Reprint from: Les Pinterís DataBase Journal Vol 1, No 6 / November,1998

Economics 101

Years ago, Chrysler built a kit that could be used to modify the carburetors of its diesel engines, so that trucks got better gas mileage. It cost over 500 dollars and involved a hefty mechanic bill as well. Mechanics loved to sell and install it, so everyone was happy. Except the users.

The military were especially interested in using computer technology to save money, so they had Chrysler design and build a computerbased module that cost twenty dollars and did the same thing that the five hundred dollar kit that took a day to install did. But they werenít required to sell it to the public so they didnít.

Finally, their largest fleet customers approached them with a proposition: Make a little modification to the circuit board so that it would boot up in such a way as to pass the EPA DOT pollution tests. Then , gradually, over the next hour, have it reconfigure itself to minimize fuel consumption and pollute like a chimney. They did so, and the trucking industry bought tens of thousands of them and laughed at the American people.

Chrysler just settled with DOT for a billion dollars. Chryslerís pricing policy wasnít illegal. Itís perfectly legal to have two products, one of which is far better, faster and cheaper than the other, and not tell your customers about it. What was illegal was the way that they went about it. It wasnít until their trade practices hurt American consumers that the government stepped in. Chrysler didnít give a damn how much more it cost their customers, or how much it cost the American people. They just wanted to increase their share price.


Microsoft sells two database technologies. One Ė FoxPro Ė is cheap and fast, and has no incremental cost. You pay for it one time. The other Ė VB with SQL server Ė costs by the seat and runs considerably slower. FoxPro users can also use SQL, but FoxPro users donít migrate to SQL Server - not even when they should. They donít perceive the need to do so, and often canít even be talked into it. But give users no alternative and theyíll go along.

Give them a language that chokes when their local tables reach fifty thousand records, and offer them SQL as the only way to salvage their development investment. Theyíll come over. Theyíll have to. I donít know whether itís illegal or not to talk your own customers into spending twenty thousand dollars for capabilities that they could have gotten for five hundred. I donít know if itís illegal to recommend a technology that represents massive overkill for most database projects. Caveat emptor. But it seems wrong. Having higher costs isnít necessarily a bad thing for a company. Your competitors simply have to bear the same costs. Being forced to move from a less expensive technology to a more expensive one falls equally on all businesses. If your programmers decide to use MS SQL, it only produces economic disadvantage if your competitors figure out that thereís something better and cheaper. If they do as you do, then your costs and those of your competitors go up by the same amount, and no one in your industry can sell their product more cheaply due to lower costs. No one loses. Except the American consumer.

What does it feel like to be the only programmer in a corporate IT department who recommends FoxPro? Remember not being part of the popular crowd in high school? Remember the power of slander by innuendo?

ďIf itís so good, why doesnít Microsoft advertise it? They must be ashamed of it.Ē Nothing could be further from the truth. Microsoft knows full well how good FoxPro is. MS SQL sales stand to be reduced by billions of dollars. And now you know the rest of the story.

Weíre the key to this strategy of benign neglect. With the Justice Department looking on, Microsoft canít kill FoxPro. But we can. If we buckle to innuendo and the campaign of silence, Microsoft wins. We, and our customers, and their customers, are the losers. By bowing our heads and accepting defeat, we allow might to make right. I know itís hard to stand up to the incrowd. Iíve been asked not to raise this issue, but I intend to stand up and be counted. I hope that you will, too.

Since Microsoft stopped advertising FoxPro, Iíve had a standing offer to compete publicly against anyone they can name in building a sample application. No takers. So try it yourself. Take any application, write it in both languages and compare the development effort; thatís one. Then dump in a hundred thousand records and run them both, comparing performance. Thatís two. Now, figure out how much it will cost to deploy your application on a network using SQL Server - the only cure for the glacially slow performance of large MDB files - paying by the seat for MS SQL. And add in the cost of a Database Administrator, unless SQL 7.0 saves you that $100,000/ year hidden cost. And there are others.

Itís not a contest; itís an IQ test.