Month Year Issue

August 1998 Issue: 4

Journal of Conceptual Modeling
www.inconcept.com/jcm

Editor's Notes: GUIe
by Scot A. Becker

Miss Me?

In case you missed last month's Editors Note's (or you never read this section), the JCM is now being published every other month. I surprised no one sent me any mail in mid-July wondering when the next issue was coming out. I take that to mean that our millions of readers read every page of this fine publication and knew when the next issue was coming out. Well, however it worked out, we're glad you're here (and read this issue slowly because the next issue will be out in October).

Welcome

This issue brings another author to the JCM. I'd like to take the chance to welcome and thank Dennis Corbo for his excellent contribution. While, of course, I appreciate the contributions of our regular authors, it's great to see new faces and get input from more of our colleagues. Want to be famous like Dennis? All you have to do is contribute. See the author's information section or drop me a line for more details.

The Rante Section

I have this nagging urge to throwe extraneous vowels all over the place. You know what I meane. I want my words to looke quaint and olde fashioned.

Then again, maybe the East Coast is starting to get into my head. They throw extra e's all over the place out here (Olde, Shoppe, etc.) To top it off, the folks out here probably made fun of Dan Quayle out here for his spelling of potato(e) too. The man probably just came out of Harrisburg; it can happen to you too. On a positive note (before I get flame male from the nor'east) I'm only poking fun; Minnesota does some goofy stuff too.

What does this have to do with conceptual modeling you ask? Well, not a lot, but I don't have much to rant about this month so you'll have to work with me on this one. What can I do? Visio's been quiet lately (as they say in the old westerns, too quiet), the database world is remaining pretty stable (aside from the mundane but hyped-nonetheless advances in writing a highly de-normalized Java-compliant web-enabled database application for Linux, as if anyone cares), and my upgrade to Window's 98 went smoothly. What can I complain about? To top it off, I'm still on the project with all that…ahem…fine…Oracle software, so I better not complain too loudly.

OK, maybe a little….

I'm glad we have these new ideas coming out. Things like repositories, common user interfaces across like applications, interoperability, UML implementations, and the like. However, after using some of these products, I have one piece of advice that I think every software vendor can use. Let's not reinvent the user interface, OK folks? I know, I know, you all aren't Microsoft, and yes the evil empire kinda sets the standard for GUI interfaces. However, let's put our egos aside for a moment -- forget about our squabbles over who has the most expensive house -- and think about our users. If I can't sit down and use your product without having to forget what I know about using a Windows interface in order to use your software, your product isn't worth much.

Think about this for a minute. You are already forcing me to learn your way of doing things, your vocabulary, and your proprietary database to store the repository. Don't make me re-learn how to use a grid. And don't make me wonder what happens when I double click on a word (Oracle, are you listening? When you double click on a single word, the word, and only the single word, is supposed to be highlighted and selected). Come on folks, let's test these user interfaces out on some professionals for a while. Don't rely on your programmers to decide what's the best way a data modeler should enter the information. And, if you want to totally go against the established user interfaces, then you are just plain arrogant. Kudos go out to companies that have user interface labs at their conferences (or via their beta program).

The only other gripe that I care to share this time is the difference between a de-normalized database and an un-normalized database. In our rush to implement the latest buzzword (data warehouses/data marts) we like to talk about de-normalizing for speed. While that in itself has some inherent problems; let's not forget that to de-normalize implies that it was once normalized. Don't expect to port data from one un-normalized structure to another (while calling that new structure 'de-normalized') and have a good system. Just a little bit of common sense that you seldom see in the rush to make the almighty dollar.

OK, enough about that. Last issue I meant to tell this joke, but I forgot. It's a great little joke that I found in Martin Fowler's UML Distilled (I don't know who really wrote the joke, but that's where I found it).

Q: What's the difference between a methodologist and a terrorist?

A: You can negotiate with a terrorist.

You may want to tell this joke to lighten the mood on your next project after you make fun of the ER folks.

Preface

We've got yet another great issue for you this month. Dr. Terry Halpin continues his series on UML and ORM. Pat Hallock discusses how to keep your integrity while de-normalizing.Dr. John Sharp talks a bit about Natural Language Modeling and requirements,supplies us with another analysis problem, and gives us last issues analysis problem solution. I discuss normalization and why you shouldn't care. And our newest contributor Dennis Corbo discusses implementing a Conceptual Information Processor.

Scot A. Becker is a software consultant and the founder of Orthogonal Software Corporation. He is also a certified ORM consultant and trainer, a certified Visio trainer, and former Editor of the Journal of Conceptual Modeling.

Contact Information:

Scot A. Becker
Orthogonal Software Corporation
scot@orthogonalsoftware.com
www.orthogonalsoftware.com