As the titans of proprietary and open source software do battle, neither side is without flaws.
In the war between Microsoft and those that support Linux, there have been many phases. Most of this war, like most wars, has been concealed by the fog of war and propaganda from both camps. Initially, both sides seemed more interested in positioning than in true discourse, and if Microsoft used undue pressure (often positioned as FUD), the Open Source community seemed willing to use outright physical attacks against Microsoft platforms to make their point.
At first, the Open Source side was overmatched but mistakes by Microsoft and others allowed OSS to gain in strength and now both sides would seem to be more evenly matched. Neither side seemed capable of creating a successful strategy against the other and both sides seemed to respond to the other tactically.
This seemed to benefit the challenger more than the entrenched vendor but, regardless of how often they competed with messages, for most of the last decade they actually seldom competed effectively for the same business opportunities. The two sides were simply too different but that fact seemed lost on both.
Even stranger is that both sides seemed to be hell-bent on damaging their own futures. And both, when attacking, seemed to be more effective at shooting themselves than at effectively moving against the other.
Microsoft Set the Stage
Initially Microsoft really didnt seem to take OSS or Linux seriously, which clearly wasnt the wisest tactic. Having lost track of their own progress down a path blazed by IBM a decade earlier, Microsoft was happily plodding along, systematically destroying the bridge of trust that existed between itself and its customers.
Seemingly forgetting the successful embrace and extend strategy that worked so successfully against Lotus and the emulation strategy that created Windows in the first place, Microsoft tried to drive Windows NT into a UNIX base that didnt want it. To showcase the power of the platform, they had a scalability day that, to this day, is unmatched in its historic failure to do anything but make the company holding the event look foolish and dishonest.
This was followed by a change in pricing which was supposed to address a complaint about Microsoft pricing complexity. And, largely because of an excessive focus on not taking even a short term revenue hit, it actually ended up being revenue positive and scaring the existing Microsoft installed base half to death by showcasing how vulnerable they were to the company. This base began to look for alternatives.
During this time one of the most common comments when a CIO was looking at increasing their Microsoft commitment was that such a move was simply not in their best interest. Yet Microsoft was convinced they could sell these same CIOs on abandoning UNIX in favor of Windows NT. This was a move that not only went against the prevailing will of these now concerned CIOs, it threatened the positions of those currently working on and managing the UNIX platforms that Microsoft wanted to displace.
This motivated the key stakeholders into looking for an alternative because all agreed that the cost savings of commodity hardware would be hard for operational managers to ignore for long.
In short, they needed something that could protect both the jobs of the managers that Microsofts products threatened, but ensure that they wouldnt be faced with the loss of control that resulted in excessive control by dominant vendors (like IBM had been) and Microsoft now was.
Linux into the Breach
Linux was far from perfect and had been floating around as kind of an internal science project for much of the 90s. Suddenly it went from internal science experiment to heir apparent and was positioned largely by the IT folks that Microsoft had put at risk as a better alternative for UNIX than Windows.
When the very people who are supposed to choose between products become the advocates of one or the other, bad decisions often result and are covered up.
Much like it was a decade before, with Client Server displacing Mainframes, so to it was again with Linux displacing UNIX. In the previous decade, it had been cost overruns that were concealed. With the current movement, the problems associated with replacing a very mature platform like UNIX by a, as yet, relatively immature platform like Linux were concealed from the operational managers by the IT departments.
Promising massive cost savings and faced with having to demonstrate these savings, IT employees put in massive amounts of time to create systems that fell well short of what similar products from companies like Sun could do easily. But, since IT owned reporting and had a direct interest in the result, it was as blind to what it was doing as Microsoft had been blind to creating this environment in the first place.
Eventually Linux grew up but not without a massive amount of heavy lifting from a customer base, which represented a cost many time what Microsoft was likely to have ever been able to charge. In addition, what they didnt realize was the repercussions of their actions as operational management looking for cost cutting trends began forcing a series of decisions which resulted in the massive off-shoring of IT jobs.
Many Linux advocates, flush with power, appeared to demonstrate that power truly does corrupt, and denial of service attacks against Microsoft and other foes like SCO (and those that seemed to support either) proliferated for a time. In many ways Linux became a bigger threat than Microsoft had been because the very people who were supposed to protect against any company gaining massive power to a degree that it put their firms at risk were now such a risk themselves.
This now seems to be self-correcting as Oracle, arguably an even scarier firm than Microsoft, moves in to seize control of Linux and is countered by the almost unbelievable combination of Microsoft and Novell. Whatever the outcome, the trend is now to once again separate the IT managers who are the decision makers from one of the platforms they must decide on.
However, at the end of all this, Linux is still closer to what the current and ex-UNIX (now Linux) shops want in a product. But progress is iffy and every little amount of it seems to come with substantial organizational pain.
From changes in the GPL, to anything that looks like it might be a product roadmap, Linux often makes the problems that Microsoft recently experienced getting Vista to market seem trivial by comparison.
Looking Forward for Both Linux and Microsoft
For Microsoft to accomplish its goals it needs a product that can embrace what Linux provides, which is largely job security for the old UNIX camp, without creating the threat of a vendor that is both too powerful and misuses that power. For Linux to succeed, it has to find a way to evolve to face emerging requirements for the platform, to hold off increasing internal and external threats, and to guard against power abuses by its own supporters.
In the end, both paths have a long future and it is probably best for the industry that both continue to exist and continue to compete. (And lets not forget the importance of additional emerging alternatives that help drive innovation.)
But it is also true that from the perspective of the IT buyer, this competition can only go so far because, in the end both sides will need to focus on interoperability. Because allowing one or the other to completely dominate a shop has, historically, not been in the best interest of the companies the IT organizations serve.
In the end, and this always seems to be the ultimate truth, the IT organization that maintains perspective and stays solidly focused on the needs of its company while maintaining balance between the tactical and strategic will be the most successful. It is that success, not the success of either camp, which should remain our primary goal.