Microsoft could sell road apples as jewelry and win awards for doing it. They are masters of marketing.

Even though the DOJ has been unable or unwilling to out lawyer Microsoft, you can still get more computing power for less money if you make educated decisions about your operating system purchases. It's still mostly a free market; you just have to walk up and down all the aisles.

I think most experienced users of both operating systems would share the opinion that UNIX-like operating systems are more stable than Windows operating systems. UNIX-like systems don't need to be rebooted so often. They require less ``fixing'' once they've been configured properly. They generally require less hardware. They're more secure, owing to their open architecture, their simplicity, their source codes being audited for security flaws, and their proven design paradigms. They are more scalable both to large as well as small implementations, generally speaking. So why is Windows so popular and why don't more people use a UNIX-like operating system?

Being marketing wizards, Microsoft has counted on the fact that only very few people possess the technical expertise to ever compare UNIX-like systems to Windows. Microsoft has known that if they could simply put their OS out in front of as many people as possible, through whatever means necessary, they would have already captured the hearts and minds of the marketplace. Making their product technically superior has never been necessary.

They were right. As the world became increasingly dependent upon Microsoft's products and used to their ``look and feel,'' the company has extracted more and more dollars for fewer and fewer goods and services. This is typically called monopoly power. (See your nearest Economics textbook for a more accurate description.)

Our main challenge today, then, is educating ourselves enough to make our own decisions and not simply following the dictates of Microsoft and their clever marketing. When choosing an operating system, knowing less will cost you lots more. Education about operating systems equals dollars saved.

Microsoft has now grown about as large as it can through normal marketing strategies. (I.e., you buy one product and you pay one license fee.) In order to support their current size and grow further, they have to get clever again and conjure up new ways of capturing revenue. And, this is where they are expert--creating new streams of revenue. For example, Windows XP is enforcing their licensing in new ways. You may only reinstall a copy of XP a limited number of times on your home hardware before you must prove that you actually own the license. And you're actually leasing your copy of XP rather than buying it. (See http://www.fullfont.com/xplicensing.htm for more information.) These are fundamental changes in how operating systems are bought by consumers.

Further, I predict that potential revenue streams from OS leasees' private consumption data will prove too tempting for Microsoft to pass up. Microsoft will eventually leverage their operating system dominance by capturing leasing users' private data into a massive database and ``sharing'' it with ``partners'' in order to ``better serve the user.'' Every time a user reinstalls, Windows XP checks a software and hardware ``fingerprint'' on the system. Windows XP wants to know whether the user is eligible for the upgrade or installation. But the data checked as criteria for the elibibility has great value in itself. What's to stop Microsoft from capturing more revenue by data wherehousing this information and charging ``partners'' who make periodic withdrawls from the bank?

Adopting An Alternative. There are almost inumerable applications where you can use UNIX-like operating systems more effectively and economically than Windows operating systems. In fact, I would say the only times it makes sense to use a Windows operating system instead of a free UNIX-like operating system is for special applications where UNIX/Linux applications don't exist. Or, where you're forced to use a proprietary document format that only Microsoft supports.

This last point, proprietary document formats, has hamstrung Linux/UNIX until recently. It's also one of the strongest monopoly tools Microsoft has. Microsoft Office has been ubiquitous in the workplace. Most people are completely hypnotized that no other legitimate file formats exist than those supported under Microsoft Office. Of course, this is by design on Microsoft's part. (Microsoft has also been trying to extend its market power onto the web by forcing web design tools to support only Microsft-based browsers and protocols. This move toward exclusivity is probably also at the heart of it's Dot Net technology. This maneuver was also underlined in the Halloween documents from several years ago. See www.opensource.org/halloween/halloween1.html.)

But this incompatibility of competing file formats is largely a myth. There are non-Microsoft tools that support Microsoft file formats. And you don't need to marry Microsoft in order to use commonplace applications like word processing, spreadsheets, and various presentation formats. While it's convenient in the short run to settle on Microsoft as a one-stop shopping place, you wind up paying a huge, Faustian-type price for this choice over time. XML-based solutions, for example, would be far wiser.

