X
Business

Sun Linux PCs--it might just work

Sun isn't the first company to hawk a Linux desktop PC, but Larry Seltzer thinks that with its considerable enterprise clout, Sun may actually succeed where others have failed.
Written by Larry Seltzer, Contributor
There is a sense among many that success for Linux on the desktop is inevitable.

True, there have been many unsuccessful attempts at Linux-based desktop computers in the past, but (so the argument goes) as the operating system inexorably improves, whatever deficiencies may exist will be remedied. Come that day, buyers will see the advantages in openness and cost and then they will switch.

I don't actually agree with this scenario, but there is some reasonable analysis to do on the markets in which Linux might be more or less successful. This is why I'm very interested in Sun's recent announcement of a client/server environment using Linux desktops. In many ways it is an environment of maximum advantage for Linux. I think it might be a real put up or shut up scenario for desktop Linux.

Two things are most important in making the ground fertile for Linux with this as-yet vaguely defined product: It's strictly enterprise-focused and it's from a real and respectable company. I have no doubt that it's only in the enterprise (or some other market, such as government, with large and competent IT staffs) that Linux on the desktop could hope to gain any traction. And while Sun has laid a few desktop eggs in the past (JavaStation is my favorite), it also has many respected, quality products. Sun will get more serious consideration than, for instance, VA Linux did.

Of course, Sun isn't the only respectable company that's been down this road. IBM has tried it before (and also tried the Java NC route--didn't work for IBM, either), and you don't get more respectable than IBM. But as loud as IBM has been about Linux, it never made a big deal about the desktop angle. Turns out, neither does Sun; it's the media that has focused on the Linux part of this solution (one day I ought to look more carefully at the rest of it). Sun views Linux simply as the "inexpensive commodity desktop" part of a larger solution where the real Sun value is in the server and Java portions.

Devaluing the desktop is central to Sun's strategy, and there's a lot to be said for this approach, which is why I think it might be well received. Even with Windows-based networks I've always thought it's a good administrative idea for an enterprise to centralize things and generally to make desktop systems as replaceable as possible.

The Sun One software approach is to create network applications, not desktop applications. The fact that the proposed "Open Desktop Client" runs on Sun Linux is almost beside the point, but it does lower the system cost. (At least one can argue that it does; you'd need to see real client numbers to know.)

So the desktop in Sun's new "100 User Solution" is just a means to an end--specifically, a secure low-cost network. The fact that the clients are sold in bunches of 100 along with the Java Card, server, and other network infrastructure, makes them solidly focused on the enterprise.

You've always been able to build this sort of solution with desktop Linux clients, but it hasn't been a mainstream thing. Linux companies and Linux advocates focus on the Linux part of the solution. Sun would have you focus on Java for the solution instead. In a way, this is not inherently better, but compared to Linux, Java is a more focused architecture for secure network solutions.

Sun should be able to leverage some of its credibility with server-based, enterprise Java to sell some of these systems. I agree with analyst commentary I've seen on this: There are a lot of large enterprises that would benefit from a solution like this.

If you're a Linux advocate looking for the Killer Linux Desktop Application (KLDA), this isn't it. These machines aren't meant to be power-user desktops, they're meant to be locked-down systems. The KLDA is probably a mirage though, and I doubt that companies like Red Hat and Mandrake will ever make a product that can compete seriously for the mainstream desktop. But when the desktop operating system is de-emphasized, Linux competes more favorably. And Sun seems to have the right idea about de-emphasizing the desktop.

Do you think that success is inevitable for Sun's desktop Linux? TalkBack below or e-mail us with your thoughts.

Editorial standards