Is Linux Desktop at the Crossroads?
Looking Left, Looking Right, Looking Ahead...
Over the last couple of years, the Linux server market has gone mainstream capturing as much as 24 percent of the market, according to IDC, but Linux desktop remains far behind with IDC estimating its market share at less than 3 percent.
Linux has failed to take hold at the desktop level in the same numbers as its server counterparts for a variety of reasons including a dearth of applications and drivers, a lack of interoperability across distros, and inconsistent user interfaces.
But with Microsoft's next generation version of Windows, code-named Longhorn, scheduled for release as much as 12-24 months away, does Linux have a window of opportunity to capture market share? The answer depends on who you ask, but even the most optimistic prognosticators don't see Linux desktop eating into Microsoft's domination of the desktop in any significant way.
Regardless, analyst Stephen O'Grady from Red Monk says his firm thinks that Linux desktop has reached a crucial juncture and it has to prove that it can gain some sort of foothold before Longhorn gets released.
"At the desktop level, the next 12-18 months will be very telling. We've been on the record as saying Longhorn is a very compelling upgrade for a number of reasons at the consumer and enterprise level. Linux needs to get firmly established prior to the introduction of Longhorn. There is a confluence of different factors from the economics, to the maturity of the Linux packages, to delays from Microsoft, and it would be tough to envision a more positive scenario for Linux desktops to pitch themselves in and I think a lot of factors are working in their favor right now," O'Grady says.
Even with these optimistic circumstances, O'Grady is careful to temper his enthusiasm about actual market share.
"We're not saying it will set the world on fire, but the opportunity is there for them to get firm foot hold," O'Grady says.
In fact, IDC projects a 16 percent compound annual growth rate between now and 2007, yet total Linux desktop operating system will only manage to garner 3 percent of the total desktop operating system market share, compared with Microsoft's 95 percent share, which is still growing as well.
Interestingly, IDC's numbers don't include free downloads or copies made and passed along to other users, nor does it include pirated copies of Windows since these types of distributions would be impossible to count.
Instead, IDC surveys only include items shipped and revenue received, but does this skew Linux's market share numbers? Probably not, says Red Monk's O'Grady, even though he says percentages of sale don't necessarily tell the whole story.
"We are on the record of saying that 2004 is going to be a significant year [for Linux desktop sales], but what significant is where some people get lost. It won't amount to much in terms of percentage of sale so there is a certain amount of truth to that," O'Grady says.
Dan Kusnetzky, an analyst at IDC who helped gather the Linux market share numbers says while he is optimistic about Linux's growth within the three percent total parameters, he still thinks many people will be just as happy to stick with Microsoft.
"I would project that Linux has an opportunity in certain segments of the desktop operating system market and in those segments could gains share. There are other segments where it would be rather unlikely that people would switch," Kusnetzky says.
O'Grady thinks most large companies will at least try Linux desktop somewhere in the company.
"I think you will see the Fortune 100 will try it for some of their constituent. That doesn't mean it will run on every desktop, but it will establish a foot hold with folks in the data center being most notable, and that to us is a significant gain," O'Grady says.
O'Grady says one place he foresees a role for Linux is in environments that use thin clients and the underlying operating system doesn't really matter.
"Linux is a good choice in environments where users are using portals anyway. What they are doing is transitioning the workforce into browser based applications that can be accessed from Web browser. Once you take out fat /rich client out of it, it's no longer a binary decision, one OS is not always better than the other," O'Grady says.
IDC's Kusnetzky thinks Linux will be a tougher sell in the consumer market where people buy based on word of mouth and there's not enough presence there yet.
"Most people are not interested in computers for their own sake. If they are a consumer, they have a tendency to go with whatever their friends bought or whatever the kids have in school so the kids can work do their homework at home," Kusnetzky says.
A bigger issue, according to Kusnetzky, is availability of key consumer programs like Quicken.
"If they are looking for an application to satisfy certain requirements such as Quicken, they will only buy it on a platform where it's available. Since it's not available on Linux, they wouldn't consider Linux," Kusnetzky says.
In spite of this, Kusnetzky still thinks that some consumers may be attracted to Linux, just not in large enough numbers to make a significant impact on Microsoft market share.
"There is a group of typical consumers who use the machine to get Internet and email, and Linux would serve their needs quite well. As HP and Dell offer Linux systems, Linux could gain [some] market share," Kusnetzky says.
In the end, Linux has a role to play, but it's unlikely at least in the short term that it will make any meaningful inroads in the desktop market. There will be places where it will be appropriate, but it doesn't appear that even the brightest prediction see it giving Microsoft anything to worry about in the desktop market.
On the other hand, Stephen O'Grady isn't sure that really matters.
"We take exception when people want to force one or the other (Linux or Windows) down your throat. Enterprises [and consumers] need to look at both, both have merit," O'Grady says.