When Dell, after exhaustively researching this topic, concluded that it would concentrate on Ubuntu on the desktop, you had one of the most powerful companies in the PC market vote for a winner.

But even Dell is not positioning this Linux as a business solution yet, and is primarily concentrating on the enthusiast market. That will likely change eventually, but I think the problem is not that the product couldn't be used in an enterprise. The problem is dealing with an offering that doesn't generate enough revenue to create a support organization that Dell can trust to either come through with a solution or take ownership of an OS-sourced problem.

The opportunity for Linux has never been greater. Windows XP is aging and Windows Vista demand appears to be declining, largely in the face of a lack of demand generation marketing from Microsoft. Whatever the cause, the combination is creating an opportunity for displacement that is unprecedented in my time covering Windows.

The beneficiaries are Apple, which has had one big apparent win recently (we'll see how the deployment goes), and Linux, with the possible nod going to Ubuntu. But I wonder if "free," as in "free beer," will be its undoing.

The Desktop Risk and the Problem with Free

If you've been in this business as long as I have, you've learned to distrust "free" as in "free beer." (It's interesting that I have to do this every time I talk about free/open source; with anything else people would simply assume I was talking about price.) This is because when vendors offer something up for free, you are generally paying for it, often at a premium, someplace else. Free software means higher-priced hardware (IBM software generally started out free), with the not-too-subtle impact of hardware vendor lock-in.

Even in the consumer space, we've learned that free software can come with malware surprises (I just ran into another Trojan-laced application last weekend). In both instances, we are more comfortable if the source of the income supporting the application is clear.

For Google, arguably the most successful provider of "free" software and services, the source of the revenue is advertising. Part of our cost is that when we search, the first things that come up may have more to do with what the related vendors have paid Google to list than what exactly we were looking for.

But we are making a personal choice with Google. If we screw up, the impact is only on our own time; it won't get us fired. But as a CIO, if we bet the employee's desktop (including the desks of the CFO, COO, SVP Marketing, and CEO -- all of whom have more actual power than we do) and get it wrong, well, the words "surprise early retirement" come to mind. That means the decision has to be very measured.

Free creates a rather large red flag that even folks deploying servers are taking issue with, and there you'd think the risks would be vastly less.

If Not Free, Then SuSE or Red Hat: Diversity Raises Its Ugly Head

If you can conclude that Ubuntu is the winner for the desktop, you have one clear choice to make. If you want Linux that has a clear economic model, you have SuSE, Red Hat and even Linspire in the mix. Diversity on the desktop is a bad thing in large companies because it adds costs that are difficult to forecast accurately, and unfavorable budget variances will result. Often a more expensive product is preferred over a less expensive one if the more expensive product is predictable. This is because executives are held to budgets and variances, positive or negative, and often see problems with negative variances becoming severely career limiting.

CEOs and CFOs like predictability because they must, with some confidence, predict financial results. This is probably one of the reasons Google avoids doing financial predictions. It may also have something to do with its costs creeping upward (though most of that probably has to do with its very rapid growth rate).

In the end, as upset as the market seems to be with Microsoft, it doesn't want to replace one problem with another that could be far worse and, if the Linux platform is going to thrive on the desktop, one distribution needs to be the clear winner.

Dell, much like it is trying to assure Blu-Ray wins the High Definition DVD wars, is trying to assort similar influence here, but I doubt any one vendor will be strong enough to make this happen for Linux.

Can Windows Be Replaced?

Certainly, but it requires that the people who support the platform make some hard choices so that the needs of those who are considering it are met. This is true of Apple as well; different simply isn't enough. Both Apple and Linux have to embrace the practices Microsoft has that IT likes and avoid the practices they don't.

This has been the historic problem in the space; to win in the first place, Microsoft had to make some really difficult choices, like not going into the hardware business and licensing to IBM at cost. Over time, the choices haven't become any easier. For Linux to really move on the desktop, it must have a revenue model strong enough to ensure the survival of the platform chosen, and one platform must be chosen as the standard for the effort to focus development and reduce personal risk.

I also think the unprecedented opportunity to displace Windows is close to maxed out, and will likely start closing with the release of Windows Vista SP1. The Windows team has largely been replaced and Microsoft has started to embrace open source, once again making the hard decisions it needs to make to compete.

Dell remains the company to watch. Its success, or failure, will either accelerate or significantly delay, if not eliminate, the move to put Linux on the desktop.

Wrapping Up

Among Linux, Apple and Microsoft, who eventually looks back at this decade with pride will have a lot to do with the hard choices the principals make and whether they made them in a timely manner. Remember: As the entrenched vendor, it is Microsoft's market to lose, and all it has to do to hold on is to keep its installed base moderately satisfied.

At LinuxWorld this week, ask yourself this: Does anything you see come anywhere close to the kind of sacrifices it generally takes to aggressively move into a market dominated by someone else? Something to think about...