Businesses which have to spend time and money correcting year 2000 issues should make the effort worthwhile and migrate to a network computing mode while they're doing it, says IBM's NC supremo, Bob Dies.
Cost should not be regarded as the main factor in going to network computers in the enterprise, says Dies.
"The one you always hear about is the lower cost of ownership," says Dies. "But really, the thing businesses should be looking at is the speed of deployment of applications, especially new applications."
Dies acknowledges that year 2000 fixes can be expensive. "So why not set up your enterprise system so it will last substantially longer? Why not get into the network computing mode while you're doing the work?"
NCs offer business a stable way to migrate from legacy applications, says Dies, because the same devices can function as terminal emulators for existing applications, and later run new Java applications as they become available.
"IBM and Lotus will be spending millions of dollars on Java development over the next few years - but we also know that many existing applications are going to be around for a long time, because there just isn't time to rewrite all your applications."
Dies says speed of deployment hit home with IBM after it bought Lotus. "After we paid $US3 billion for Lotus, the chairman stood up and said, well, for that money we'd better standardise on Notes. We did - and it's taken us 18 months to two years to roll out Notes globally, because you have to manage every desktop. It would have been a different story if we had already standardised on a network computing model. After all, I have a NetStation now, and I still have exactly the same access to my productivity applications.
"In fact, a network computing model makes it more economical to make available all the productivity tools your employees might need. You don't have to weigh up the cost of putting a more powerful PC on that employee's desktop, because the application is on the server. You will often find, too, that your NC runs those applications faster than a highly-spec'ed PC, because it is co-processing with the server."