Were the Health Insurance Portability and Accountability Act of 1996 (HIPAA), the sun setting on the U.S. tax code by the year 2002, and 10 years of government surpluses all just a bunch of bureaucratic babble? It seems like there's more and more of this going on lately. These political-hot-buttons-turned-law have essentially come and gone--and nobody noticed.

Does anyone remember that the U.S. Congress passed a law to eliminate the tax code about six years ago? That was supposed to occur this year. What happened? What about HIPAA? It was supposed to provide portable database and EDI transaction for health insurance claims processing and provide accountability. It was due in stages this year and later in 2003. Also, in 1998, the government was trying to figure out ways to spend the 10 years of revenue surpluses they were about to collect. Funny thing, we only had a couple of years of surplus revenue then dropped back into deficit spending.

What does all this have to do with AS/400 and iSeries? Planning. Planning makes a huge difference in our IT budgets and future strategy. One of the key ingredients in the planning process is to use historical patterns as a guide for future behavior or future expectations. For example, suppose that over the last 40 years we have had a slight recession every five years. Then, it's a pretty good idea when doing a 10-year plan to expect two recession cycles during the lifecycle of the plan.

The problem is, we often don't plan like this. Often, we use the new legal methodology for planning by using a precedent, or what I like to call...a pattern of one.

When something happens once, it is not a pattern; it is merely a blip. For example, we had one year of government surpluses and suddenly the government was planning on 10 years of surpluses. Is that smart?

Over the last year and a half, the iSeries market has been flat or slightly dropping. Sure, there were 4,000 new iSeries customers during the last year, but so what? How many iSeries customers have left the iSeries in favor of less-capable platforms? Probably more than 4,000, but I have no data, since IBM doesn't release that information.

So let's say there are roughly 250,000 customers with a total install base of somewhere between 400,000 and 600,000 system units. Some customers have multiple machines; others have upgraded or consolidated and are represented in those figures as multiples. For example, I have personally purchased two AS/400 systems and one iSeries system myself, so I guess I account for three of the units in that 600,000 unit figure.

Back in 1988, there were roughly 25,000 (plus or minus a couple thousand) System/38 systems installed. There were rumors that there were about 200,000 System/36 units installed. So that would mean roughly 225,000 units in all. So, over the last 14 years, the number of installed systems has more than doubled. Yet, over the last year or so, that number has been relatively flat.

So when planning the next several years' budget, do we expect that the iSeries will continue to be a viable solution? More importantly, will OS/400 be a viable solution? I think the answer is yes.

If you look at the long-term history, the number of new installations stalls just prior to IBM announcing a new platform. IBM has indicated that they will merge the pSeries and iSeries sometime in 2003.

So should you plan on being on an OS/400 platform in two to five years? Absolutely. It may be a much less expensive platform at that time, but it will still be OS/400 and all the great features that go with it. In fact, the CPU used for the iSeries and pSeries is also used in the Apple Mac (all models) and in an upcoming consumer home gaming system. When that happens, the cost of manufacturing the PowerPC CPU will drop so substantially that 7 Ghz chips will probably cost less than $100. Since today's high-end chips are only about 1.1 Ghz, a 7 Ghz machine running OS/400 should be quite a spectacle.

Also there's a rumor that IBM will soon port the Mac OS X Server operating system to the iSeries--but will run it in its own partition. Makes sense to me. The Mac OS is already targeting the PowerPC chip, the same chip as in the iSeries box. So it would be logical that, if Mac OS is ported to a partition, then the MS Windows shell that runs on the Mac will also run in that partition, which means Windows will run in a partition, no longer needing the separate Intel-based processor card.

This leads me to my final thought. If IBM is really smart, they'll get Microsoft to port Windows to the PowerPC platform. Then they'll start turning out desktop and Notebook computers with that chip in it. Of course, they'll also have to sell chips to Dell and Gateway and the others. Effectively, this could create another PC boom, and this economy could certainly use that kind of injection. What are you waiting for, IBM?