Feature:Solaris and the Desktop

Shawn T. Amundson has sent us a feature on Solaris and the Desktop. Most of his arguments are cut and paste translatable to Unix as a whole, but in Sun's case, he's quite right. If they want to grow, they need a new desktop. Anyway, hit the link below to see what he has to say about making that happen.

The following is a feature written by Slashdot Reader Shawn T. Amundson

Solaris and the Desktop

by Shawn T. Amundson

Commercially, Sun has been successful at providing a large portion of high-end UNIX servers. They have also done will in workstation market until the recent bombardment by Microsoft. Solaris is an excellent operating system, manageable, and scalable.

It is easy to determine that what is used as a workstation can and will be used as a server. Microsoft knows this very well, and so they don't need to compete even with Sun's servers. They take away the perception that UNIX is the best solution by providing a "better" desktop solution. Windows NT users will think for themselves that NT is also a good server, even if it is not. People like to use what they know, what they can visualize.

If you don't know UNIX, it is hard to visualize that it is better. Visualizing a command prompt makes UNIX seems like it is old, stale, and outdated. We all know all these things are a perception problem, but managers do not necessarily have our insight.

Therefore, it is important that Solaris and the other UNIX vendors fight back in the workstation and very low-priced market or they risk being squeezed off as Windows NT becomes more capable to do previously UNIX tasks. NT doesn't need to do it better, or even do it good - if it can do it, there is competition. Managers have a tendancy to get technical things incorrect, and so the real best solution tends to fade into invisibility.

Without direct competition from Sun, it's market will eventually shrivel up to only the top most servers. And Compaq and Dell are not just going to sit around there either. Of course they rely on Intel coming out with a more scalable processor, but that is under way.

Java is a good step toward this defense against Windows. Java's purpose is clearly to get more applications to run on Sun Workstations, and to bombard Microsoft by a new technique: trivialize the OS. But this doesn't go far enough.

When users log into an NT workstation, the company has usually bought up extras like office and outlook. The user has a feeling thus that these are standard, even if they are not. Sun needs to get competative products *shipped with the OS and already on the desktop*. This is key to success against Microsoft.

CDE is not a good desktop to compete with Microsoft. To compete you have to be better, a lot better. MacOS better. A merger between Sun and Apple would destroy Microsoft. That's not likely to happen, but the reasons why it would are important.

Sun needs a revolutionary desktop with a lot of application shipped with the operating system. Revolutionary doesn't mean difficult to come up with. Any good desktop with the stability of UNIX is near a revolution. Add to that UNIX and X-Window's ability to have multiple people use the machine and that *is* revolutionary, at least to most people. The missing component is a good desktop.

Projects like KDE or GNOME should be mentioned now, because they don't fit the need of what I'm talking about. What I'm talking about has to be fresh (KDE and GNOME are a mish mash of other things like Windows and CDE, at least so far). It has to be designed not from the bottom, but from the top with the consistancy of the MacOS or Windows. That might mean coming up with a new GUI toolkit or morphing an existing one to be unique but good.

Sun's logo: "The Network is the Computer". Think about that. That logo is everything UNIX, and describes how we use UNIX. How about the thought that "The Desktop shows the Network". That's the next step, and one that Microsoft knows about - thus the integration of IE with Windows. It isn't entirely to snuff out Netscape.

Keep in mind that what benefits a UNIX vendor can very directly benefit Linux, and the opposite is true as well. With that in mind, I think Sun should develop the next thing in desktops (or buy it), and work with the free software community to get it on Linux and installed in everyones homes. (Release it under a BSD license, or something similar.) Sun needs to use Linux as a marketing tool; Linux can penetrate the home almost as easy as Windows. All it takes is a lot of advocates, something Linux has in quantity.