Guest column: There is some hope for Windows alternative

I believe Mark Twain said, "Everybody talks about the weather but nobody does anything about it." Sometimes I feel that talking about operating systems is about the same. Just two short years ago it seemed like all was said and done in the world of operating systems. For many, the OS battle seemed like a question of whether IBM's OS/2 would smite Microsoft's Windows 95 with pure technology alone. Well, we all know the outcome of that one. Despite the fact that IBM can blame itself for giving up on OS/2, this certainly hasn't stopped many IT professionals in their search for an alternative to Windows.

During the past few years, Unix has succeeded in recapturing the eye of corporate IT departments. And Microsoft's failure to convince large-scale IT shops that Windows NT has the right stuff seems to guarantee a future for non-Windows options. The continued growth of Sun Microsystems and the grass-roots success of Linux prove this theory, at least in the server OS market.

But the desktop belongs to Microsoft. In the past two years we saw the near collapse of Apple, and the rise and fall of network computers, which so far have failed to move beyond anything more than a great marketing plan. Like it or not, we still rely on Microsoft for our desktop OS needs. And why not? As much as people love to hate Windows (3.x, 95, 98, NT, or all of the above), it is still one of the best overall environments, with the greatest support for desktop productivity applications.

I hate to disappoint you Macintosh fans, but in the corporate space, Apple has succeeded only in showing us what it cannot deliver. Remember Copeland? What about Rhapsody? Instead we got Mac OS 8.5. What a letdown. Kudos to Apple for bringing us the iMac, but there isn't much corporate implication there. And until Apple puts out a great OS that supports Intel-compatible platforms, its reach will be limited.

I'm inviting flame mail when I say that Linux isn't ready for the corporate desktop, but let's look at it seriously. Many of you can cite how Linux has solved some of your problems, and it has solved some of mine, too. But it will take more than Netscape Communicator and WordPerfect for me to recommend pushing Linux onto the average desktop.

One of the main problems with Unix desktop OSes is the same one that has plagued Unix for more than 20 years: No one can agree on how to do things. Regardless of which Unix variant you are running, the configuration files and command-line syntax are never consistent from one platform to another. Even Unix administrators will agree this is not ideal.

OSes are like personal information managers for me. I've tried just about all of them, but I always come up short. Lately, there is one desktop OS that I have my eye on -- Be Inc.'s BeOS. Like Linux, BeOS still needs a lot of work, but so far it is the most polished desktop I have seen in a long time. BeOS is the easiest OS to install since, perhaps, MS-DOS 6.22. Overall, BeOS' interface is clean, its performance is good, and its product definitely has a lot of promise.

Be has designed its OS to be a righteous multimedia platform. Designed from scratch, BeOS is unencumbered by the woes of "legacy" support. And Be does not attempt to make legacy compatibility a design goal, but rather produce a new OS for a new generation of hardware. One of the legacy issues that Be does need to address, however, is forward compatibility of applications. Developers now need to recompile applications when users move, for example, from Release 3 to Release 4 of the OS. This should be seamless.

Although Be sees its OS as a complement to Windows or Mac OS, I question that thinking over the long term. There is no reason why BeOS cannot be scaled into a full-fledged Windows alternative. Well, almost no reason: Be still faces the same problems as every other OS company on the market; they need to build momentum through extensive hardware support (even if only for new hardware), and by enabling an extensive battery of polished applications. Because, no matter how good the trap is, you have to bait the mouse with some old-fashioned cheese.

Will BeOS weather the storm? Tell me I'm crazy or tell me I'm right -- just tell me.

Senior Analyst Jeff Symoens (jeff_symoens@infoworld.com) lives on the edge at the InfoWorld Test Centre, where he covers operating systems and other enterprise platform products.

Join the newsletter!

Error: Please check your email address.

More about IBM AustraliaIntelMicrosoftSun Microsystems

Show Comments