As I write this piece, it's springtime down here in New Zealand, and being a bit of a backyard gardener, I've a dozen or so little seedling trays sitting under glass, soaking up the newly warm sunshine. A few weeks back, I left some young tomatoes, just an inch of stalk and a couple of leaves, out in the open for the night. This, of course, was too much like temptation for the local snails, who promptly made a salad out of them-my tomatoes became just several inches of stalk poking out of the soil.

My second reaction after this discovery (my first reaction I leave to you to surmise) was to ask, "Now just what are these little fellows going to do about this? Do they have some kind of disaster recovery plan, or will they just drop dead on the spot?" So I took them inside, put them back under glass, and watched them.

After a week, to my great surprise, none of the original six were dead. In fact, they didn't look much different-six stalks, a little fatter and a little greener maybe. Still didn't look much like tomatoes. I guess getting half your appendages lopped off does demand a little time for reflection. But after another week passed, they had clearly gone from contemplation to action. One was dead, and five were sprouting new leaves, albeit in the most unusual places. Now, the surviving five are in great shape, with a new growing stem firmly in place and leaves forming just as if nothing had happened. Bear in mind that these humble plants figured out how to reconstruct themselves, how to manage despite that their food producing factories (their leaves) were out of commission, and how to go on to become something without which pasta would be unthinkable. And along the way, they'll also be assembling thousands of tiny encapsulated versions of themselves, all waiting to do the same thing again next year. All this without any assistance from IBM or Bill Gates.

Now, I've seen some pretty smart code and a goodly share of dumb code as well, but compared to what these tomatoes did, nothing I've ever seen even rates. How many systems would manage if a couple of key files got wiped? Or if a program started overwriting where it shouldn't? And what happens if the power drops with a couple of hundred users signed on? I've heard that a lot

of systems can't even handle the fact that our dates will soon start with 20 instead of 19. How dumb is that? And all the while, there's always some "hot" new item out there that will revolutionize the way we work with and use computers. Ten years ago, it was UNIX and C; five years ago, it was CASE tools, C++, and OO. Now, it's Java. I wouldn't say these developments are without much worth, but they need to be seen for what they are. They are simply tools, and, yes, better tools than were available to us in the past. But I don't believe they are the embodiment of the future-they are a means of improving on what we do badly in the present. But you'll still be able to write dumb code in Java, just as you can in RPG II.

Don't think I want to get embroiled in a language war here. My beef isn't with languages; it's with intelligence-or more to the point, the lack of it-on operating systems themselves. Many of us feel pretty comfortable with OS/400, especially those of us who have had to work on other systems. I think it could be a lot better. A simple example to make the point: When I create a file, I go to a lot of trouble to name fields, saying what kind of data they hold, how big these fields are, even how these fields should be presented should someone view them with a query tool. Clearly, when a program opens one of these files, a wealth of information about it is already known to the operating system. Shouldn't I then be able to ask if field DWDATE is six-packed or one of those new-fangled date things? And if it is six digits, shouldn't I be able to convert it into
a date type? This way, I could code all my Year 2000 problems away without having to change a single file. Then, when I get around to changing the file itself, I could do so with full confidence that all my programs would be sufficiently smart to handle it. Yeah, right. Just so much as look
at that file the wrong way and you'll get level checks, decimal data errors, and highly excitable people calling you unpleasant names.

So why is it that a simple plant can cope with major obstacles in the way of its purpose in life, but a multibillion-dollar industry endowed with the best brains of our time is in crisis over a date change? I guess if I knew the answer to that one I could start building the biggest house in Seattle. But not knowing the answer doesn't stop me from looking toward where I think it will come from-and I'm not talking about Rochester or the Sun labs. Natural systems are absolutely packed with the mechanisms for dealing with the unexpected that our computer systems so sorely lack. Maybe on an evolutionary scale, the computer scene as we know it is still at the single-cell level, and developments like the Internet and parallel processing are the first signs of digital amoebae teaming up for that lurch out of the murk onto dry land. Whatever's going on, I expect we will borrow much from the living world in making the electronic one more lifelike- and that we will do this quite unwittingly.

John V. Thompson is a technical consultant to Honda New Zealand. He has worked with a range of IBM systems since 1983 and is currently involved with Year 2000 and new systems development using RPG IV and ILE. He can be contacted by email at This email address is being protected from spambots. You need JavaScript enabled to view it..

Also Read

IBM i Shops Pivot to Battle COVID-19

Energy, Vibe, and Atmosphere

Murphy’s Security Law