Why Linux On The Desktop Is Toast?

Linux succeeding as a desktop operating system as been discussed forever. I’m a Linux software developer by trade, use Linux as my desktop at work and home, but I see the future for it on the desktop to be quite bleak. I discuss the major reasons why below.

Before we discuss this issue, we need to determine what a Linux desktop operating system is. Linux distributions such as Redhat, Suse, and Ubuntu are clearly Linux desktop OSes. However, what about Chrome OS? Google’s upcoming operating system utilizes the Linux kernel, however it has a much different user interface than the typical Gnome or KDE environment that comes with most Linux distributions. Although it is not out yet, it also appears that they will not be offering common Linux utilities such as Gimp and OpenOffice, favoring their own web based applications instead.

For the purposes of this discussion, I am doing to define a Linux desktop OS as an operating system that meet two criteria. One, changes and modifications to the components, both user space and kernel space, are actively shared amongst all the Linux desktop distributions. Two, it is actively advertised and described as Linux. In my opinion, a “Linux desktop” refers to a common set of open source components and the philosophy of open source sharing, and thus is advertised that way. While Chrome OS utilizes the Linux kernel and open-source, and contributes back, my feeling is that Google is doing their own thing. They are intending to make Chrome OS its own identity, that isn’t identifiable with other Linux distributions. For that reason, I’m not going to consider Chrome OS as a Linux desktop operating system. Another way to think about it, how many times have you heard Android referred to as a “Linux OS”? Or how many times have you heard that Apple products run a “Mach kernel OS”?

I will ignore the question of what market share Linux must obtain to call it a success on the desktop, since opinions will vary greatly. At the minimum, all Linux advocates would agree more market share is required for it to be a success.

Linux Can No Longer Compete on Price

At the time of this writing, the cheapest Windows netbook that Dell sells is $300. Dell’s equivalent Linux Ubuntu netbook is listed at $300 dollars. The cheapest desktop offered by both Dell and HP is $300 dollars. For desktops, I have yet to find an online retailer with a comparable Linux desktop for atleast the same price. While this data is limited, I hope these numbers illustrate a simple point. The advantage that Linux once had on price has evaporated. This has happened for a variety of reasons, including more competitive pricing, but one of the major advantages Linux had is now gone.

I recall an article I once read on disruptive technologies. In it, Henry Blodget states, “Disruptive technologies take advantage of a new manufacturing/business process or technology to provide a cheaper, more convenient, simpler solution that meets the needs of the low end of the market. Low-end users don’t need all the features in the Incumbent’s product, so they rapidly adopt the simpler solution.”

Linux on the desktop was expected to be this disruptive technology. The old saying was, “If a grandma needs a computer just to check e-mail from her grandkids, there’s no reason for her to pay more money for a PC.” Well, grandma no longer needs to pay more money, so why would a person unfamiliar with Linux now buy it? Grandma has no incentive to switch.

Granted, when you add other applications such as MS Office, a PC will be more expensive. At the time of this writing, a home version of MS Office is listed for $150 dollars, and can be found for $100 on Amazon. It’s a far cry from the $300-$400 dollars it once cost. The key point is the price gap is significantly shrunk.

Linux Must Compete On Features

The flip side of the previous point is that Linux must now compete on features and functionality to succeed, price is a much smaller competitive advantage for Linux. Microsoft and Apple are aggressively developing better experiences on their operating systems and making better applications that run on it. It does not appear Linux has the resources to compete. Google, perhaps Linux’s biggest supporter from a huge tech company, is the one company that could perhaps push that forward. However, as I mentioned earlier, it appears Google has limited interest in supporting a user interface or Linux applications outside of their cloud.

There are Too Many Linux Distributions

While open source is one of Linux’s greatest attributes, it’s also one of it’s greatest flaws. With so many different distributions of Linux, it ultimately makes it more difficult and costly for software developers to develop software for all flavors of Linux. In addition, individual customers may modifying their Linux distributions for their own internal needs, which simply leads to more portability problems.

This may be the smallest issue of all the ones I’ve mentioned in this list. If one of the distributions can become the clear winner over the other on Linux desktops, this problem could dissipate.

Third Party Applications

Ultimately, for the Linux desktop to succeed, major third party applications must be ported over to it. Classically, people discuss video games as the major limiting “killer app” that must exist on Linux. However, other major software from Adobe and Oracle must also be ported. This single handedly will prevent many customers from switching to Linux. There are multiple issues why this exists, such as the fragmentation problem discussed above. However, I believe the issue at its core is just a chicken and the egg problem. Developers will be unwilling to port to Linux as long as it has such a small market share. However, in order to get people to switch to Linux, more developers need to support Linux.

OS Consolidation

The question is, how many desktop operating systems can the market have? Historically, the answer is not a lot. The need for interoperation of systems will ultimately limit the number of operating systems that can exist. Software developers want to have their software work for as many people as they can, while consumers want familiarity. With Google entering the desktop OS war, I am reluctant to believe that Redhat, Suse, or Ubuntu are going to be able to make inroads in an even tougher environment. I doubt they will become the next BeOS or OS/2, but Google’s entry into the market won’t help.

Summary

This article does not make the point that Linux is dead on the desktop forever. It’s only for the relatively foreseeable future. However, the above points illustrate that Linux has a lot to overcome to become a success, and nothing over the past few years has made it easier or cheaper to accomplish.