Why the IBM Linux desktop will fail

Stan Beer
Sunday, 07 December 2008 15:12

Opinion and Analysis

The answer to why so many enterprises stick with MS Office despite its shortcomings and relatively high cost is simple. It's called lock-in.

It's fine to make the shift to OpenOffice or Lotus for wordprocessing, spreadsheets and presentations if you have never done anything more than basic office productivity tasks. This is not to say OpenOffice and Lotus aren't every bit as powerful and feature rich as MS Office - they obviously are.

The problem is that many if not most medium to large enterprises have years of intellectual property tied up in complicated Excel spreadsheets, Powerpoint presentations, Access databases and Word documents riddled with embedded macros and formulae. In a high percentage of cases, making the transition to a new office productivity suite would be quite problematical and unacceptably expensive.

The second problem with IBM's brave new Microsoft-free world is even more fundamental as it involves shifting users off Windows and onto a Linux virtual desktop running on a server. This of course resonates nicely with IBM's dream of a return to the good old days when all processing was centralised on a large mainframe server.

This is not meant to be a philosophical argument over whether centralised computing is better or worse than distributed processing - whatever takes your fancy. The issue is can you successfuly, easily and most importantly cheaply shift an organisation completely off Microsoft Windows and onto a Linux desktop?

Unless you plan to run Windows in a virtual window as a guest operating system on the Linux server - in which case you're not Microsoft-free - the answer is obviously no.

One of the major reasons Linux desktops - Ubuntu, Suse, Red Hat or any other flavor - have not penetrated the enterprise market is once again that old bugbear lock-in. How one earth are enterprises going to run all their Windows applications on an operating system that doesn't support them?

Linux advocates can espouse the wonders of the Wine project until the cows come home. The fact is after all these years it's as far away from being a real all encompassing solution as ever.

In fact, the only practical way for IBM to accomplish their goal of wooing enterprise users over to the virtual Linux desktop is to embrace Microsoft. Instead of trying to convince users to be Microsoft-free, IBM should be encouraging them to run their "legacy" Microsoft applications on a virtual Windows desktop side by side with their new Linux desktop applications, including Lotus.

Unfortunately for IBM and other Linux advocates, after nearly a decade and a half of trying, they still don't seem to have got the message that in most cases trying to get enterprise Microsoft users to go cold turkey simply doesn't work.