OPINION
How Linux Saved Microsoft

As I look at how Microsoft is changing to address the Linux threat, one that may actually turn out to be no more real than Netscape's was, I can't help but see how Microsoft has dramatically benefited from it -- and much more broadly so than they did from the rise of Netscape.

Recalling Pre-Linux Microsoft

We so often get caught up in today's events that we forget what it was like before. We take for granted improvements that have dramatically changed our lives. The same can be said for some of today's problems, too. For instance, when I grew up we didn't have heavy traffic -- but now it just seems like an unavoidable fact of life.

In the '90s we had Windows 9x products which had memory leaks (you couldn't run them for much more than a day without crashing), hardware driver issues (graphics drivers in particular were a nightmare), and Microsoft was behaving badly.

The best example of bad behavior was their release of Windows NT for the desktop. Here was a product designed for workstations, real workstations, and servers. Priced at a premium for a market that would buy a premium product, suddenly Microsoft felt it was ready for the desktop. It wasn't, and the desktop market sure as heck didn't want to pay the premium, but there was no other choice and we had what amounted to a contained OS war with Microsoft on both sides. The Windows 9x side lost.

During that time Microsoft wasn't particularly happy with me, as I was outspoken in saying that Windows NT wasn't ready. Imaging products didn't work with it, compatibility was bad (particularly with notebook computers) and, worst of all, virtually every deployment came in well over budget and well under expectations.

Right Time

It would be as if Ford, after putting all of the other car companies out of business, decided that everyone should stop driving pickups and start driving dump trucks. There was no competition; OS/2 (which was basically an earlier version of Windows NT anyway) was dead, Apple (Nasdaq: AAPL) was clearly no longer a threat, and none of us were taking Linux seriously. It, like Plan 9 (an arguably better but more obscure OS), existed just for discussion between people who wanted to show off how geeky they were.

Towards the end of the decade Microsoft made a pricing adjustment that was supposed to address the biggest complaint of the time, pricing complexity, but the execution was so badly botched -- largely because of the fear that Microsoft would lose money -- that customers all but revolted.

Because of declining competition, Microsoft let its advocates nearly die out and they adopted a set of policies that, to me, felt much like IBM (NYSE: IBM) had a decade or so earlier which seemed based on the "we are selling air so you don't have any choice" marketing and sales concept. Microsoft was on the bullet train to disaster, and Linux gave them a much needed kick in the rump.

Linux's Positive Impact

Now look at Microsoft. They have adjusted prices downwards sharply in third world countries which, before this, couldn't even afford Microsoft's products. They have improved the quality of their products to a degree that many of us run Windows for weeks without crashing. The products are vastly more secure and are wrapped with an infrastructure of services that makes them much harder to penetrate and vastly more capable.

Patching is largely automatic, problems with hardware have all but been eliminated, and much of the development and break fix processes have been fully automated with testing that is magnitudes better than it was only four years ago. Microsoft is listening as well, as all you need to do is use Linux in a sentence and your Microsoft rep will remain glued to your side.

Microsoft is also going though a massive effort to, once again, capture the costs associated with their products and find ways to reduce these problems so that the products better compete with Linux. The company has gone so far as to hire large numbers of Linux programmers to implement policies and processes that have many of the advantages that Linux promises.

An example is its Shared Source Initiative which provides source code to an ever-increasing group of companies that feel they need to look at source, something that Microsoft would never have even considered a few years earlier.

Virtually none of this would have happened had it not been for Linux, and had it not, we would have seen a different version of Longhorn (the next version of Microsoft Windows) than we will now see, and we probably would have liked it vastly less as a result. Microsoft is even aggressively working on its culture, which has been the core to Microsoft's appearance of untrustworthiness and arrogance. Those of us who watch the company have seen dramatic improvements over the last 12 months in Microsoft's general behavior as a result.

I'd like to ask the Microsoft folks to stop reading here, and to remember that if you don't continue to make the changes you are making, Microsoft will fail much like IBM failed. In other words, most of us really like that you are changing, so for goodness sakes don't stop; you still have a ways to go.

False Threat?

Even the fact that we are talking about Linux, which not only isn't a company, but isn't even an operating system, is a sham. There is no Linux product; not really, Linux only refers to the kernel which, by itself, doesn't actually do much of anything. It is more the core point of a concept that surrounds "open source" which, in turn, is based on a false concept. This concept is that people actually want to look at source code.

Think about it: For decades we have surveyed companies and for decades, except for those who are actually in the software business, the vast majority have said they don't want to be in the software business. Yet open source , as it is supposed to be practiced, puts you squarely in the software business. We also know that IT buyers want the vendor to enjoy all of the product liability associated with an offering, but open source, at best, passes some of that liability to the customer, and, at worst, all of it.

Finally, we know that what is largely holding the open-source community together is a dislike for Microsoft. As Microsoft improves, the reality of what Microsoft is will slowly penetrate the increasingly artificial reality that the open-source community has created and, much like it was with Apple, non-aligned buyers will avoid the related platforms and aligned buyers will change sides as their perceptions shift to the new reality.

Those of us that have been following this market for decades have seen this pattern over and over and over again and while it is never certain (click here to read an interesting abstract discussion on the misuse of history to prove a future event) it would seem likely that, unless something dramatically changes, by 2015 we'll be largely wondering what all the fuss surrounding Linux was really about.

Sometimes no matter how big a balloon looks you have to just wonder if the important part isn't simply all of the hot air.