You have my apologies for not responding sooner. I forgot to subscribe to this discussion, and didn't notice your response.

Mostly what it boils down to is marketing and fear of change. Most corporate middle-management is used to a slower-moving world than the IT industry. In IT industry terms, Microsoft is a venerable leader, but in financial services industries its twenty-year rise to market dominance would be something of a flash in the pan.

The IT industry relies on other industries for its market. The pointy-haired bosses in those other industries are sold on the necessity of computers in the workplace, but they are very cautious about changes in general, and thus tend to be very slow on the uptake where adopting new technologies is concerned. Linux is two things to these people: 1) Unix's little brother, and 2) a buzzword. Knowing the power of the buzzword, Microsoft is using the ephemeral quality of buzz against Linux to portray it as being entirely smoke while continuing to produce its own buzz about Windows-platform systems. Unix, meanwhile, is seen as "outdated".

Finally, MS puts a lot of pressure on people with decision-making powers to buy Microsoft products and reject competing products. MS has a powerful marketing presence and a lot of financial leverage with which to work.

Of course, part of the problem lies in the fact that we, IT professionals, who are supposed to know the facts of the industry and give good advice, are often at least as susceptible to buzz as anyone else. As a result, it's difficult for your average pointy-haired boss to find someone to go to for advice on the matter that won't lead him astray.

As a result of all of this, the biggest reason that Microsoft has such a strong market presence and Linux sometimes isn't as well-recognized as it should be is simply that Microsoft has the money and cohesive market presence to throw around that it needs to perpetuate its own market saturation.

Ironically, Linux acts as sort of a flagship for a whole slew of other products that never get noticed by many managers, but should be. An example is the various BSD versions, which have their uses as well in low-cost servers. BSD doesn't get media mentions, however, the way that Linux does, even though it really should. Depending on your needs, BSD is every bit as valid an operating system for enterprise deployment of low-cost servers.

The American public is somewhat wed to the idea of having two choices in any matter of importance. Those who take note of Linux view it as the only real alternative to Windows. The same can be said of those who take note of MacOS. There are as well those who view MacOS and Linux as the only viable solutions to their computing needs, often grouping BSD in the same category as Linux (even though MacOS X has more in common with, for instance, OpenBSD, under the hood).

There are a great many misconceptions in the IT industry. They won't go away in a day. I'm sure I have a few of my own. Linux will continue to gain market presence, as long as Microsoft doesn't manage to effectively outlaw it by lobbying with Congress. Before it becomes a true equal contender for the end-user's desktop, though, we may see something else entirely on the rise that will supplant both Windows and Linux on the desktop -- maybe BeOS will see a resurgence. Frankly, BeOS would probably have been a better desktop OS than anything else on the market if it had been allowed to mature before it went the way of the Dodo, and if it gets picked up in force by the open source community it may still become a major player.

Only time will tell.