The Back Office. As servers go, there is little that cannot be done with UNIX-like operating systems. These are machines that must be trusted to operate continuously without complaint. UNIX and Linux are great at this. Naturally, Microsoft has tried to leverage their strength in the desktop arena into the back room server by making their proprietary desktop protocols, such as Exchange and Outlook integrated scheduling, be supported only through their server technologies. Again, it's brilliance in how to capture revenue streams. But, by removing yourself from dependence on these proprietary protocols and replacing them with open protocols, such as good old SMTP, HTTP, or SSL, you can accomplish more work for less money.

Instead of deploying an Exchange server, why not deploy a similar implementation over secure HTTP? You avoid the cost of dealing with Microsoft and gain host of alternative avenues of support available from the open source community.

The Desktop. Sometimes, there is no software alternative to Windows software. Examples are childrens' edutainment games (however, see linuxforkids.org), do-it-yourself household software, certain audio/visual file formats, such as Sorenson MPEG support, and others. If you are committed to one or more of these applications, or similar unsupported niches, you'll need Windows or a Mac to do your job. But, really, the vast majority of work often does not include these software niches. For common web surfing or email reading or office work, you can easily use open source equivalents.

Alternatively, you can adopt a two-computer solution, or even a dual-boot or multi-plexing solution. You can run Windows on those few occasions when only Windows will do, and you can run a UNIX clone the rest of the time. If you have a fast computer, you can multiplex it by running a virtual machine that can host more than one operating system simultaneously. (See vmware.com, netraverse.com, bochs.sourceforge.net, plex86.org.) I personally have several computers in my office that I switch between as suits my needs at the moment. Normally, there is a computer running Windows I can use if I have to. It's rare that I cannot accomplish something with a free UNIX OS, though. For me, the only thing I can't yet do with a Freenix OS is let my daughter play certain edutainment games.

Some userland tools for *Nix are less developed than their Windows counterparts. But the chances are very good that whatever product category you've wanted probably exists now for some form of *Nix. Everything from technical analysis and stock trading software to animation or music composition. If you haven't looked in a while, check out www.linuxapps.com.

There are some Outlook clones out there (without the security weaknesses, hopefully): Evolution (http://www.gnome.org/gnome-office/evolution.shtml), TradeXCH (http://www.bynari.net/Support/Downloadable_User_Manuals/TradeXCH_Press/body_tradexch_press.html), and Aethera (www.thekompany.com/projects/aethera/), to name a few.

Openoffice.org and www.gnome.org/gnome-office/ will provide more information on two possible substitute office application suites. Koffice (www.koffice.org) also provides a full suite of applications, but they don't promise Microsoft compatability yet. You can also check out Applixware (www.vistasource.com/products/axware/). I've had very good luck with that in the past.

Un-Microsofting Yourself. Of course, it would make far more sense in the first place to avoid proprietary formats for filetypes you use frequently. It would be better to use open file formats so that your chances are fewer of being stranded with an unsupported format. That's my biggest complaint with Microsoft formats. Once they have you hooked into their file formats, they change the format, forcing you to upgrade to their latest version of Office. It becomes apparent after time that capturing more of your money was the primary reason for the upgrade. It makes much more sense to adopt some sort of open format, such as SGML/XML, where documents can be reused and output into multiple formats or media when required.

The costs of moving more jobs to a UNIX-like operating system often can be offset with benefits of avoiding common Windows problems. No more viruses, no more forced upgrades, no more $200 tech-support calls that still don't answer your questions. Many more benefits accrue that save many dollars over time.

Perhaps the largest cost of migration is the anxiety of doing something unfamiliar. Changing the way you currently do things forces real costs in production and on morale. But, when you consider that most computer users have become used to rebooting computers more often than they refil their coffee cups, that they believe viruses and worms are ``just part of life'', that they expect to pay large amounts of money for inadequate technical support, and that they just have to accept whatever inconveniences and costs Microsoft sees fit to inflict, it's clear that users have already learned a lot. One company has already trained most all of us to adopt their own Kafka-esque vision of computing. If we can be trained to use broken and expensive Microsoft products, surely we can be trained to use products that cost less and perform better. But we have to keep an open mind.

In conclusion, you are the master of your computing destiny if you want to be. If you don't want to be, Microsoft is eager to control your destiny for you.