Now, select the 20% of your Office users who you would classify as light users. Set them up with the most recent version of OpenOffice.org.  Change the default 'Save As' option to save documents in Microsoft Office formats. This reduced confusion amongst simple users. Change all your business document templates so that they look the same in Microsoft Office as they do in OpenOffice.org.

Setup a single copy of Microsoft Office on a sacrificial PC running VNC, so that when they absolutely must have access to it, these OpenOffice.org users can click an icon and remotely access that copy of Microsoft Office. If you have Citrix or Windows Terminal Services, even better. If you have complex VBA macros, use the new OpenOffice.org macro converter, or the one in StarOffice, to convert these across. You may need to buy a single copy of StarOffice to do this.

See how this integration of these two office suites user-bases gels over the next 6 months. If it does, then move the next 20% of Microsoft Office users. Then the next. Then the fourth 20%. You may find that the remaining 20% of Microsoft Office power users can't be moved across without strenuous effort. So be it. Leave them on Office for the foreseeable future.

Once you have done all these steps, perhaps over the course of 18 months or 3 years, you are now in a position to replace the operating system which underlies the application space, with Linux. If done with some nous, your users will be less affected by the change than they would be in migrating to the next version of Windows. What about costs you ask. What about them? All your migrations to the next versions of Microsoft's software technologies have costs, too. The costs of migrating to open source equivalents may be higher than upgrading to the proprietary incumbent in the first software refresh, but not on the subsequent ones, the costs will be less. Much less.

This means that if you factor in the costs of two Microsoft-to-Microsoft software refreshes, and compare them to a Microsoft-to-open source-then-open source refresh, the open source costs will be lower. And on each subsequent refresh, you will gain even more savings. And even greater flexibility and control. Further, if businesses actually factored in 'migration away from specific technology' costs when calculating TCO, open source would win every time, hands down.

So, where does that leave us? That all depends on you. Are you serious about gaining the cost, flexibility and control leverages that open source offers all businesses? If so, you have a lot of work to do. But know this, if you succeed, you'll be the master of your own IT space for the first time, and the benefits can be immense.

Con Zymaris is CEO of Cybersource, an open solutions company operating since 1991. http://www.cybersource.com.au/ {moscomment}