Gregory (Scotland Yard detective): "Is there any other point to which you would wish to draw my attention?"
Holmes: "To the curious incident of the dog in the night-time."
Gregory: "The dog did nothing in the night-time."
Holmes: "That was the curious incident."
That comes from Sir Arthur Conan Doyle's Silver Blaze, a Sherlock Holmes short story. I've been plowing through all of Doyle's Sherlock Holmes stories over the past month and was struck by this famous dialogue last week, especially as it pertains to the Linux desktop.
There are, of course, the constant reports of how easy Linux is to install and use on the desktop. Then there are the more pragmatic posts like this one from Steven J. Vaughan-Nichols, calling out a few things Linux needs to succeed on the desktop (device drivers, applications, and power management are his suggestions), despite its otherwise strong credentials.
And yet the dog isn't barking. Few are buying. Why?
It could be that a dearth of applications that consumers really care about (e.g., Microsoft Office) is stifling Linux's voice on the desktop. I buy that.
But I think there's a more subtle problem: no one cares about replicating their existing experience for a slightly smaller price tag. It's simply not worth the risk.
Contrast this with Macs. The Mac is on a tear, gaining market share at a torrid pace. It solves Linux's core application problem by embracing a native version of Microsoft Office, but that is an important precondition to success, and not the determining condition.
No, the Mac is exploding in popularity because it's cool. Look at Apple's Mac ads. They're not about "equal functionality for less." They're about "we're cool and the PC is lame." People are willing to take a chance on cool. They're less likely to take a chance on "good enough and cheaper." No one wants to date you just because you're frugal, but they just might if you're cool.
It's possible that the Linux desktop's future is to appeal to enterprises: boring but cheap works in the enterprise. Sort of. Hence, IBM recently clamored for not-even-close-to-being-sexy vertical applications on Linux as a way to make Linux more relevant on the desktop.
Prediction? It won't work, even if Linux gets such applications. Why? Because enterprises still have to deal with people, and people want to use what's easy, looks nice, etc. I'm not suggesting that Linux doesn't have these attributes: I'm suggesting that consumers simply won't care until the Linux desktop appeals to something more than a few pennies in their wallets.
The Linux desktop is going to have to be much cooler and much better marketed before it's going to resonate with a wider audience. It's not really about being technically better. It's about something more (and less). It's about people wanting to use it, rather than having to use it.
That's what is carrying the Mac. People are looking beyond the Mac's applications dearth (compared with Windows). They're looking beyond its cost. They're looking beyond its flaws. Why? Because it's a beautiful machine that does really cool things like easy integration with their iPods and consumer productivity (iMovie, iPhoto, etc.).The company also has done a great job of marketing Mac security.
Put most simply, the Linux desktop needs much better marketing if it ever hopes to make a splash. Unfortunately, the developers who prize its stability and security are going to be the wrong ones to market its consumer-facing attributes.