The end of civilization as we know it

When civilization ends, it won’t be because of nuclear holocaust, viral outbreaks or flooding caused by global warming. It will be because of the telephone companies.

Actually, the telephone companies are merely part of a much bigger issue: the systems and processes that run civilization will become too complex to be understood even by systems that are designed to attempt to control them and, therefore, civilization will become unmanageable.

Infrastructure will fail, trains and planes will not run, factories will be without raw materials, and power and water supplies will dry up. A long, sullen silence will fall across the Western world, broken only by the sound of people tending their vegetable gardens and scholars scratching away at smug treatises on the value of structured systems design.

Be that as it may, back to my evidence – the behaviour of telephone companies. The sheer level of complexity in simply ordering any service beyond a single line (which also excludes any attempt to transfer service from another residence) is mind-boggling.

Before you accuse me of needlessly obsessing over my travails with Pacific Bell, perhaps I should explain why I am having so much interaction with the telephone company in the first place.

The answer is simple – we’ve moved again. Yep, 20 months and 10 days since we moved into the previous house, the Gibbs Institute has relocated to our dream house.

Anyway, back to complexity. You see, the problem is that as systems get more intricate, we lose our ability to understand them as a whole, and when they become really convoluted, even the subsystems are a nightmare.

Sure, for most computer and network systems we understand the parts, but anyone who has dealt with the likes of SAP or Baan software will know that even well-defined applications (that is, well-defined in terms of their scope and purpose) have a degree of intractable complexity that is astounding.

And when you connect data flowing to and from complex software linked to real-world processes, the probability for chaos, disaster and really, really ticked-off customers goes from a slim chance to a dead certainty.

I could go on at some length about the number of calls I made to get my phones moved and the sheer waste of time involved, but the detail isn’t really that interesting. What is interesting is the implication of thousands of people all going through similar exercises every day. The countless hours spent listening to: “We’re sorry, but due to unexpectedly high call volume all of our operators are busy having lunch or otherwise ignoring your pathetic call.” (OK, I admit it, I made that last bit up.)

And then when you get to an operator or whatever they call them (as in “customer care advocate”…what crap!) you find that the information they need if you call on the weekend is different from what they require on a weekday. That is basically because it is a different office, and despite spending an hour on the phone ordering six lines the engineer only turns up with two and…

The same nonsense is typical of airlines, the gas company, credit card companies – the list is endless. And it gets worse every year.

So how does this relate to you? Well, other than you could just as easily be a victim as anyone else, because you are in the IT industry you are also part of the problem. If you don’t keep an eye on how your systems interact with other companies’ systems as well as with people, you may well have to accept your share of the blame when civilization comes crashing down. And if you are with a telephone company, I’d practice saying something like, “No, I left before things went wrong.”

Gibbs is a contributing editor to Network World (U.S.) . Send complicated comments to