The more money Microsoft's friends could make, the more money they could spend on advertising. Therefore, supporting Microsoft meant filling pages with ads not only for Microsoft products, but for a variety of related office suites, utilities, servers, workstations, and so on.
Here's what has changed: vendors are realizing they can make more money by supporting Linux than by supporting Windows. There are a variety of reasons for this, most of which stem from the fact that Linux is open source and free. Perhaps the most important reason is that a vendor can develop a product for Linux without needing to worry that Microsoft will usurp the vendor's market. Almost as crucial is the fact that vendors can sell software and machines without paying a Microsoft tax for every unit.
Whatever their motivations, the money vendors have for advertising becomes more dependent upon the success of Linux as they discover the gold mine in our favorite OS. Therefore, it is in the best interest of trade publications to be careful not to cut off this potential source of future revenue.
The razor's edge
Right now the future remains uncertain for many trade publications. They aren't ready to buy into the guaranteed success of Linux. For that reason, Microsoft certainly hasn't lost all of its influence. The company still generates a lot of press with targeted announcements about the future of Windows NT/2000. They are targeted at those specific areas where people find Linux more attractive. And the press dutifully publishes these announcements and supporting quotes without applying critical thinking.
For example, consider the fact that many of the people who are abandoning Windows NT for Linux are doing so because Linux is far more stable. In that light, don't you find it interesting how many news stories include quotes about how stable the Windows 2000 betas/release candidates are when compared to Windows NT 4.0? As a result of such buzz, Windows 2000 is already gaining the reputation for being infinitely more stable than Windows NT, and therefore, by implication, at least as stable as Linux.
This is despite the fact that a final version of Windows 2000 has not yet been released. This is despite the fact that there are millions of lines of new code and a constantly changing feature set in Windows 2000, rendering any conclusions about the current beta or release candidate irrelevant. This is despite the fact that Microsoft promoted Windows NT as rock solid which, if true, would render such comparisons meaningless. This is despite the fact that anyone with an IQ greater than his or her shoe size knows that most Windows installations are relatively stable for the first few months, until people install new drivers or applications -- and know that gradually increasing instability is a symptom of poor design.
In spite of all these factors, the press sees fit to endorse such meaningless quotes about how much more stable Windows 2000 is than Windows NT. No, the press hasn't suddenly developed a conscience.
The silver lining
What the press has developed is a sense that Microsoft isn't always going to be paying its bills. And it has developed a suspicion that Linux may pay more of them in the future. If the press has a little more testosterone as a result, this is why.
As a result, although Microsoft has always used underhanded marketing tactics, you're now reading more about them. You're reading more about Microsoft PR blunders. You're reading more stories about how institutions are ripping out Windows NT and replacing it with more secure and stable alternatives. And you're reading fewer stories that are obviously manufactured by Microsoft's commissioned data.
So fans of the truth, as well as fans of Linux, have cause to celebrate. Microsoft may still have a good portion of the trade press in its pocket. But it now has to fight a little bit harder to get its propaganda on the front page. And as evidenced by the Linux Myths Web page, Microsoft sometimes has to resort to publishing some of that propaganda itself. Not all of it, perhaps, but some of it.
It's not a total victory for truth. But it's a trend worthy of a bottle of champagne.