Recess is over

Last month we came to the conclusion that we’re all kids at heart, and the real reason we’re in IT is so we can play with all these wonderful toys and gizmos. Fair enough. We can make a strong case that “playing” is just another word for research. Provided, of course, that when we’re finished playing, we put what we’ve learned to good honest work.

After more than two decades in IT, I’ve come to the inescapable conclusion that, with some notable exceptions, the IT industry is guilty of perpetual play. We’d rather play with something new, than apply what we’ve learned. We’d rather sell a client a new technology, than teach them how to use what they’ve already bought. That we consider “development” more respectful than “maintenance.”

The truth of this accusation is found, in of all places, humour… An optimist looks at a hard disk and says “It’s only half full!” A pessimist looks at the same hard disk and exclaims, “It’s already half empty!” Then the computer geek looks at it, and grinning eagerly, says, “It’s time to upgrade!”

This push to upgrade is costly. One organization upgrades all their 20,000 PCs every 2 years. What’s their business case? They don’t have one. IT has convinced business management that they “need” to do this, and management, trusting their IT advisors blindly, sign the cheques. That’s an awful lot of money to satisfy the need to “play” with the latest and greatest.

The vast majority of users, those that use mainly E-mail, word processors and spreadsheets, have no need for the obscene amount of processing power at their desks. I’m writing this article on a 233MHz computer using a word processor released in 1995. As I hunt and peck my way through the article, the unused computer cycles (read: most of those 233MHz) are reading my regular e-mail every minute and simultaneously searching for e-mail, as in extraterrestrial mail, from other planets, with the SETI@Home application.

Do I really need to buy a 1GHz machine? I only get so much e-mail… even if I do include the rest of the universe in my search.

There are obviously limits to “not upgrading.” If I were still on my old North Star Horizon, a Z80A based computer, or even a 286 PC, then I’d be incapable of communicating with most of the world in any reasonable manner. An upgrade is mandatory in either of these situations.

So? That leaves us with a question. How do we decide that one upgrade is mandatory and the other is frivolous? The wrong answer will incur a loss of potential productivity or an unnecessary expense.

In deciding whether or not an upgrade is necessary, there are two different types of questions we have to answer. The first is the easiest. Does the upgrade enable me to do something I cannot currently do, that I need to do? If the answer is “Yes!” (and assuming we haven’t confused “need” with “want”) then I open the chequebook and upgrade.

The next question is more complicated and therefore more difficult to answer. Does the upgrade guarantee a payback, over a reasonable amount of time (whatever you chose to mean by that), that covers the cost of the upgrade, including training and the inevitable loss of productivity during the transition period?

It’s obvious that this question will mean different things to different organizations, but it’s better than the simplistic notion that we upgrade because “everyone else is upgrading” or because “it’s kewl.”

A written statement on upgrade justification isn’t a straitjacket, it’s just a way of placing some checks and balances on the never-ending upgrade process.

Now having said all that, I need to upgrade this old klunker… any recommendations?

de Jager is a well-known, Luddite apologist, keynote speaker/consultant on issues relating to change and the management of technology. You can contact him at