Comment: What perch for the penguin?

The Linux bandwagon has been gathering pace ever since it was first constructed with four odd wheels and an ageing horse. Its first success was in hosting Web pages, where a combination of old hardware and an open source operating system let IT departments support the then emergent Web at very little cost. Since then Linux has gained pace on the back of the Web's success, and its position is now assured.

It is easy to lose sight of the limitations of any product with such success - even with an open source product. Linux does some things very well. But others not so well. Like it or not, Linux today does not provide a complete alternative to Microsoft. But what I've noticed is that now that it is established in one area, we are starting to see it creep into others...

Take PDAs. Not a very fertile place for Linux to take hold you would think. Especially with Palm and Microsoft already locked in brutal and punishing mortal combat. But Sharp has already shipped a Linux PDA, the SL-5500. It has been with the developers for a while and so arrives with a wide range of applications software. The hope is that because it is an open platform, the open source development community will do the rest. We'll see.

On desktops, Linux is a rare sight, though Sun's StarOffice and its open source cut-down version OpenOffice promises Microsoft Office functionality on a Linux platform. But personal computing is Windows' heartland and I think it will take a lot to topple Microsoft. Don't expect it to happen soon.

In Web servers, as we've seen, Linux has made amazing progress as the Web has itself made amazing progress. Most firms have Linux within their walls serving intranets, or serving their Web sites.

But what about business servers? How does an open source Linux alternative compare with Microsoft's slowly and painfully developed enterprise offering. The answer is that, at least at the lower end of file, print and application serving, Linux is well up to the task. Scale up beyond four-way servers and you'll lose the advantage however, so Linux's strength is at the lower end of the server market.

And support has certainly improved now that a handful of suppliers are focusing on the business market. Red Hat is the best known of these, and as if to prove its business intentions it is working on a version of Linux that will offer-load-balancing cluster support and built-in failover support.

But the question remains, does all this activity represent a real threat to Microsoft in the enterprise? Predictions are tough to make in this market, where it takes 10 years to move corporate IT from one network platform to another. Ten years ago, few people believed Microsoft had a chance of moving into the back-office of the enterprise, so it would take a braver person than me to say a similar revolution couldn't happen with something else.

But I do think that the perceived weaknesses of Microsoft's enterprise offerings - security and reliability - are largely overstated. IT strategists are not going to be driven to anything but a demonstratively better alternative. It takes too long and costs too much to make a move - and the purchase price of the software is just a tiny component.

Have your say: contact IT Week

More IT Week Comments