Can Linux Do What Netscape Couldn't?

A few years back, a lot of folks got really excited about Netscape. They were going to be the next Microsoft, their founder was lauded as the next Bill Gates, their IPO started the .com boom. The decade ended with a lot of us looking rather silly. Netscape had apparently had little direct impact on Microsoft at all. In fact, Netscape's biggest impact came after it was acquired and killed by AOL -- that action was a driver for the U.S. Department of Justice antitrust prosecution against Microsoft. The prosecution had more to do with Microsoft's behavior than Netscape's; apparently, even Microsoft had been convinced that Netscape was a threat and, unfortunately for Microsoft, the company misbehaved unnecessarily to stop a threat that never actually existed in the first place.

In hindsight, while Netscape could have taken some Microsoft business had Netscape even had the inkling of a successful strategy, Netscape couldn't do serious damage to Microsoft because Netscape simply didn't have what it took.

Since that time Microsoft has actually become stronger. Given the new hype that Linux is the next big Microsoft threat, I figured it was time to revisit this subject and put this so-called threat in proper perspective.

The Pillars of Microsoft's Power
Many people assume that the core to Microsoft's power is the monopoly control it has over the operating system and productivity suite. As a result, competitors including IBM, Novell and Sun focus on that software, and fail miserably.

That's because those products aren't Microsoft's strength. The products are a visible representation of Microsoft's power but they aren't a source of power in themselves. If they were, they would have been displaced years ago by something else.

Microsoft's power lies in three groups of people. You can confirm this by going to three events. Those events are the Microsoft PDC, WinHEC, and Tech-Ed conferences. We will look at each pillar in turn.

Pillar 1: Developers
The PDC, or Professional Developers Conference, showcases the work Microsoft does with developers. Microsoft shares product roadmaps and increasingly makes third-party developers part of the decision process for new products and tools.

Microsoft's tools are most important for this group. Developers like to create, and Microsoft concentrates on making the creation process as easy as possible by removing much of the painful gruntwork that goes into making a product.

Most recently, we saw this effort pay off in the blending of Windows and XBox development tools, coupled with vast enhancements to the tools themselves. These enhancements shifted the workload for developers, which had been 80% building the plumbing and 20% building the game, to the opposite -- 20% plumbing and 80% game. This is likely to vastly reduce both the cost and the risk of creating a new title.

Compare Microsoft's relationship with developers with other vendors'. Apple simply doesn't share information anymore, and IBM historically keeps information, like roadmaps, concealed to prevent a timely competitive response.

The Linux community shares everything - unfortunately, they share property they don't actually own (or so SCO alleges). The Linux community often doesn't even seem to know what a road map is. Tool development is haphazard and ease-of-use is an afterthought, if it exists at all.

Still, Linux's greatest threat to Microsoft is in the developer community, because Linux is largely driven by developers.

Pillar 2: Hardware Manufacturers
WinHEC, or the Windows Hardware Engineering Conference, has no real parallel outside of Microsoft. Apple has no reason to share hardware interfaces because they build their own hardware. The same goes for IBM. Linux distribution providers don't seem to have the bandwidth to share with hardware developers, and the hardware groups seem to know that they are on their own with Linux.

Microsoft has, over the last few years, increased its ties with not only PC makers but peripheral manufacturers by communicating detailed reports about hardware conflicts that crash Windows. Improvements to the reliability in the last two versions of Windows come largely as a result of these efforts, which created much better drivers and reduced service costs suffered by these vendors that were clearly eating into everyone's bottom line.

Microsoft will also prototype hardware to give the hardware vendors ideas on where markets may be. Microsoft is currently active with Tablet PCs, Media Center PCs, Portable Media Centers, automotive technologies, smart phones, and classes of embedded devices too numerous to mention here. Microsoft isn't just creating a platform, but it often funds to create the market. Clearly, not all efforts are successful, but none of the other vendors or platforms are doing as much to create markets where third parties can participate as Microsoft is. The one way to win over a hardware vendor is to make money for the vendor and Microsoft seems to understand this better than anyone else, better sometimes than even the hardware vendors themselves.

Pillar 3: IT Buyers
Tech-Ed is targeted at the folks that use Microsoft technology in corporate settings. These are buyers and advocates of Microsoft technology and represent the third and most demanding of the three pillars. Users often don't know what they want individually or as a group and often live tactically focused on current problems without any sense for what they may need one or two years out. They often demand change but often don't recognize that change generally comes at a price, and then they complain loudly when that price becomes due.

This group is cost-focused, but has a conflict of interest when it comes to improving efficiency, because cost-cutting can result in downsizing, which puts their own jobs at risk. Downsizing also results in loss of stature for CIOs, who often measure their own importance by the size of their organizations.

In short, this group wants to do more with less as long as the "less" doesn't mean fewer people.

While many in this group see Linux as a way to take power back from users and recreate the technical priesthood that existed in the mainframe era, most seem to recognize that Linux is simply a lower cost version of Unix, and use Linux appropriately. Much is made of Microsoft's ties to this group being at risk, but feedback from the recent Tech-Ed conference in San Diego, coupled with failures of Linux installations at Home Depot, and problems with high-profile installations in Munich and Thailand, indicate this is just Netscape all over again. Home Depot, after a lot of flash, ended up deploying Windows, Munich is stalled due to compatibility problems, and the kids that got the Linux machines in Thailand ended up putting pirated copies of Windows on them. Linux is beating Microsoft for some customers migrating away from Unix, but little, beyond bad press, is really resulting.

By the way, you may note that we are actually missing a potential pillar and that is the consumers themselves. We'll take this up at a future point when we chat about how a company might go about displacing Microsoft.

This brings us to the end of this month's column. Remember, just because a lot of folks say something is true doesn't mean that it is. The world isn't flat after all, and you actually do need a business plan to be successful, especially if your company's name ends in .com.

Rob Enderle heads the Enderle Group. He has been an external IT analyst since 1993. He is contemplating building an open source-free saferoom in his solar-powered home. He can be reached at [email protected]. Contact the editor of Linux Pipeline at [email protected].