- The Washington Times - Thursday, June 1, 2006

I find it interesting to talk to a friend who does computer support for businesses, much of it involving databases. He spends a fair amount of time being exasperated, not by the weaknesses of the technology but by mistakes that companies make in using it.

The price of my friend’s candor was anonymity, because his comments were sure to offend some valued customers. His thoughts:

“First, remember that I work with small companies, with 20 or 30 employees. General Motors has professional IT staffs who know what they are doing.

“With databases, a company has three choices: Set up a database itself, buy commercial software, or hire someone — me, for example — to design a custom database. Usually it is an enormous mistake to try to do it yourself. People try, and end up doing things that look more like spreadsheets. Then Joe Blow, who did it, leaves, and no one else understands it, so something critical isn’t accessible. It can be a disaster.

“A fundamental problem with home-brew databases is that people usually think in two dimensions, by which I mean in terms of rows and columns. Unless you are talking about something trivial, there are multiple pieces that have to work together, with complex relations. Most people, even if they know how to use Access, can’t set up relations. You need to think in several dimensions.

“Often, they have someone they think is qualified. The biggest problem I face, whether in databases or network management, is people who think they know more than they do. It’s an ego question. They get in way over their heads. Usually the result, aside from being clunky, isn’t useable by others, and just barely functions.

“Ego is a big problem. Often, after I design them a database, they want to change things on their own, and they don’t know how. If they try, they don’t have me as a consultant. When I build a system, I lock it down and don’t let them mess with it, period. In fact, I don’t give them the source code, so they don’t have a choice.

“Sometimes they get someone to build them a custom database, but it still doesn’t turn out well. The somebody, perhaps even somebody very good, doesn’t know the industry. And the commonest mistake programmers make is doing what the customer says he wants, not what he needs.

“Usually in a company you are dealing with someone who really doesn’t understand what is really important. They don’t know their own jobs that well, though you can’t tell them so. You have to find out what they really do, how they need to do it, and systematize it. You have to be diplomatic: Someone who has been in a business for 20 years doesn’t want to be told that he hasn’t figured it out very well.

“Another common problem is that companies buy a solution that works for the moment, but their needs change. Suppose that you are a trade association. You can buy, say, a membership database, some of which are really good, but really expensive. An upgrade can cost $25k.

“But your business changes. You can’t modify commercial packages much. You tweak and tweak. Then finally you have a custom system built. It’s the difference between a tailored suit and one off the rack. The real advantage that most get addicted to is that it works the way they work. You can’t do that with commercial packages. Instead, they give you a clunky workaround.”

Always, he says, the technology is there and works. The trick is getting managers, often egotistical, to listen.

“For some reason, computers are a vanity issue. Everybody’s an expert. Well, they aren’t.”

Copyright © 2018 The Washington Times, LLC. Click here for reprint permission.

The Washington Times Comment Policy

The Washington Times is switching its third-party commenting system from Disqus to Spot.IM. You will need to either create an account with Spot.im or if you wish to use your Disqus account look under the Conversation for the link "Have a Disqus Account?". Please read our Comment Policy before commenting.


Click to Read More and View Comments

Click to Hide