There are more or less valid reasons for this attitude.
- It's going to me a one time expense which is harder to take than the steady bleeding with Windows
- The bleeding isn't that bad yet, because licensing and upgrade enforcement hasn't taken effect yet. My clients are split between Office 97 and Office 2000 with no intentions to upgrade. Very few have any XP except on a few new notebooks. Worms have just started to become a significant expense - not yet fully digested.
- It's seen as high risk. Spend the money and it doesn't work and you have to go back to Windows anyway - more expense.
- Resistance from employees is extreme. Irrational, but extreme. Just suggesting an alternative to Outlook sends the sales/marketing people on the warpath, and they have tremendous influence. Owners just don't want the internal strife.
- Linux isn't ready because critical specialty applications aren't available yet. Does Linux have an equivalent of Act! or Maximizer? Not that I've seen.
Linux must enter the small business market in a low profile way. I've successfully deployed a number of Linux servers, and resistance from software vendors and consultants seems to be declining. Beyond that, it's going to have to come in as a task specific item.
Right now, I think it's more important to move small business people to OpenOffice and Mozilla, so when the disaster does come, or licensing costs become unbearable, the transition will be a lot easier. If it does come, I've made my case and my clients know where to turn and who not to blame.
As for OS/2, it only runs DOS/Win3.1 software - and with some restrictions. DOS worms are pretty few and far between right now, and the last DOS boot sector infector I've seen was at least 4 years ago.
Thanks for your comments on patching. I'm going to incorporate at least the one on patches being impossible to back out.