I’ve seen several publications as of late harkening to an Apocalypse of the Desktop – portending the doom of Windows 8 before it’s even made one retail sale. Probably not on that big Mayan wheel that everyone was freaking out about, but it’s been trending the closer we get to Microsoft’s late October release of Win8. What’s more interesting is many of these same authors suggest a resurgence of Linux as a viable desktop alternative. I honestly don’t think that this is going to be the case, and that these claims are about as reliable as those indicating that 12/21/12 is going to see a cataclysmic end to life as we know it. Here’s why:
1. Linux is still not consumer or enterprise ready.
Much as I like (and use) Mint Cinnamon and Ubuntu 12.1, I still cannot see these surpassing Windows as a desktop OS for use outside of very small pockets of Linux enthusiasts, tech-savvy SMB’s, developers, or perhaps an unsuspecting consumer. OEM adoption of Linux will remain low or nonexistent until the distros can come up with a coordinated effort or consortium on things like drivers, GNOME vs. KDE, etc. Keep in mind that for things like GPU’s, etc. drivers for Linux distros are still not supplied by the chip or component makers, but instead are supplied by the Linux community. This isn’t going to be an acceptable model for any sort of widespread enterprise or consumer adoption. The distros themselves also lend themselves to a six to eight month refresh cycle. That’s way too often for consumers or businesses to deal with – it’s far too risky. Support would be yet another reason this won’t happen. I’ve heard the argument that OEM’s wouldn’t have to incur and pass on the Microsoft licensing costs to the consumer with Linux, making the end products cheaper. I contend that’s not true once you factor in the support, maintenance, and training costs associated with completely changing platforms. If the claim for the death of Windows 8 is the new UI and no start button, why would anyone assume that a completely different UI with a Linux distro would make that any better than learning Windows 8? Skins are the answer, you say? Well, there’s already 3rd party Windows developers waiting to publish a shiny new Start Button and “Classic” style UI skins for Windows 8 too.
2. Applications and Infrastructure
If anything, businesses and consumers will stick with Win 7 or even WinXP as long as they can, even if they decide not to move to Windows 8. They will not shift wholesale to a Linux desktop environment. Investments in applications and maintenance/assurance programs that are currently Windows-based will likely not change. The R&D costs associated with researching, evaluating, piloting, and implementing new applications for a Linux environment would make the effort cost prohibitive, or in some cases technically prohibitive (no application equivalents) for many environments. This will hold true for enterprises with multi-million dollar Microsoft or Microsoft-based software deployments as well as for the home consumer with a $495 investment in Office 2010.
Infrastructure is also built for Microsoft clients in most corporate environments as well. You’re not going to displace Active Directory or Exchange overnight. It’s just not going to happen. You’re not going to see massive adoptions of Linux-based VDI or application virtualization/presentation just to protest a UI that has yet to see adoption (or failure).
3. It’s more than just the desktops
Enter consumerization. Many of us are already in the weeds trying to strategize and architect ways to securely and efficiently incorporate BYO programs, SaaS applications, and personal/enterprise cloud service integrations. Having to do this at the desktop level would introduce a huge effort that I just don’t think would be cost effective for most organizations with more than 50 desktops. The cost of R&D for such a project in most cases would make it prohibitive for large-scale deployment. It’s not just designing the desktop build and application delivery strategy, it’s the soft costs of R&D, testing, training, management and support, and all the other man-hour costs associated with such a project. Those of you who’ve had to deal with app delivery with various flavors and screen resolutions of Android know what is entailed in this type of an effort.
I’m not saying that this is not a good idea, or an option, but I’d definitely think long and hard about the position of Linux desktops in the enterprise workspace currently and how expensive and hard it really would be for many users. Will this change over time? Quite possibly. Younger workers today are much less tied to a familiar platform experience than those of us who have only worked with Windows variants for our entire careers. Thank mobility for that, and it honestly could be the foundation upon which this could become a possibility in the future. I’ve been using Windows 8 since the beta, and I have been reserved in my commentary as I’m inherently biased. The true test of adoption and acceptance will come from business IT and increasingly for consumers after the release of Win8, Surface, and Windows Phone 8. Microsoft’s trying to build that unified ecosystem to match Apple and Google/Android. To a lesser extent it will come from vendors like Citrix, VMWare, and others who will supply the bridging to applications. Microsoft’s licensing and SPLA restrictions on desktop licensing for the service provider market might have a significant effect also. Many of us who actively protested the anti-hosted VDI restrictions have already looked at ways to deliver Linux desktops as an alternative. In summary, we’ll have to wait and see. A number of factors could make a mainstream Linux desktop for some use cases (I’m thinking SMB’s as a ripe proving ground and incubator) to start, and perhaps greater adoption if it can be “done and spun” right. Like it or not, marketing’s going to have to be a huge part of it as well to get Linux out of the techie market and into mainstream view by consumers and CIO’s. Time will tell. Just don’t expect a “desktopocaplypse” in the near future. I don’t think the end times have come for Windows desktops quite yet, but there’s no doubt that Linux could gain ground depending on how the desktop landscape changes after Windows 8 is released.