ACCESS was:hierarchical

I like ACCESS, for certain purposes.

For a small amount of data, that is not to be shared widely (disregarding copies), and where
the model is relatively simple and straight forward, and where the ease of using the Query wizards pays for itself, ACCESS is easily the shortest route to rapid productivity.

ACCESS fails as a small scale model of a larger database, because it doesn't illustrate the main points of a large, multi-user, relational, database management product. People who start out with ACCESS nearly always think they know more than they really do when it comes time to learn SQL, data modeling,
database design, and data sharing.

A similar problem happened with the BASIC language, back in the mid to late 1960s. On the Compatible Timesharing system at MIT, a user could crank up BASIC, and write a program interactively, and see some results in minutes, that would have taken weeks using FORTRAN and batch mode development. Also, people who were not programmers could be taught the rudiments of simple programming in an afternoon, instead of going through a semester of difficult, abstract learning process.

There is no doubt in my mind that a lot of good and useful work was done in BASIC that otherwise would not have been done, until VisiCalc came along. There is equally no doubt in my mind that a lot of very bad software has been written by people who learned BASIC in an afternoon, and then proceeded to evolve their way to senior software engineer level, without ever learning the real art of programming from someone who knows.

So the problem with ACCESS, IMO, is not that it's a bad product. It's that, by calling it a "database product" it obscures the existence of a real discipline in database application engineering.

So what to do? The alert learner, after learning ACCESS, will learn where "the wall" is, where the upper limit of practical scalability for ACCESS implementation lies. When they hit the wall, they will be prepared to start over again with a "real" database product, like SQL Server, Oracle, or DB2 (to name a few).

If people are interested, I could give a thumbnail sketch of applications that I've seen using ACCESS,
and why I though ACCESS was a good choice. I'd equally be interested in disasters.

[snip]
>
>>ACCESS is a terrible product
>

[snip] Received on Sat Mar 04 2000 - 00:00:00 CET